bcg technology advantage november 2018 · 5% before the start of the transformation. so how,...

bcg technology advantage

• Digital Maturity Is Paying Off

• The $1 Trillion Opportunity in Digital Support Functions

• The Big Leap Toward AI at Scale

• How CEOs Keep Agile Transformations Moving

• Agile Traps

• Renault’s Billion-Euro Digital Transformation

Take Control of Your Digital

Future

NOvEMBER 2018

The Boston Consulting Group (BCG) is a global management consulting firm and the world’s leading advisor on business strategy. We partner with clients from the private, public, and not-for-profit sectors in all regions to identify their highest-value opportunities, address their most critical challenges, and transform their enterprises. Our customized approach combines deep insight into the dynamics of companies and markets with close collaboration at all levels of the client organization. This ensures that our clients achieve sustainable competitive advantage, build more capable organizations, and secure lasting results. Founded in 1963, BCG is a private company with offices in more than 90 cities in 50 countries. For more information, please visit bcg.com.

Preface

FEATURETakeControlofYourDigitalFuture 2

ANALYSISDigitalMaturityIsPayingOff 8

FOCUSThe$1TrillionOpportunityinDigitalSupportFunctions 14

FOCUSTheBigLeapTowardAIatScale 18

VIEWPOINTHowCEOsKeepAgileTransformationsMoving 27

VIEWPOINTAgileTraps 29

Q&ARenault’sBillion-EuroDigitalTransformation:AnInterviewwithCIOFrédéricVincent 31

Preface Contents

What does it means to be a leader in a workplace profoundly impacted by technology?

Though digital technologies have tremendous promise for so many as-pects of life and business, they are leaving enterprises challenged to find employees able to fully embrace these technologies. Moreover, the pace at which these technologies are being introduced can be unsettling for many, including managers. In this environment, it’s more important than ever to build effective teams and a workplace culture that provides the workforce with opportunity and inspiration. It’s up to leaders to figure out how to make the most of digital resources as well as human resourc-es, for the sake of both the business and the workforce.

The feature article in this edition of BCG Technology Advantage, “Take Control of Your Digital Future,” explores the payoff that can come when an enterprise focuses a move to digital on its human resources by recog-nizing the value of internal employees and understanding how people will work with digital. From there, they can develop the skills of existing employees (through coaching and upskilling), selectively hire new talent, and incorporate human-centered design in the transformation initiative.

Other articles in this edition explore the value of digital, including AI, as well as the value of agile ways of working. These articles too consider the challenges that enterprises confront in their transitions to digital, in-cluding the aforementioned “people challenge.” No surprise, the pace to-ward digital varies across industries. “Digital Maturity Is Paying Off” shows which industries lead and which lag. This is a useful lens by which to view the urgency of addressing the people challenge—but given the generally fast pace of digital, companies across industries should be turn-ing their attention to these issues.

Our final article offers a glimpse into a real-life digital transformation. In an interview, Renault’s Frédéric Vincent describes the automaker’s ap-proach to coaching and upskilling its workers: Renault Digital, a subsidi-ary that serves as an incubator to help teams learn new skills and partic-ipate in the digital transformation.

Renault’s solution sounds like a promise of both opportunity and inspi-ration. Just the kind of thing that leaders should be considering for their digitally enabled workforce.

What’s going on at your workplace? Please share your observations and insights with us at [email protected]. We are always look-ing to hear stories and comments from our readers.

The Boston Consulting Group | 1

Companies today understand that digital transformation is no longer an

option. As markets, business models, and customer expectations evolve by the minute, firms face a stark choice: transform or die. But because such transformation can be tremen-dously hard, many businesses are tempted to use the wrong approach—allowing external vendors to drive the effort.

Yes, outsourcing can help you access new ca-pabilities and tools. But a digital transforma-tion isn’t something a company can simply plug in. Firms that look at transformation purely from a technical perspective struggle to achieve the speed, scale, and scope that a full-blown, fully effective digital overhaul re-quires. Too often, initiatives fizzle after the pilot stage. New digital applications aren’t used as expected—or at all. Productivity gains that seemed certain at the outset never materialize.

Successful transformation is as much about people and processes as it is about technolo-gies. That’s why businesses must take a com-prehensive and systemic approach that focus-es on all three dimensions simultaneously. The best way to do this is from the inside. An external vendor may know a technology but won’t likely understand how a business, not to mention its culture, works at the deepest level. Without a rich internal perspective, compa-nies can’t easily achieve the kind of fast, at-

scale transformation that provides a competi-tive advantage. Businesses need applications that are “just right” for users; the ability to ex-tract value from data throughout the organiza-tion; and processes and organizational struc-tures that fit with new technologies—and vice versa.

Instead of outsourcing their digital transfor-mation—and in turn the fate of the busi-ness—companies need to develop the capa-bilities that let them take full control of it. Our work with companies across a range of industries revealed five key things they must master to unlock the full potential of their digital future. It also shed new light on why some businesses stumble at the threshold.

Escaping the Complexity TrapSurprisingly, productivity growth has general-ly not kept pace with the growth in IT spend-ing. This despite the fact that technological progress has been shown as a key driver of productivity growth—a finding that won economist Robert Solow the Nobel Prize in 1987. Yet the computer age has—so far, at least—seemed to contradict Solow’s theory.

Experts continue to debate what causes this “productivity paradox” and whether it is tem-porary or permanent. But one idea is particu-larly compelling: while companies have made massive investments in technology, they have

FEATURE

TakE ConTrol oF Your DigiTal FuTurEby vanessa Lyon and Anne-Françoise Ruaud

2 | BCG Technology Advantage

not focused on true integration—that is, inte-grating tools with the way people actually work. As a result, companies regularly fall into what’s known as the complexity trap. They deploy sophisticated tools and plat-forms that they ultimately don’t use effective-ly or simply avoid in favor of more informal “shadow” processes. This, in turn, leads to overly complicated information systems or to organizations that adapt poorly to the new tools. Either way, expected productivity gains fail to materialize, and in the most severe cas-es, entire systems—and investments—are abandoned.

transformations that emerge from the digital era may spell the end of the complexity trap.

Fortunately, digital technologies and methods are supremely flexible. They enable business-es, end users, and IT departments to work in-teractively to design applications and user journeys that are “just right,” and to adapt processes accordingly. In short, the transfor-mations that emerge from the digital era itself may well spell the end of the complexity trap.

In our work across many industries, we’ve seen companies unlock significant value from digital, with massive impact in terms of mar-gin improvement, revenue growth, and em-ployee engagement. While these businesses are quite dissimilar, they all have taken a ho-listic approach to transformation, tackling the three key dimensions—people, process, and technology—simultaneously. (See the side-bar.) These companies deploy new technolo-gies, user interfaces, and data architectures in a way that reflects how people work. At the same time, they adapt their existing processes and organizational structures to unleash the full potential of new tools and platforms.

Consider one leading industrial conglomerate we worked with whose CEO understood that to stay competitive, industrial firms would need to transform into software and analytics companies. By leveraging data—whether about customers or equipment on the factory

floor—companies could quickly identify and respond to patterns that revealed opportuni-ties and trouble spots. To spur such a change, the CEO transformed his large, established company into one that thought—and worked—like a startup. It adopted an agile culture that stressed experimentation, learn-ing, and iterating. It created several digital “centers” to foster deeper collaboration among software engineers, digital specialists, and industry experts. It brought all of its digi-tal capabilities, tools, and data into a single organization.

Significantly, the company did not simply de-ploy new technology. It factored in how people worked—and identified changes in processes and organizational structure that could help them work better, be more productive, and, ultimately, create more value in a digital world. Together these moves helped the busi-ness develop new, successful digital offerings. Thanks to this approach, within five years the company’s digital offerings accounted for 15% of the company’s revenue—up from less than 5% before the start of the transformation.

So how, exactly, can your company ensure that kind of success?

Five Things Companies need to Do WellBy developing five capabilities internally, your business can achieve a true digital transformation—one that occurs at a speed, scale, and scope that bring real competitive advantage. Granted, our list might feel intimidating to even the strongest manage-ment team. In fact, COOs we’ve met question how feasible such changes actually are. And they’re not the only ones with doubts. CIOs might be uncertain about their ability to transform a landscape of older, not easily adaptable legacy systems, especially if the required skills and knowledge are in short supply. And HR heads—who are already fighting tough battles to attract and retain digital talent—may question the efficacy of any “solution” to the recruitment challenge. The good news is that there are ways to help develop each of the five capabilities internally—and unlock the full potential of digital.


4 | BCG Technology Advantage—Feature

Mostcompaniestodayviewdigitalasapathwaytogrowth.Butspottingtheroadisonething;propellingalongitisquiteanother.Manybusinessesstruggletounleashthevalueofdigital,yetothershavemet—andevenexceeded—theirexpecta-tions.Howdidtheydoit?Herearesomeinsightsfromthreedifferentsuccessstories.

The Consumer Goods Company.Initsmassivetroveofcustomerandbusinessdata,aconsumergoodscompanysawpotentialvalue—andplentyofit.Butunlockingthatvaluewouldrequiresignifi-cantdigitalcapabilities.Notonlydidthecompanyrequireanewdigitalarchitecturetomoreefficientlyandeffectivelyusedata,italsoneededacarefullythought-outplanforintegratingitspeopleandprocesses.Thecompanylookedintotraininganddataaccessrightsaswellasthekindofimpactthatdatacould—andshould—haveondecisionmaking.Intheend,thecompany’ssystemicapproachenabledmorethan50,000employeestoaccessthecompany’sdata,increasingusageandspurringnewapplicationsforthatinformation.Thecompanyalsodeployednewdigitaltoolsandmethodologies,suchasdigitalmodel-ingandsimulation,anddata-drivendecisionmakingbecamethenorm.Theupshot?Decisionsineveryaspectofthebusiness—fromproductdevelopmenttomarketing—arenowbackedupbydataanalysis.

The Card Payment Company.Akeyplayerinitsfield,thiscardpaymentcompanysawcompetitors—traditionalrivalsaswellasinternet-basedupstarts—utilizedigitaltechnologiestolaunchsuccessfulnewproducts.Determinedtoretainitsfront-runnerstatus,thecompanytransformeditselfbydeployinginnovativecustomer-centricdigitalcapabilities.Forexample,itemployedpersona-basedusecasesandbigdataanalyticstodevelopoffersmorecloselytailoredfor—andmore

relevantto—specifictypesofcustomers.Italsodeployedamoremodern,flexibleITarchitecturetohelplaunchnewbusinessapplications.Andbyembracingnewmarketingmodels,aswellaspartnershipsinareassuchassocialmedia,thecompanylearnedtointeractwiththeoutsideworldinwaystheworldnowpreferred—whilegaininginsightsthatcouldprovevaluablewhendevelopingnewproducts.Sincelaunchingitsdigitaleffort,thecompanyhasseenasignificantandrecurrenttop-lineimpactofhundredsofmillionsofdollarsperyear.

The Fast-Moving Consumer Goods Company.Aseriesofpilotprogramsconvincedtheleadersofthisbusinessthatdigitaltechnologiescouldtransformitsmarketingefforts.Butthecompanydidn’tsimplydeploynewtoolsandmethods.Italsoreimagineditsoperatingmodelandthewayitsmarketingprofessionalsworkedandwereorganized.Replacingacurricu-lumthathadchangedlittlein40years,thecompanyofferedmorethan3,000ofitsmarketingstaffanewcoachingandupskillingprogram.Italsocreatedorgani-zationalstructurestofosterandincreasecollaborationbetweenbusinessfunctionsandwithdigitalagencies.Sincelaunchingitsdigitalinitiative,thecompanyhasgainedbettercontroloveritsmediastrategy,butithasalsoseenmorequantifi-ableresults:itdoubleditsreturnoninvestmentofdigitalmarketingspending;reducedcostsby30%;increaseditsaudiencereachbyafactoroffour;andachieveda50%higherengagementlevel.

ASYSTEMICAPPROACHTOUNLOCKINgDIgITAL’SPOTENTIAL

Stick to Clear Business Objectives, Even as the Digital Environment EvolvesFor all the benefits the digital era brings, it also carries uncertainty. Along with continual-ly evolving technologies—through advances in performance and miniaturization—come rapid changes in customer expectations, mar-ket standards, and the competitive landscape. Your company might feel challenged, there-fore, especially if it’s used to developing and launching new products or services in a tradi-tional three- to five-year cycle. You’ll still need to meet midterm business objectives, yet an evolving, uncertain environment means you must also maintain tactical flexibility.

The key to acquiring such flexibility is a test-and-refine approach. (See “The Digital Imperative,” BCG article, March 2015.) Product designs—and, ultimately, proto-types—should go through frequent cycles of testing and then incorporating feedback. And the process shouldn’t end with a product’s release. Instead, continually seek customer reactions to identify potential improvements. In effect, the details—unknown at the out-set—are revealed and integrated as you go along.

Companies that do a good job working to-ward a future that is hard to see share certain traits. They maintain a portfolio of opportuni-ties. They work in an open environment and within an ecosystem of partners—the better to fuel their innovation engine. They look ahead when making choices. They communi-cate to employees and partners a clear over-all purpose and a framework of priorities and guiding principles. Such communication keeps everyone aligned across the company while enabling flexibility in daily operations. And without fail, they define the criteria for success—and evaluate all business decisions against them.

Acquire Digital Skills Despite a Talent ShortageNot surprisingly, the demand for digital prod-ucts has created a shortfall of talent—and not only on the technical side, where many new jobs are needed, such as data scientists and developers. Increasingly, business-side roles also are acquiring a digital flavor. Sales

representatives, for example, need to learn how to use predictive tools and data analytics to better target customers.

Your company can pull different levers to help bridge the skills gap. It can, for instance, build up its talent base through acquisitions of other businesses, particularly startups. But developing the skills of existing employees will be critical. These are your key assets, the people who know the business best. What’s more, businesses that make skills develop-ment a priority can reignite the “technical pride” of IT and engineering teams—which smooth the way to digital transformation.

developing the skills of existing employees will be critical.

All of this points to the need for companies to renew their social contract with employees, the implicit deal that defines the duties and obligations of both employer and employee. More than previous generations, today’s workers seek purpose and autonomy in their jobs. If you can empower and inspire your people, you will bring out their best while reducing attrition.

In a digital world, the role of your managers needs to change too, as digital tools offer teams more autonomy. By acting more like coaches than controllers, managers will help teams adopt new tools and ways of working, while drawing on the wealth of information that is accessible. Moreover, such managers will need to provide employees with regular feedback and promote a culture of experimentation.

Simplify Decision MakingMost large companies today function, at best, with a one-month decision clock. Committees meet every four weeks or so to make vital de-cisions about the business. But today’s digital- native companies simply make decisions at the moment they need to make them. In such agile environments, teams take ownership of their projects and are empowered to make de-


cisions, which in turn enable companies to move much more quickly, often rolling out solutions within weeks rather than months.

to shift into digital speed, look at decision-making processes and bottlenecks.

To shift into digital speed, then, look closely at your company’s decision-making processes and tackle the bottlenecks. For instance, a global automotive player with whom we worked discovered that monthly meetings and multilayered processes hampered the company’s decision-making agility. So the company streamlined its processes for digital, notably by implementing weekly business-decision committees for every division. It tasked each committee, comprising a senior business lead for digital and his or her IT counterpart, with building a long-term digital portfolio, prioritizing initiatives, keeping project delivery on track, and maximizing end users’ adoption. To boost the pace of initiatives, the pair owns the digital budget for the business unit and is empowered to make investment decisions. The result of this new model? A double-digit margin increase for the company.

Accelerate the Delivery of Customer-Focused ApplicationsTwo powerful approaches that your company can use both internally and externally—agile methodologies and human-centered design—align nicely with user needs and preferences. Agile methodologies emphasize speed and flexibility. They include frequent iterations, short timelines, and minimum viable prod-ucts. They stress continuous testing and feed-back cycles, which give developers the ability to implement changes as needed. So if user requirements evolve during development—no rare event in the digital era—products can evolve with them.

Human-centered design, meanwhile, puts the user front and center when companies are creating new products, platforms, and interfaces. Like agile methodologies, such

design involves frequent testing and feedback, with designers typically creating numerous prototypes, watching how testers respond, and making tweaks to better align with user behaviors and needs. This approach proves especially useful for companies creating internal digital solutions that employees actually use—and use effectively. Companies that treat employee-users like customers, seeking their feedback on prototypes and pilots and incorporating it into new iterations, wind up with digital solutions that not only work, but work for employees.

Consider the experience of an industrial com-pany that sought digital technologies to re-duce the amount of time plant supervisors spent walking between their desks and the workshops. Using agile methodologies and human-centered design, developers created an intuitive tablet-based app that provided, in real time, the specific KPIs that supervisors used to manage their teams. With this any-where, anytime dashboard of metrics, super-visors now spend less time “commuting” and more time on value-generating tasks. Within six months of its initial launch, the company rolled out the app to several hundred users at a number of sites; within a year and a half of the launch, the app will be fully deployed to 3,000 people. Significantly, the company has continually refined the app in response to user feedback.

Build on Existing IT Systems—Even the Most Rigid OnesTraditional companies often have to find ways to connect new digital applications to legacy systems that, while crucial for running the business, can be old and complex. Inte-grating them with new platforms, services, front ends, or sources of data is rarely seam-less. Compounding the challenge, different business units and departments usually have their own systems and associated data. So deploying a slick new application—or lever-aging data stored across the company—can become a grueling task.

If your company is pursuing its digital ambi-tions in a legacy environment, think about adopting a flexible, staged approach. (See “Generating Value While Transforming Insur-

6 | BCG Technology Advantage—Feature

ers’ Legacy Technology,” BCG article, Decem-ber 2016.) That way, you can renovate your company’s legacy IT while boosting its digital quotient for immediate value creation. In a staged approach, you first build the data foundations for a digital transformation; in effect, decoupling data from specific systems and sources and creating a single data layer. Using this data layer, you can then reimagine and automate many processes, improving the customer journey—and experience—while boosting efficiency. Savvy companies will also make use of APIs (application programming interfaces) and modular architecture, which enable you to integrate new front ends and applications more easily.

It might take you several years, but in the end this process of incrementally renovating your IT landscape can result in a digital-ready en-vironment for your company—with minimal disruption to the business.

Today’s demand for digital talent and skills presents businesses with a unique

opportunity to mobilize and repurpose work-forces to catch the digital wave. Although transformation is never simple, many compa-nies have accumulated vast—and valuable—experience. They know the business and its processes and goals inside out, and this knowledge will serve them well as they im-plement change.

Vanessa Lyon is a partner and managing direc-tor in the Paris office of The Boston Consulting Group. She is a member of the leadership team of the Technology Advantage practice and a core member of the Industrial Goods practice. You may contact her by email at lyon.vanessa @bcg.com.

Anne-Françoise Ruaud is an associate director in the firm’s Paris office. You may contact her by email at [email protected].


The performance gap be-tween digital champions and

laggards is widening, and digital’s continued contribution to company performance means that gap will likely grow. But any company can improve. By taking steps to become more digitally mature, a company can boost cost efficiency, time to market, competitive advantage, and market share.

This holds true across a wide range of industries, according to a second annual study conducted by The Boston Consulting Group. The study, involving more than 1,900 compa-nies in Europe and the US, relied on BCG’s Digital Acceleration In-dex (DAI) to derive comparisons. (See the sidebar.) Specifically, the study found that while 25% of the companies surveyed qualify as champions, almost one-third ap-pear significantly behind in their digital capabilities. Among indus-tries, the biggest share of digital champions comprises telco, tech-nology, banking, and automotive.

Meanwhile, industries with a higher share of digital laggards—such as health care, energy, and manufac-turing—are looking for ways to

improve their standing. Our study correlates specific digital initiatives with specific positive outcomes—such as boosting market share—and shows how digital champions are starting to see a return on their

digital investments. These findings could provide valuable guidance for companies that have fallen behind, suggesting how they might up their digital game in the most efficient way possible.

ANALYSIS

DigiTal MaTuriTY is PaYing oFFby Michael Grebe, Michael Rüßmann, Michael Leyh, and Marc Roman Franke


Forour2018study,weasked1,900companiesinEuropeandtheUStoestimatetheirdigitalmaturityonascaleof1to4in37categories.Wethenaggregat-edthoserawscoresandcalcu-latedresultingvaluestotheirresponsesonascalefrom0to100.Weweightedthemtodetermineeachcompany’soverallperformanceonourDigitalAccelerationIndex(DAI).

CompanieswithaDAIof67to100qualifyaschampions,whilethosewithaDAIof43orlessarecategorizedaslaggards.Championshadanaveragematuritylevelofatleast3outof4onallmaturitydimensions,whereaslaggardsreportedaveragematuritylevelsoflessthan2in66%oftheassessed

dimensionsandanaveragematurityof3in33%oftheassesseddimensions.Champi-onsinthisstudyhadanaverageDAIof80,whilelaggardsscored30,onaverage.

Wecollecteddatafromcompa-nieswithatleast2,500employ-eesintheUS,germany,theUK,andFranceacrosstenindus-tries:manufacturing,chemicals,technology,banking,telecom-munications,consumergoodsandretail,automotive,energy,healthcare,andthepublicsector.Mostlyseniorleadersparticipated:30%wereC-level,34%weredivisionleaders,and26%weregeneralmanagers(10%didnotreporttheirrole).

OURMETHODOLOgY

Digital Drives Performance We found that increased levels of digital maturity significantly im-proved competitive advantage along multiple performance indi-cators, such as time to market, cost efficiency, product quality, and customer satisfaction. In fact, an in-crease in digital maturity of 25 DAI points improves the likelihood of reaching a superior performance for time to market, and cost effi-ciency doubles. This connection holds for an increase of 50 and 75 points, respectively, with the likelihood tripling and quadru-pling. We observed less strong, yet significant, effects for product quality and customer satisfaction.

For gaining competitive advantage overall, we found that new digital growth is the main driver. Champi-ons that offer new digital services and products and incubate start-ups seem to have stronger compet-itiveness than laggards in those di-mensions. (See Exhibit 1.) As for individual performance indicators, we identified the most relevant digital initiatives as follows:

• To improve time to market, building go-to-market capabili-ties such as digital marketing or

personalization is essential. Startup incubation also has a major effect in addition to contributing to competitive advantage.

• To enhance cost efficiency, digitizing the technology enablers is imperative—for example, making the tech function ready for the future, building digital platforms, and setting up DevOps for digital delivery.

• To upgrade product quality, people and organizational dimensions such as digital leadership are important, as well as a strong digital culture and governance to foster digital initiatives and digital talent. In other words, investing in digital talent naturally yields higher-quality digital products.

• To perform better in customer satisfaction, it’s critical to focus on digital customer jour-neys and personalization of offerings as well as prototyping digital products and supporting lighthouse projects.

All of these effects on competitive advantage manifest over time. In a

DAI deep dive study, we assessed 81 leading telcos from more than 40 countries to compare the mar-ket share evolution of champions and laggards. From 2012 to 2017, champions increased market share by 7% and laggards saw their mar-ket share drop 11%.

applying Digital BoostersGiven that digital pays off, the question becomes how to digitize fast. In last year’s study, we iden-tified three boosters that cham-pions have adopted: investment (more than 5% of operational expenditures [OPEX] on digital), recruitment of digital experts (more than 10% of full-time equivalent [FTE] employees hav-ing digital roles), and embedding digital in their organization (in a hybrid or built-in model). In our previous study, champions that applied all three boosters had a DAI score that was 16 points high-er, on average, than the laggards’. This gap has actually widened to 21 points in the recent study. Even applying just one booster improves DAI scores: 8 (recruiting digital experts), 9 (spending more than 5% of OPEX on digital), or 14 (em-bedding digital in the organi-zation) points on average.


CHAMPIONS OUTPERFORM PEERS1 KEY DRIVERS BEHIND PERFORMANCE

60% 25%

78% 29%

24%62%

Cost efficiency

Time to market

Competitiveadvantage

Digital champions Digital laggards

Go-to-market

Technology

New digitalgrowth

… such as future-ready tech functions, digital platforms, and DevOps

… such as new digital services andproducts and startup incubation

… such as digital marketing, sales, and personalization

Source: BCG analysis.1Percentage of digital champions and laggards that said they outperformed peers over the past three years.

Exhibit1|ActionsThatDriveDigitalMaturity

In this year’s study, we found the effects of these boosters varied by industry. For example, a high share of digital FTEs has a strong impact on manufacturing (12 DAI points) and health care (15 DAI points). Given that these industries severe-ly lack digital skills, it’s logical that hiring any new digital talent accel-erates the companies’ digital journeys. Meanwhile, spending more than 5% of OPEX on digital significantly improves DAI scores in banking, tech, and health care. Embedding digital in the organ-ization and empowering business units to drive digital initiatives is the strongest booster across all industries.

industries struggle with Digital Maturity In line with last year’s results, the US outperformed Europe, with more champions (25% versus 22%) and fewer laggards (31% versus 33%). (See Exhibit 2.) Much of this edge in US performance is due to stronger US technology and telco sectors, which support other

industries’ digitization efforts through partnerships and close collaboration. Among US telcos, 46% score as digital champions, compared with only 27% of European telcos. The difference, based on self-assessments, is less dramatic in the tech industry, where 33% of US and 31% of EU companies report a high digital maturity; that’s surprising, given the large number of leading tech companies based in the US.

It’s logical that tech, telco, and banking companies lead in digital since their main products and ser-vices are already mostly digitized. For all other industries, the digiti-zation journey is more challenging, given their nondigital product portfolio. An exception is automo-tive, an industry traditionally fo-cused more on products than on service, with a strong share of champions in both the US and Eu-rope (27% in both regions).

Compared with last year’s report, we expanded the scope of our study, taking a closer look at addi-

tional industries, including health care and energy, where we found a comparably high share of laggards. These industries face significant challenges to bringing their digital strategies to life, particularly when it comes to driving innovative tech-nologies into core processes or pur-suing new business opportunities. (See Exhibit 3.)

Energy Has the Most laggardsAs a whole, the energy sector re-ported the highest share of lag-gards among all surveyed indus-tries: 61%. The picture by subindustry provides more differ-entiation: 71% of oil and gas (O&G) companies assessed themselves as laggards, while the share in utili-ties (57%) was much lower. It’s somewhat surprising that O&G re-ported such a high percentage of laggards, given that companies have invested significantly in mak-ing the exploration process more efficient. One explanation could be the high degree of autonomy in operating units. Enterprisewide,

10 | BCG Technology Advantage—Analysis

UNITED STATES EUROPE1

075100 2550

31%25%

100 75 75 050

22% 33%

DAI

DAI

46%33%

27%23%

19%

18%15%

22%24%

21%27%

33%

55%

40%52%

2080100 60 40 0

Telecommunications

Public sector 15%Manufacturing

EnergyHealth careChemicals

27%Automotive

Banking 30%Technology

46%19%

Consumer & Retail

10%

6%

39%65%

80 2060 40100 0

16% 35%

25% 30%

27%

29%

17%31% 31%

27%

39%60%

29%16%

19%8%

Percentage of digital champions Percentage of digital laggards

Source: BCG analysis.Note: DAI = Digital Acceleration Index.1Europe includes Germany, France, the United Kingdom, and Austria.

Exhibit2|TheUSHasMoreDigitalChampionsThantheEUinMostIndustries

end-to-end digitization is challeng-ing to achieve because each oper-ating asset has its own require-ments and initiatives.

Interestingly, O&G companies are now reviving digital initiatives that had been on hold since the oil price crash, when the industry turned its attention to consolida-tion and cost reduction. For O&G players, the biggest challenge is tapping into their large pools of data. Both the data platforms and infrastructure and the data and analytics governance dimensions in our survey show very low matu-rity (both with average DAI scores of 20). On the other hand, O&G players reported the greatest aspi-ration for these dimensions by 2020 (DAI scores of 35 and 39, respectively).

As noted, there are fewer laggards among utilities, which are closer to end customers and have stronger go-to-market capabilities driven by their stronger B2C footprint. But like O&G, utilities tend to think more on a project-by-project basis (for example, building a new pow-er plant), which can make end-to-

end digitization initiatives seem less urgent or necessary.

That said, market forces such as the disruptive power of renew-ables, the decentralization of ener-gy creation (the smart grid), tech competition (such as Google Nest), and the increase in demand from electric cars are pushing utilities toward digitization.

This has led to important initia-tives, such as recruiting experts from other industries and building innovation labs to explore new op-portunities (such as electric car batteries and peer-to-peer market-places for power). Furthermore, some utilities started leveraging their vast customer data to devel-op new products and services, and many are advanced in analytic modeling (for example, weather information for predicting the util- ization of windmills). Some have even started to pilot artificial in-telligence and blockchain technol-ogy in the trading space—reflect-ed in the high score for the artificial intelligence dimension (DAI 39). By comparison, O&G companies are only beginning to

use AI and other emerging tech-nologies and so scored much low-er (DAI 13).

All of these technologies require a digital and IT function that will act as a partner with the business to identify the right use cases for these technologies and enable the digital future.

Health Care’s Wide-ranging MaturityThe health care sector has a big share of laggards (43%), but like energy, the picture is more nu-anced when looking into the vari-ous subsectors: medical technology (medtech), providers, and biophar-ma/pharmaceuticals (bio/pharma). All three subsectors score highest in having a business strategy driven by digital (medtech DAI 64, provid-ers 52, bio/pharma 46). Even though all have a solid digital strat-egy, when it comes to digitizing the core, we found that medtech (DAI 61) is far ahead of providers (DAI 40) and even more so bio/pharma (DAI 25). Success in this area is what seems to truly differentiate laggards and champions.


Energy

Automotive

Health care

SELECT OPPORTUNITIESMAJOR CHALLENGES

• Disruption from renewables, smart grids, e-cars, and tech competitors

• Industry under consolidation and pressure to significantly reduce costs

• Low degree of process digitization, especially in the supply chain

• No systematic approach to leverage partner ecosystem and wealth of data

• Asset-heavy business with long cycle times for development and production

• New car companies like Tesla or Uber, which compete in asset-light business models

• Use of data for predictive analytics toforecast energy consumption

• Application of AI1 and blockchain technology in energy trading

• Offering digital customer interaction like patient portals and remote monitoring

• Using ecosystem data, for example, for clinical decision support and preventive care

• Establishing digital platforms and ecosystems around mobility

• Digitization and automation of core processes with analytics and AI1

Source: BCG analysis.1Artificial intelligence.

Exhibit3|AllIndustriesFaceUniqueChallengesandOpportunities

Medtech is the most advanced sub-sector, with only 10% laggards, which makes sense since technolo-gy and software for diagnostic pur-poses are at the core of its busi-ness. Being close to the customer and regularly receiving feedback on product innovation helped this subsector achieve a respectable DAI score of 63 in reinventing cus-tomer journeys. Meanwhile, it is also experimenting with connected devices (for example, for remote monitoring) for new growth, which is reflected in the high score for new digital services/products (DAI 50), compared with a score of 29 for bio/pharma.

Medical providers (such as hospi-tals and medical centers) are much further behind on the digital jour-ney, with 54% categorized as lag-gards. But the subsector with the most laggards was bio/pharma (70%). This is due to a very compli-

cated supply chain that some com-panies still manually manage on Excel spreadsheets. Bio/pharma’s customer journey score (DAI 28) shows room for improvement, es-pecially compared with that of providers (DAI 44).

But the outlook for health care is positive, especially if it can better utilize two inherent advantages: a broad ecosystem and tons of data. Today, many health care compa-nies have myriad partnerships yet no strategic approach to leverage these networks for new growth. To turn this potential into a reality, they will need digital capabilities and real-time data access, which is currently still an area with lower digital maturity.

For example, bio/pharma scored DAI 13 on data and analytics gover-nance and DAI 7 on digital and arti-ficial intelligence. They will also need to ramp up data analytics ca-pabilities to apply AI and other emerging technologies like block-chain. (See “A Prescription for Blockchain in Health Care,” BCG article, April 2018.)

To provide one example, a multi-national biopharmaceutical com-pany launched a digital initiative to integrate data sources, gain greater visibility into the supply chain, and generate insights be-yond what is possible from tradi-tional enterprise resource planning (ERP) data. This project allowed the company to map the supply chain end-to-end, expose sources of volatility, and get real-time up-dates on material flow and disrup-tions. By applying machine learn-ing and smart analytics to this

deeper well of data, and leveraging those to guide their process optimi-zation and operational excellence initiatives, the company lowered costs by 3% to 5% and became more efficient, reducing working capital by 5% to 7%.

automotive’s Digital know-HowThe automotive industry has the highest share of champions, 27%, after telco, tech, and banking. Digi-tal champions scored particularly high in digital roadmap (DAI 87) and manufacturing/industry 4.0 (DAI 85). Automotive is very experienced—and way ahead of other industries—when it comes to digitizing R&D and production.

With the emergence of connected and autonomous cars, digital has become essential to the product roadmap.

Today, fast-moving newcomers pose a threat to the industry. These new entrants have very fast internal cycle times, or, like Uber, they compete using asset-light digital business models.

Some of the champions in automotive started responding early to these threats. This is reflected in their high scores for developing digital capabilities (DAI 75), digitizing their core processes (DAI 77), and launching digital businesses (DAI 75). Car lenders, dealers, and OEMs should move quickly and partner with startups or build their own digital plat-forms or ecosystems to create revenues from adjacent or new businesses. For instance, Daimler and Volkswagen funded Auto-Gravity, a site where potential car buyers apply for loans and compare rates directly on their smartphones.

The 25% of laggards in automotive face challenges foremost in process digitization and robotics, next- generation sales, and DevOps (each dimension with average DAI score of 25). These laggards risk losing long-term competitiveness with their peers, as well as digitally sav-vy new entrants.

steps Moving Forward Every company has its own starting point. While we found that most companies are incor-porating digital in their strategic roadmaps, what truly sets apart the champions is how they bring that strategy to life on an oper-ational level. Here are some findings that hold true for all companies.

automotive is very experienced when it comes to digitizing R&d and production.

12 | BCG Technology Advantage—Analysis

Create the next-generation tech function. Companies should pursue digital maturity to create a technology function of the future. Besides improving efficiencies and costs, a future-ready IT and tech function is critical to enabling emerging technologies such as AI and blockchain.

Drive customer-centric digitization of the core. Companies need to create value fast with digital initiatives that always keep the customer top of mind—for exam-ple, by designing digital customer journeys. This leads to customer satisfaction, of course, but it also enhances product quality. In addition, through personalization and digital marketing, companies can accelerate their time to market.

Focus on new digital business opportunities. Companies that digitize the core while simultane-ously identifying new growth areas gain competitive advantage. Opportunities are manifold and

may involve startup collaboration, ecosystem expansion, and leverag-ing assets such as customer data.

Becoming digital pays off in su-perior performance. Across all

industries, we see great opportuni-ties to improve financial perfor-mance and gain competitiveness from digital investments. Compa-nies must move away from seeing IT as a cost center but rather as a source of economic and competi-tive advantage. Only then will they have the confidence and commit-ment to move their digital trans-formations forward.

Michael Grebe is a senior partner and managing director in the Munich office of The Boston Consulting Group. He is the leader of the Technology Advantage practice in Germany and Austria and a member of its global leadership team. He is also the global leader of the Next Generation Tech Function product. You may contact him by email at [email protected].

Michael Rüßmann is a senior partner and managing director in the firm’s Munich office. He leads DigitalBCG in Central and Eastern Europe and in the Middle East and Africa. You may contact him by email at [email protected].

Michael Leyh is a lead knowledge analyst in BCG’s Düsseldorf office. He is the operational leader for the Digital Acceleration Index (DAI) and the IT benchmarking in banking topic. You may contact him by email at [email protected].

Marc Roman Franke is a project leader in the firm’s Berlin office. He is the operational leader for the Digital Acceleration Index (DAI) and a lead member of the Technology Advantage practice. You may contact him by email at franke.marcroman @bcg.com.


Most executives understand the power of digital technology to improve

performance in support functions, and many companies have begun the process of imple-menting it. However, they are at widely different points in the journey, and most have far to go. A recent BCG study found that only one in five companies, across industries, is actively pursuing a digital agenda and putting the right foundation in place. (See “What to Do When Support Functions Aren’t Ready for Digital,” BCG article, July 2018.) Unfortunate-ly, organizations that wait will have a hard time catching up.

Digital technology provides a number of benefits. It reduces costs and creates transpar-ency, empowering leaders to make smarter decisions more quickly. It also boosts service levels, producing a better experience for internal customers and resulting in fewer errors. (This is particularly true as the digital experience continues to improve for consum-ers in their personal lives; they have equally high expectations of digital experiences from their employers.) Perhaps most important, leveraging the power of digital allows support functions to help business units generate new revenue and improve access to working capital.

We believe that, overall, digital represents a $1 trillion opportunity for companies in the US and European Union alone. (See Exhibit

1.) To seize this opportunity, support function leaders need to develop a bold vision and pursue it relentlessly through a three-part approach:

• Tailor the operating model and processes to harness the power of digital.

• Apply intelligent automation tools, such as computer vision, robotic process automation (RPA), and artificial intelligence (AI).

• Focus on creating business value beyond the recent push for cost reductions.

stopping the squeezeIncreased competition has pressured compa-nies to hit their financial targets, and support functions have often been squeezed as a re-sult. Managers are asked—or ordered—to do more with less. In response, many support functions have adopted an overriding goal of efficiency and cost reductions. They have fo-cused primarily on decreasing costs through brute-force standardization measures, putting in place self-service tools to push work back onto business units, reducing the catalog of services, bundling services, and outsourcing or offshoring activities.

Some executives have been tempted to ap-proach digital with this mindset as well, view-

FOCUS

THE $1 Trillion oPPorTuniTY in DigiTal suPPorT FunCTionsby Fabrice Roghé, Sukand Ramachandran, Yann Letourneux, and Nicholas Clark


ing it as simply a means to further reduce costs. To be clear, digital can dramatically in-crease efficiencies that have significant finan-cial impact. Depending on where a company starts, implementing the technology can lead to reductions in headcount of 20% to 30% and cut costs by up to 40%. If one considers that companies in the US and EU spend some $2.5 trillion on administrative expenses, a re-duction of 40% represents a potential savings of $1 trillion.

In addition, digital can lead to dramatically improved service quality. Digital allows customers to have 24-7 access, and compa-nies can “mass customize” processes, rather than just standardize them, through smart technology. Data is more accurate and error rates drop significantly. Our experience shows that improved process design can free up three days per month for managers in both support functions and business units, allowing them to focus on activities that add more value.

Through such shifts in resources, digital can unlock real value. More transparent data leads to better and faster decision making. Leaders in support functions and business units have better insights, with a clear view of how their decisions will translate to im-provements in such critical business metrics as days sales outstanding, price realization, and procurement terms.

Common PitfallsSo, why aren’t companies pursuing digital more actively? In fact, many are trying, but they often struggle to realize value because, unfortunately, there are a lot of ways to get digital wrong.

For example, some organizations try to take small, measured steps. They focus on incre-mental improvements, rather than on achiev-ing a bold transformational vision. Or they wall off one part of the business and try to transform just that function rather than the entire organization. This is a bit like trying to turn a restaurant into a three-star operation solely by replacing the kitchen equipment. The skill of the staff, the quality of the food, and other factors are all important in improv-ing the customer experience. The large poten-tial gain from digital requires a correspond-ingly orchestrated change effort, and small pilot tests won’t be enough.

Other organizations try to overlay digital on top of existing processes. Given that most support function processes were designed for an analog, manual, standardized mode of working—and are driven by the need for scale—digital won’t truly transform them; it will simply accelerate some segments while leaving much of the complexity and legacy issues in place. For example, adding RPA to a flawed process with an error rate of 40% won’t fundamentally fix the problem; it will


US

10 million full-time employeesworking in support functions

$1.2 trillionin associated costs

EU

15 million full-time employeesworking in support functions

$1.3 trillionassociated costs

30% to40%

cost savings throughautomation

$1trillion

opportunity

2–4fewer days sales outstanding

5%–20%lower supplier costs

50%–75%faster reporting cycles

30%–40%faster recruiting of new employees

10%–20%improvement in customer satisfaction

Potential business benefits

Source: BCG analysis.

Exhibit1|DigitalSupportFunctionsArea$1TrillionOpportunityintheUSandEUAlone

just generate errors faster. Instead, companies need to use the digitization process as an op-portunity to redesign processes from the ground up and from end to end, making sure to prioritize business objectives and the user experience.

Still other organizations get digitization wrong by focusing on customer-facing units and ignoring support functions. Support func-tions often perform tasks that are core to the business (such as collecting cash), and they should be recognized as the digital transfor-mation engine that powers the whole organi-zation. By improving internal processes—such as unlocking cash through more efficient billing processes or resolving internal queries faster—digital support functions can show what’s possible and serve as a model for customer-facing units.

a Three-Part approachTo overcome these pitfalls, companies should follow an integrated, three-part approach. (See Exhibit 2.)

Tailor the operating model and processes for digital. Support functions need to put the right foundations in place by rethinking their operating model and processes. Traditionally, support functions have been organized as a set of isolated departments with concentrat-ed, internal expertise. What organizations need instead is an operating model built around customer journeys. Companies should identify the critical support functions that internal customers need and then determine how departments can best provide them.

This often entails inverting the traditional approach to process design by putting the customer first, streamlining bureaucracy, and

fostering collaboration throughout the department. Critically, the new processes should capitalize on the advantages of digital from end to end, to improve data accuracy and enhance productivity. And they should be customizable wherever possible to accom-modate market-specific factors, such as local regulations.

Apply intelligent automation tools, such as computer vision, RPA, and AI. Rather than an all-of-the-above approach to digital, support function leaders should understand the universe of digital solutions and how they can best be applied to specific objectives. For example, RPA programs use rule-based algorithms to automate and streamline straightforward tasks, such as pulling data from a form or populating cells in a spread-sheet. Similarly, computer-vision algorithms can extract data from nontext sources, such as images and videos. AI and machine learning, by contrast, are smart enough to make decisions independently, and because they use experience to improve their perfor-mance, they get better over time. These tools—which can either replace employees or complement them—are enabled by cloud- storage solutions and big-data tools that increase the availability and accessibility of data, leading to lower costs and higher satisfaction levels.

Create business value beyond cost reduc-tions. Support functions should think of digital as an opportunity to move beyond cost cutting toward a mindset of creating value. This shift typically follows an evolution from standardized and centralized transactions (where many companies are today) to processes that use intelligent automation (further reducing costs) to completely rede-signed processes that increase service levels

16 | BCG Technology Advantage—Focus

Design a new operating model and revamp processes end to endEstablish customer journeys, single decision makers, and data consistency for improvements in quality and productivity

Apply intelligent automation toolsReplace employees, or assist them, to reduce costs and increase service satisfaction

Focus on creating business value, not just cutting costs“Mass-customize” services to increase flexibility and adapt to the needs of the business


Exhibit2|AThree-PartApproachtoSupportDigitalinSupportFunctions

and free up employees and managers for more value-added tasks (the ultimate goal).

getting startedIn the past, support function expertise was primarily focused in technical areas, such as HR and finance. Today, those skills are still required but are concentrated among a small-er number of employees, and companies have a growing need for digital talent that can apply the principles of data-driven solu-tions across various support functions. Most support function organizations, therefore, will need to begin their digital journey by estab-lishing skills in other areas, such as data sci-ence and data engineering. They can do so by either developing those skills in current em-ployees or hiring new talent. This is a marked shift from the traditional paradigm. Some or-ganizations have even created a focused cen-ter of excellence for data science and engi-neering, with the specific mandate of empowering staff to create value in support functions.

What’s more, rather than plotting a tradition-al, sequential, “waterfall” transformation (with a long period of strategizing and imple-mentation), many companies benefit from implementing an agile approach. They choose a few lighthouse projects that can deliver val-ue quickly and prove the concept, then begin to build capabilities and generate momentum for more ambitious measures. Because digital technology is advancing so quickly, choosing to wait is a bad idea. Companies need to jump in, learn through experience, and cor-rect their course when appropriate.

One company, for example, was struggling with outmoded manual steps and high vol-umes—compounded by a complex pricing mechanism and little transparency—in its order-to-cash process. Each year, as the com-pany processed about 200,000 deductions (for such things as large-order discounts), some were mistakenly accepted, others were mis-takenly rejected, and customers were often frustrated by the discrepancy in their bills. The company began digitizing the process by looking at it from the customer’s perspective, identifying the ideal experience, and then working backward to create it. As a result, the

company was able to redesign the process us-ing tools that could anticipate deductions for standard situations, dramatically reducing the error rate and improving the customer expe-rience. The entire effort, from starting point to a working prototype, took just 12 weeks.

In another example, a leading global bank launched a broader transformation with cor-respondingly larger benefits. By digitizing its entire shared-services center, the company was able to reduce the number of full-time employees from 40,000 to 25,000 and—more impressive—to create value of $600 million to $900 million.

In sum, digitizing support functions repre-sents a tremendous opportunity to not only

reduce costs and increase efficiency but also unlock new value. Most organizations are closer to the beginning of this journey than the end. By following the three-part approach discussed here—designing a new operating model and processes, putting the right tech-nologies in place, and relentlessly focusing on value—companies can give themselves a boost and a clear edge over the competition.

Fabrice Roghé is a senior partner and manag-ing director in the Düsseldorf office of The Boston Consulting Group and the leader of the firm’s People & Organization practice in Europe and the Middle East. You may contact him by email at [email protected].

Sukand Ramachandran is a partner and managing director in the firm’s London office. He is a coleader of AI@BCG and is an active member of the Technology Advantage practice in Western Europe. You may contact him by email at [email protected].

Yann Letourneux is a director in BCG’s Paris office and a core member of the Operations, Technology Advantage, and Financial Institu-tions practices. You may contact him by email at [email protected].

Nicholas Clark is a principal in the firm’s London office. You may contact him by email at [email protected].


After decades of false starts and unfulfilled expectations, artificial

intelligence (AI) has now gone mainstream. Companies are successfully applying AI to a wide variety of current and novel processes, products, and services, and this success comes none too soon; AI is essential for business to address the ballooning complexity brought on by digitization and big data.

It is deceptively easy to launch aI pilots and achieve powerful results.

Visionary executives have started to imagine the potential for implementing AI throughout their companies. For example, BCG estimates that AI could help the top ten banks generate an additional $150 billion to $220 billion in annual operating earnings.

As pioneers across industries strive to reap these rewards by scaling up AI, however, they are stumbling against what we call the “AI paradox”: it is deceptively easy to launch AI pilots and achieve powerful results. However, it is fiendishly hard to move toward “AI@scale.” All sorts of problems arise, threaten-ing to undercut the AI revolution at its in- ception.

The paradox is easy to explain but hard to re-solve. AI forces business executives to deal si-multaneously with technology infrastructure as well as more traditional business issues. The core challenge is a tightening Gordian knot. Typical IT systems consist of data input, a tool, and data output. These systems are relatively easy to modularize, encapsulate, and scale. But AI systems are not so simple. AI algorithms learn by ingesting data—the training data is an integral part of the AI tool and the overall system. This entanglement is manageable during pilots and isolated uses but becomes exponentially more difficult to address as AI systems interact and build upon one another.

What sounds like a mere technical issue has multifaceted implications and challenges for companies. For example, vendor manage-ment becomes both more strategic and more complex. For the foreseeable future, vendors will play a large role in deploying AI because they have hired so much AI talent and have even promised to help with entanglement. Companies need to prepare for working with AI vendors in ways that do not put their data at risk or create long-term dependencies but instead strengthen competitive advantage.

People challenges also loom large for AI@scale. On one hand, companies confront a scarcity of AI data scientists and systems en-gineers; on the other hand, current employ-

FOCUS

THE Big lEaP ToWarD ai aT sCalEby Philipp Gerbert, Sukand Ramachandran, Jan-Hinnerk Mohr, and Michael Spira


ees are concerned about interacting with ma-chines and even about losing their jobs.

Also, the organizational demands of AI re-quire a delicate three-dimensional balancing act. Data governance, core AI expertise, and system management should be centralized. The development of use cases, learning, and training should occur in business units or functions and be managed by agile, cross- functional teams. Finally, AI action remains decentralized—in the marketplace, on the shop floor, or in the field.

Below, we provide a roadmap to systematical-ly resolve the AI paradox. The AI@scale pro-gram is a full transformation to a new operat-ing model.

a systematic approach to ai@scaleWhile strategy and AI use cases are not the focus of this publication, they are critical to AI@scale transformations. So that is where we start. (See Exhibit 1.)

Strategy for AI@scaleStrategy setting in an era of disruption is daunting, and AI is the epitome of disruption.

So while it makes sense to initiate an AI strat-egy within or close to the current business model, companies must also take into ac-count the evolution of value pools in the wid-er market as other players embrace AI.

The first step in that process involves study-ing the potential and overall effect of AI use cases for both internal processes and external offerings. When we ran such an analysis for the top ten banks worldwide, it generated the estimate (referenced earlier) of up to $220 billion in additional annual operating earnings with AI. (See the sidebar “How AI Could Fat-ten the Bottom Line.”)

The second step—understanding the evolu-tion of value pools—is equally important. The rest of the world is not standing still, pas-sively watching these rewards being realized. Other companies both within and outside a company’s traditional industry are focusing on how AI can help them attack value pools.

In a prior publication, we illustrated potential value shifts in the health care industry among biopharma companies, insurers, providers, medtech companies, new entrants from the technology industry, and consumers. (See Put-ting Artificial Intelligence to Work, BCG report,


DEVELOP STRATEGY FOR THE AI@SCALE ERA Competitive advantage Shifting value pools

BUILD ANDDEPLOYAI USE CASES

Data Processing Action

AI building blocks

IDEATE AND PRIORITIZEAI USE CASES Va

lue

Speed

TRANSFORM OPERATINGMODEL FOR AI@SCALE

People, skills, and processes Organization

Industrializing (data) infrastructure AI ecosystem


Exhibit1|DeployingAI@scaleRequiresTransformation


AIcandrivetremendousvaluecreation.Weillustrateitspotentialbyestimatingthebottom-lineimpactithashadonthetenlargestglobalbanksthroughincreasedoperatingrevenuesandreducedoperatingexpenses.AIcanalsoimproveriskmanage-ment,butitsvaluedependsoncountry-specificregulations.Inindustriesotherthanbanking,moreover,AIcanhelpoptimizecapitalexpenditures.TheexhibitbelowshowstherevenuegainsandcostsavingsthatAIcanproducewithincorpo-rateandretailbanking.

AIcanimproverevenuesbyattractingnewcustomers,increasingtheirshareofwallet,andoptimizingpricing.SomeoftheAIap-plicationstoboostrevenuesincludemicro-targeting,personalizedofferings,chatbotsandrobo-advisors,portfolio-basedriskalerts,loanlossestimation,willingness-to-payevaluations,andadvancedfeeschemesimplementation.Thesemeasurescouldliftrevenues(orwhat’scalledoperatingincomeinbanking)by$120bil-lionto$180billionannuallyatthetenlarg-estbanks.Assuminga50%ratioofcoststorevenuesatthetoptenbanks,$60billion

to$90billionwoulddroptothebottomlineonanoperatingbasis.(Inbanking,thisisknownasthecost-to-incomeratio.)

AIcanlowercostsbyautomatingavarietyoffront-andback-officeactivities,includingsalesandservice,riskmanagement,accountcreation,transactionprocessing,operations,andsupportfunctions.SomecommonusesofAItoreducecostsincludeapplicationstoaddressmoneylaunderingandfrauddetection,workforcemanage-ment,andexpenseauditing.AI-enabledsavingscouldreach$90billionto$130bil-lionatthetenlargestbanks.

Thecombinationofrevenueboostsandcostsavingscouldgenerateanongoingannualincreaseinoperatingearningsof$150billionto$220billion.Ofcourse,bankingisanindustryinflux,soprojec-tionsbasedonthecurrentmarketenviron-mentmaynotbefullyrealized.

HOWAICOULDFATTENTHEBOTTOMLINE

$120 billion to $180 billion annual increase in operating revenues translates into a $60 billion to $90 billion increase in earnings1

$90 billion to $130 billion annual decrease in operating expenses

$150 BILLION TO $220 BILLION IN ANNUAL EARNINGS POTENTIAL FROM AI

Increase incorporate revenues

116–124%

Current operatingrevenues

100%

9–13%

Increase inretail revenues

7–11%

Revenuepotential with AI

Current operatingexpenses

16–24%

Corporatebanking savings

10–15%

61–74%

Expenses with AIRetail bankingsavings

100%

Support

Service

Operations

Sales

Accounting

Share of walletNew clients

Risk Transactions

Pricing

Sources: BCG corporate banking digital excellence benchmark; SNL Financial; BCG analysis.1Earnings calculation assumes a 50% cost-to-revenue ratio for the top ten banks (referred to as the cost-to-income ratio in banking), not including additional potential from AI-driven operating-expense reductions for additional revenue.

AICouldgenerate$220BillioninAdditionalAnnualEarningsfortheTopTenBanks

September 2017.) A foreshadowing of such shifts occurred in January 2018 when the stocks of major retail pharmacy chains and insurers dropped by up to 5% relative to the S&P 500 after Amazon, JPMorgan Chase, and Berkshire Hathaway announced their entry into health care. Financial services, media, re-tail, as well as mobility and logistics are other industries susceptible to major AI-triggered disruption.

Companies formulating their strategy need to consider both the potential of AI uses and the risk that today’s value pools may shift.

Two warnings are in order in setting strategy.

It’s not all about amassing data. While criti-cal for the success of AI, internal data is not the Holy Grail. For one thing, companies in-creasingly will need to negotiate access to data outside their walls—from suppliers, partners, and customers, for example. Also, new approaches are emerging that avoid the need for massive troves of data. Generative adversarial networks (GANs), for instance, are virtual learning environments in which algo-rithms train one another to solve problems—such as creating and discriminating false pic-tures of faces. “Digital twins,” meanwhile, are virtual replicas of physical assets that enable algorithms to learn without the same real- world data requirements of traditional ma-chine learning.

It’s also not enough to be agile and fast. Flex-ibility and speed are important, but they are not excuses to engage in a random walk through the AI wilds. To achieve competitive advantage, companies need to point in a spe-cific direction. As research we have conduct-ed with the MIT Sloan Management Review shows, the gap between companies believing in the transformational potential of AI and those actually having an AI strategy in place is stunning.

Identifying, Prioritizing, and Deploying AI Use CasesCompanies that have successfully piloted sev-eral AI use cases can get seduced into think-ing that the move toward AI@scale is not that hard. Use cases can be powerful levers to boost performance. And they can help com-

panies adopt agile test-and-learn methods; build transparency and cybersecurity mea-sures into AI systems; and satisfy internal and regulatory requirements.

But isolated use cases provide limited guid-ance for the challenges of navigating through the AI paradox. They can sputter and grind to a halt when interacting at scale unless com-panies transform their operating models.

While critical for the success of aI, internal data is not the holy grail.

As a consequence, companies should initially prune the number of use cases, focusing on the most promising. (See the sidebar “How to Address AI Use Cases.”) There is peril—not safety—in numbers when scaling up AI use cases and pilots.

Preparing for ai@scaleAgainst that backdrop of strategy, priority setting, and piloting, let’s explore how companies can plan for a successful AI@scale transformation.

Creating a Robust AI ArchitectureIT is often treated as a standalone function that is brought in to implement decisions al-ready made by business leaders. This ap-proach does not work with AI. If executives want to deploy machine intelligence to solve essential business problems, they must priori-tize machine architecture equally to “people architecture” or traditional organizational is-sues. Otherwise, the entanglement concerns described earlier will disrupt their plans.

As companies move beyond the pilot phase, AI algorithms systemically interact with one another by learning from data, as explained earlier. Researchers at Google raised the en-tanglement issue several years ago, leading many prominent AI users to develop compre-hensive machine-learning architectures. For example, Uber’s Michelangelo, in the words of the company, is “an end-to-end system that


enables users across the company to easily build and operate machine learning systems at scale.” (See Exhibit 2.) Likewise, a leading European online fashion retailer has built an AI platform that allows individual AI systems to work together and reuse components.

Vendors such as Amazon, Google, IBM, and Microsoft have all now launched or an-nounced platform offerings to AI users. Whether organizations buy one of these emerging platforms from a vendor or build their own, companies that want to deploy AI

at scale need a rigorous and consistent frame-work and system to manage, document, and monitor workflow from data input to final ac-tion. In addition to addressing issues such as entanglement and reusability, for example, companies need to monitor the security of these systems as well as the integrity and com-pliance of automated actions generated by AI.

Moreover, companies need to ensure that they have the storage, computing, and band-width to handle multiple AI engines and their timely actions. Owing to its flexibility, the


Manycompaniesarenotyetapplyingaprovenmethodologyforidentifyingandprioritizingusecases.Instead,theyrunmanyuncoordinatedpilotsandtests,thenfacetremendouschallengesinovercomingtheAIparadox,andoftenmustrevisitandrestructuretheirpreviouswork.

Aglobalautomakeraddressedtheproblemupfrontbycarefullyreviewingandevaluat-ingawidevarietyofoptionsandthenprioritizingthemostcriticalfunctionsandusecasesforachievingthemostimportantbenefits.Thecompanytookatwo-stageapproach:itaimstohitanintermediatesetofmilestonesby2020andreachfullpotentialby2025.

Theexhibitbelowshowshowtocategorizeusecasesalongthedimensionsofspeedofimplementationandpotentialvaluecreation.Easywinsshouldbeaprioritybecausetheyarequickandhighvaluebutalsorare.Theunicorncategorypromisesthemostpotentialbuttakesthelongesttimetoimplement.Smallerinitiativesthatcouldeventuallybecomeunicorns(whatwecall“incubationfields”)makesenseaswaystogainknowledgeandexperienceandembarkonthejourney.Allotherusecasesaredistractionsandshouldbeavoidedordiscontinuedassoonaspossible.

HOWTOADDRESSAIUSECASES

SPEED OF IMPLEMENTATION

EXPE

CTED

VAL

UE

Hig

hLo

w

Low(More than three years)

High(A few months)

Medium(One to two years)

Med

ium

Incubation fields

Unicorns Easy wins

Distractions


ByPrioritizingUseCases,CompaniesCanFocusonEasyWinsandUnicorns

cloud is often a preferred option to address these needs. In industrial automation and In-ternet of Things networks, however, latency and bandwidth constraints may prevent the cloud from serving as a complete solution. These settings require novel structures, such as edge computing, in which part of the pro-cessing power is kept closer to the action in the periphery.

Ultimately, the intelligent algorithms are a relatively small part at the core of the overall system. Consequently, many companies ex-panding their AI operations need only a small number of data scientists and AI ex-perts at the core but require a large number of data and systems engineers with AI experi-ence to ensure the performance and resil-ience of the pipeline and peripheral systems. For instance, with experienced data engineers and architects, a leading oil and gas company has built its own integrated-operations center, leveraging AI to enable predictive mainte-nance and optimize production. The project-ed efficiency gains amount to several billion dollars over the next few years.

Structuring an EcosystemThe question is not if companies will work with vendors and partners but how. At least for the short term, it is their only way to ac-

cess necessary AI talent and capabilities. Even companies with strong AI backgrounds rely to some extent on vendors.

With AI, however, these relationships could jeopardize core sources of competitive advan-tage for the companies. Vendors need to train their AI tools, often using sensitive client data.

Companies can work with AI vendors in many ways, from outsourcing an entire pro-cess to buying selected services, seeking help in building in-house solutions, or training in-ternal staff. Executives should view these op-tions in light of two questions:

• How valuable is the process or offering to our future success?

• How strong is our ownership, control, or access to high-quality, unique data, relative to the AI vendor’s?

By analyzing the AI landscape this way, com-panies will discover that their AI efforts land in one of four quadrants that each requires a different strategy:

• Commodities (low-value potential, low differentiated data access). Manage vendor relationships to reduce costs and


THE RISKS OF SCATTEREDAND ISOLATED AI DEPLOYMENT

No central data and pipelinemanagement for AI systems

Fragmented storage of features,models, and outputs

Few employees with required technical know-how

Unsystematic deployment process for AI systems

HOW UBER'S MICHELANGELO INDUSTRIALIZES AI

Central data lake with data source and pipeline management

AI platform that handles various use cases with central feature store

Many teams building and deploying AI models at once

End-to-end centralized deployment support

MICHELANGELO'SCRITICAL END-TO-END MANAGEMENT OF AI

ENVIRONMENT

Manage data

Train models

Evaluate models

Deploy models

Make predictions

Monitor predictions

Sources: “Hidden Technical Debt in Machine Learning Systems,” Proceedings of the 28th International Conference on Neural Information Processing Systems, Sculley D, et al., 2015; Uber website; BCG analysis.

Exhibit2|Uber’sMichelangeloPlatformHelpsHarmonizeAIDeployment

improve performance in support functions and elsewhere.

• Danger Zones (high-value potential, low differentiated data access). Develop or acquire competitive data sources to turn danger zones into gold mines.

• Hidden Opportunities (low-value potential, high differentiated data access). Lev- erage technical expertise of vendors to generate quick wins and insights.

• Gold Mines (high-value potential, high differentiated data access). Capitalize on superior data access in strategically relevant areas and develop AI independently.

Beyond the critical issues of trust and confi-dentiality, the overall commercial offerings of vendors—including their newly launched platforms—differ widely in capabilities, inte-gration abilities, and ease of use. So compa-nies need sufficiently solid AI capabilities to select and manage these relationships. Many business leaders ultimately choose to hire an independent advisor to assess the tradeoffs while they build up the internal capability to bring work in-house.

Developing People, Skills, and ProcessesCompanies moving toward AI@scale should carefully consider how the transformation

will affect their workforce through the cre-ation of skills shortages, elimination of jobs, or both. This complex topic requires intensive strategic workforce planning, retraining, and process re-engineering to accommodate a human-machine world.

The most immediate need is critical technical skills. While many companies focus on hiring data scientists, they are encountering a great-er shortage of people who have both business skills and an understanding of AI, as well as systems and data engineers, as mentioned earlier.

Longer term, companies need to understand the fundamental and nuanced shifts in the workplace that occur as humans and ma-chines work side by side. (See Exhibit 3.) When AI is focused on the augmentation of employees in their current roles, changes to the workforce and processes are relatively modest. Indeed, the introduction of augmen-tation in many internal processes and service functions often improves employees’ experi-ence by enhancing the quality of information at their disposal. In financial services, for ex-ample, the use of AI is lowering the cost of many processes.

These changes become more extreme as pro-cesses start being reimagined for true human- machine collaboration. In radiology, doctors and hospitals are already starting to prepare


Inte

ract

ion

Operationalimplications

Example

Dominant decision maker

Underlying approach

Humans’contribution

New superfast approachusing virtualized-learning

environments and digital twins

Flight simulators for drones

HYPERLEARNING

Machine

New

Principles

Enhancementof current process

AI document management

HUMAN AUGMENTATION

Human

Existing

Details

Systematic change managementincluding process redesign

Medical-imaging diagnostics

TRUE HUMAN-MACHINECOLLABORATION


Exhibit3|ThePatternsandConsequencesofHuman-MachineInteractions

their processes and business models for a time when regulators accept machines as a second opinion—and ultimately a first opin-ion—for diagnostics, with a human in the loop to correct potential errors.

Finally, in what we call “hyperlearning,” machines are capable of teaching themselves in a virtual environment with little or no human intervention and without relying on massive data input from the real world. As discussed earlier, GANs and digital twins are such environments. It is conceivable that algorithms will not need real-world data and training—or even human intervention—in these environments in the near future. This evolution has already occurred in the relatively restricted environment of chess but also, for instance, in flight simulators for drones.

As companies move along this continuum, they will need to address new and novel change management and reskilling demands. Renault has already embarked on this path with its digital transformation and with the creation of its Digital Hub, a large center ded-icated to the development of digital products.

Designing Governance and Organizational StructuresGovernance and organization of any major transformation are heavily dependent on context and goals. However, all AI@scale transformations require the general three-dimensional structure mentioned in the introduction. (See Exhibit 4.)

Corporate Center. Certain types of expertise and governance should be centralized. These include excellence hubs that house experts in many of the (horizontal) AI building blocks—such as machine vision, natural-language processing, and some general frontier topics in machine learning—and support AI initia-tives and vendor assessment throughout the company.

Data is the raw material of AI, but it also con-tains some of the company’s most sensitive information. A world-class data governance function, which sets data permissions across the organization, is critical in an AI world, to ensure both competitive advantage and regu-latory compliance.

Finally, data architects, who manage global topics such as industrialization frameworks and platforms as well as vendors, should be centralized along with cybersecurity experts.

Business Units and Functions. As a general rule, cross-functional agile teams situated within business units or functions should be responsible for the development of AI-enabled processes, products, and services. These teams should include both AI experts and topic specialists. Roving “SWAT teams,” including HR experts and change specialists, should also be available to help decentralized units affected by AI use cases with implemen-tation, training, and other issues.

Field Level. The employees responsible for managing the processes and actions modified


CORPORATE GOVERNANCEAND SUPPORT HUBS

Datagovernance team IT architectsAI center of

excellence

CENTRALIZED LEARNINGAND USE-CASE BUILDING

Business ownersand use-caseteams

SWAT teamsfor change

DECENTRALIZED ACTIONAND IMPLEMENTATION

Organizationsaffected by AI


Exhibit4|AI@scaleRequiresChangesinOrganizationandgovernance

by AI are, of course, decentralized, located close to the market, shop floor, or field. These units will need to understand the new tools and how they affect processes and skill requirements and will carry the principal burden of change. The SWAT teams will be available to bring these people up to speed on how to leverage the new technology and assist with the transformation.

Pulling Together the Transformation ProgramAn AI@scale transformation should occur through a series of top-down and bottom-up actions to create alignment, buy-in, and follow- through. This has ensured the successful in-dustrialization of AI across companies and their value chains:

• AI Ambition and Maturity Assessment. This top-down step establishes the overall context of the transformation and helps prevent the company from pursuing disconnected AI pilots. The maturity assessment is typically based on a combi-nation of surveys and interviews.

• Evaluation of AI Initiatives and the Operating Model. This bottom-up step provides a baseline of current AI initia-tives. It should include goals, business cases, accountabilities, work streams, and milestones in addition to an analysis of data management, algorithms, perfor-mance metrics, and cybersecurity. A review of the current AI operating model should also be conducted at this stage.

• Priority Setting and Gap Analysis. The next top-down step prioritizes AI initia-tives, focusing on easy wins and unicorns. This step also identifies the required changes to the operating model.

• Outline of AI@scale Transformation Program. This top-down step consists of both the transformation roadmap, including the order of initiatives to be rolled out, and the creation of a program management office to oversee the transformation.

• Detailed Implementation Planning of AI@scale Program. The last step covers

implementation, detailing the work streams, responsibilities, targets, mile-stones, and resources.

By systematically moving through these steps, the implementation of the AI@scale transfor-mation will proceed with much greater speed and certainty. Still, companies must be aware that the transformation can take one to three years, depending on the complexity of the overall organization.

AI is a promising and alluring technology. Many companies have had early success

with pilots but have fallen victim to the AI paradox. An AI@scale program helps ensure that companies will methodically address the multifaceted dimensions of a full transforma-tion and reap the outsized benefits.

Philipp Gerbert is a senior partner and managing director in the Munich office of The Boston Consulting Group and a BCG Henderson Institute Fellow exploring the impact of AI on business. He coleads the firm’s AI@scale initia-tive and leads its global digital strategy topic. You may contact him by email at gerbert.philipp @bcg.com.

Sukand Ramachandran is a partner and managing director in the firm’s London office. He is a core group member of BCG’s Technology Advantage practice and coleads the AI@scale initiative. You may contact him by email at [email protected].

Jan-Hinnerk Mohr is a principal in BCG’s Berlin office. He is a BCG Henderson Institute ambas-sador exploring the impact of AI on business and a core member of the firm’s Technology, Media & Telecommunications practice. You may contact him by email at [email protected].

Michael Spira is a project leader in the firm’s Munich office. He is also a BCG Henderson Institute ambassador exploring the impact of AI on business and a core member of BCG’s Energy practice. You may contact him by email at spira [email protected].


Here’s a complaint we’re hearing more and more

frequently from client CEOs. “I’ve got a bunch of agile pilots going. Many are actually working. I’m seeing results, but they’re not transforming the company. They’re not having the full impact I expected.”

There’s no question that making the transition to agile ways of working, especially at scale in a large and complex organization, is a tall order. Agile transformations that stall or fail to reach the trans-formational tipping point are common.

Successful lighthouse projects can establish momentum and achieve scale as the results roll in and the organization gains understanding of what agile can accomplish. How-ever, the extra push from the CEO—as well as the leadership team—can play a crucial role in getting over the inevitable hurdles. On the basis of our experience, we have identified five things that suc-cessful leaders do not only consis-tently but also consistently well to keep their transformations on track.

start with WhyNew ways of working are not goals in themselves: they are a means to an end. To mobilize the troops and keep everybody on track when times are rough, leaders articulate why they want to change: What are the objectives and desired out-comes? Changing is hard work, and in the face of difficulty, people tend to gravitate toward the old and familiar ways. Moving beyond the tipping point in agile transfor-mations requires a shared convic-tion that continuing the way the company has always done things is not good enough. Setting new standards in customer service, get-ting products to market before competitors, disrupting the indus-try, achieving step changes in pro-ductivity, and winning the war for talent are all examples of objec-tives and outcomes that organiza-tions can rally behind. People need to know why the new ways of working are critical.

adopt the Principles, adapt the Practices At its core, agile is a set of cultural values, principles, and behaviors, rather than a set of specific prac-

tices. Adopting values, principles, and behaviors can be done suc-cessfully only by shaping the con-text in which people work.

Depending on the nature of the underlying work, the contextual solutions will differ. New ways of working on the interface of busi-ness and IT will likely include ele-ments of, for example, agile scrum and design thinking, while custom-er service and operations activities may benefit more from lean or the self-management practices of Holacracy.

For an individual team or a start-up, specific practices and ceremo-nies go a long way toward leading people to “live” the aspired-to cul-tural values and principles. But spreading values and principles across hundreds of teams in a large and complex organization re-quires addressing almost all ele-ments of the operating model to set the context in which agile be-haviors can take hold and thrive. This level of change almost cer-tainly needs to be driven by the CEO, as it often includes wide-spread alterations in such areas as governance and funding models,

VIEWPOINT

HoW CEos kEEP agilE TransForMaTions

Movingby Martin Danoesastro, Benjamin Rehberg, and Grant Freeland


organization structures, incentives and performance management, lo-cation and sourcing strategy, mea-surement frameworks, and techno-logical enablers.

Change Your leadershipNothing leads like an example. Leaders should push themselves to adopt agile ways of doing things and publicly demonstrate their own changing behaviors. The pub-lic aspect is important. To signal change at one company, executives in certain divisions gave up their offices so that the company could create team rooms. At another, ex-ecutive team members committed to holding weekly “town hall” meetings in the company café. In addi-tion to providing general business updates and answering questions at these meetings, individual exec-utives spoke about their personal- development agendas.

Getting experienced leaders to change their behaviors is not easy. After all, agile behaviors are not the behaviors that propelled most of them into senior leadership po-sitions in the first place. Many leaders find that they are required to unlearn what made them suc-cessful, and for more than a few, this is a bridge too far. CEOs need to recognize that agile transforma-tions almost certainly require at least some changes in the composi-tion of the leadership team. These are tough decisions because the executives in question haven’t done anything wrong. Still, making the necessary changes has two benefits. Most important, it puts the people who will help drive the transformation into critical roles. It

also sends an unmistakable mes-sage to others who might be sitting on the fence of change: they need to get with the program before they encounter a similar fate.

align to Empower Small, cross-functional, empow-ered teams are at the core of every agile organization. The ability to act autonomously spurs ownership and creativity, enabling teams to make quick decisions and move fast. But a high degree of autono-my works only when there is also a high degree of alignment in and among teams. A key role of agile leaders is to ensure strong align-ment around overall company pur-pose, strategy, and priorities. Lead-ers need to communicate their intent, explaining both the why and the what, and they need to let go, releasing their teams to figure out how to address their specific

assigned challenge. The more alignment that leaders are able to establish, the more autonomy they can afford to give. Leaders can spread and reinforce alignment in a variety of ways that include mod-eling their own behavior and strengthening governance mecha-nisms, measurement frameworks, and performance management practices.

learn and adapt—at speedAgile puts a premium on feedback and lessons learned: adapting to change is more important than fol-lowing a plan. This rule applies even to the agile transformation itself. Of course, the transforma-

tion must be well thought through and carefully planned, but leaders must also be open to modification and adjustment along the way. In-evitably, there will be setbacks and challenges, but strong leaders are the ones who have the ability to learn, adapt, and change course when things go awry.

Transformations, which are often described as journeys, can take two or three years. We are seeing more CEOs tighten up that time frame—in some instances to less than a year. The biggest counter-weight to resistance is momentum, and nothing builds momentum like speed. Like loosening controls, moving fast can seem risky. But the bigger risk to the transforma-tion lies in not changing fast enough.

Martin Danoesastro is a senior partner and managing director in the Amsterdam office of The Boston Consulting Group and global colead-er of agile at scale for the People & Organization practice. You may con-tact him by email at danoesastro [email protected].

Benjamin Rehberg is a partner and managing director in the firm’s New York office and global coleader of ag-ile at scale for the firm’s Technology Advantage practice. You may contact him by email at rehberg.benjamin @bcg.com.

Grant Freeland is a senior partner and managing director in BCG’s Bos-ton office. He is a core member of the People & Organization practice and previously served as global leader of that practice. You may contact him by email at [email protected].

leaders should adopt agile ways and publicly demonstrate their own changing behaviors.

28 | BCG Technology Advantage—Viewpoint

From industry to industry, agile is spreading in fits and

starts—from the IT department to other functions and, in more and more cases, companywide. Or it is trying to. Lots of companies attempt the transition to agile ways of working but end up with either something that is agile in name only or a hobbled hybrid organization that exacerbates the problems they set out to solve.

There’s good reason for the inter-est in agile. When large companies get agile right, the results can be stunning. Productivity can improve by a factor of three. Employee en-gagement, measured in quantita-tive surveys, increases dramatically as well. New product features can be released within weeks or months rather than quarters or years. Rates of innovation rise, while the number of defects and do-overs declines.

There’s equally good reason for the failures. Agile is hard—really hard. Done right, the transforma-tion affects everything from inter-nal processes to how employees spend their day to how people in the organization interact with one

other. It requires rethinking struc-tures, reporting, compensation, and career paths.

But most established organizations like the status quo and fear change. So they try to kill the transforma-tion before it gains traction.

Many management teams under-stand this, so they’re hesitant to take on an agile transformation. Then one company in their indus-try gets agile right, others see the power of what agile can accomplish when it’s done right, and naturally they want to do the same thing. But they tend to overlook the fact that the leader that made the successful agile transformation spent years planning and executing its agile journey. Its top management was fully committed to making the change and was willing to experi-ment and learn from lots of failures along the way. When the followers don’t make the same commitment to planning and execution, they are likely to fall into one of three traps.

name onlyThe name-only trap may be the most common: companies under-

take an organizational change that they label agile, but they don’t make the kind of fundamental shifts in ways of working—estab-lishing cross-functional teams and institutionalizing a try-and-fail ap-proach, for example—that are the basis of agile. In large and complex companies, speeding things up means reducing reliance on hierar-chies, and leveling hierarchies re-quires reliance on collaboration. For many companies, this is a big organizational and cultural change, and implementation is dif-ficult. Companies that get it right establish the boundaries within which empowered employees can collaborate and make great things happen. But most companies con-tinue making decisions the way they always have—slowly.

Two TierA variation on the name-only trap is the two-tier trap. There are a couple of manifestations. One is that the organization is redesigned around agile ways of working, but senior management continues to do things the way they have for years. Another occurs when a com-pany shifts some functions to agile

VIEWPOINT

agilE TraPsby Grant Freeland, Martin Danoesastro, and Benjamin Rehberg


while other functions continue op-erating in the old ways of working. In either case, if the transformation is successful, the organization, or parts of it, moves much more quickly and nimbly. But the bene-fits are lost when the results of ag-ile teamwork—a product innova-tion, for example, or a faster internal process—run hard into tra-ditional processes and deliberate, drawn-out management approvals. The result is not dissimilar to a sprinter running into a wall—pain and even injury are inevitable.

Half MeasuresCompanies that fall into the half- measures trap get agile partially right. They are successful at cross-functional reorganization. Multidisciplinary teams start to work in scrums and sprints. But these companies do not follow

through with critical organization-al enablers such as redesigned ca-reer paths and incentive programs. Employees adopt the new ways of working only to encounter uncer-tainty about the impact these nov-el methods will have on them as individuals. The initial enthusiasm stalls. Organizational change is not linear; companies need to reach a critical mass of change in order to reap the benefits. Rather than speeding up processes, decision making, and results, the half-measures trap—like the other traps—leaves the organization less productive than it was before.

Golfers will tell you that traps are easy to slide into but hard

to hit out of. The key to avoiding bunkers is good planning and bet-ter execution. That’s the only way to avoid agile traps too.

Grant Freeland is a senior partner and managing director in the Boston office of The Boston Consulting Group. He is a core member of the People & Organization practice and previously served as global leader of that practice. You may contact him by email at [email protected].

Martin Danoesastro is a senior partner and managing director in the firm’s Amsterdam office and global coleader of agile at scale for the People & Organization practice. You may contact him by email at [email protected].

Benjamin Rehberg is a partner and managing director in BCG’s New York office and global coleader of ag-ile at scale for the firm’s Technology Advantage practice. You may contact him by email at rehberg.benjamin @bcg.com.

30 | BCG Technology Advantage—Viewpoint

Digital disruption in the automotive industry has created enormous op-portunities for automakers. However, to keep pace with change in the digi-tal era, companies must modernize. For Renault, a French automobile manufacturer with roots that reach back to 1898, the imperative to inno-vate is in its DNA. That doesn’t mean it’s easy.

As CIO of the Renault Group since May 2016, Frédéric Vincent has over-seen the company’s comprehensive digital transformation. Entering a multinational company with more than 2,000 legacy applications, most of which dated back 10 to 15 years, Vincent had his work cut out for him. But he maintained a sharp focus on creating measurable, tangible value for the business and the consumer, and his efforts are paying off hand-somely. Renault anticipates that its digital transformation will garner €1 billion in value by 2020.

Recently, Vincent sat down with An-toine Gourévitch, a senior partner and managing director in the Paris office of The Boston Consulting Group and the global leader of data and digital platforms. Edited excerpts from that conversation follow.

Thanks very much for joining us today, Frédéric. Tell us a bit about what you do at Renault.

I have two roles at Renault. The first, as CIO, is to manage IT, and the second, as president of Renault Digital, is to ensure that the Renault Group is transforming itself toward digital. We are at the heart of a unique transformation. Digital is massively entering the automotive industry—both upstream, in vehicle design and manufacturing, and downstream, with customer relationships. So there is an enormous digital transformation to be made at Renault.

Please describe your role in Renault’s digital transformation.

The first step was to communicate the value that digital could bring to the entire company. We did a lot of work to identify—and quanti-fy—all digital projects that could generate value for the Renault Group. We found that digital could deliver around €600 million per year: 50% from gains in efficiency and productivity and 50% from new businesses. So we had a clear objective: capture €600 million in value. It was really the first step to-ward onboarding the whole com-pany. We focused on value rather than the transformation itself.

Q&A

rEnaulT’s Billion-Euro DigiTal

TransForMaTionAn InTerVIew wITh CIO FrédérIC VInCenT

At a Glance

Education•1992,BA,ÉcoleCentraleParis

Career Highlights•2016–present,Renaultgroup,CIO•2013–2016,CANAL+group,chieftechnologyandITofficer

•2007–2013,CANAL+group,variouspositions

FrédérIC VInCenT


Once we’d calculated the value of digital for Renault, we asked, How do we capture €600 million in the different business lines over the next three years?

What steps have you taken to capture that value?

We have three main levers.

The first is technology oriented. We had to build a digital platform that would work alongside our leg-acy systems and allow us to get value from all of our data. Our data is spread across more than 2,000 applications, so we needed a digital platform that would consol-idate the data and make it usable.

Second, we changed our ways of working. We decided to use agile for all of our new projects. This is quite new for our staff, so scrum masters and agile coaches were

needed to explain how to work in an agile way. We also upgraded our working tools within the com-pany to facilitate real-time collabo-ration and information sharing for our teams, since so many of our employees are traveling and work-ing all over the world.

Finally, we had to hire new IT tal-ent, as well as upskill our existing teams to work with the new tech-nologies. To accelerate the trans-formation, we created a digital hub called Renault Digital. It’s a Re-nault subsidiary, composed of 300 people. Its purpose is to pro-vide the Renault Group with the skills, methodology, and working environment that can facilitate digital transformation.

When a company creates a digi-tal hub, there’s a danger that it can become isolated. How did you integrate the digital hub within the company as a whole?

Indeed, we were aware of two po-tential pitfalls in creating a digital hub. The first risk was that the hub could become a completely isolat-ed silo and largely irrelevant. This is not hard to imagine, since a group of 300 employees is obvious-ly a small number within a compa-ny of 130,000 employees. At the other end of the spectrum, there was a risk that the hub could be completely absorbed into the com-pany and could get lost in the in-tricacies of Renault’s organization and processes.

We worked continually to find a balance so that Renault Digital could be both autonomous and well integrated. To ensure its auton-

omy, we made it a 100% Renault subsidiary, installed it in a dedicat-ed building, and gave it its own hu-man resources and purchasing structures. Because I chair the sub-sidiary, and I’m also in charge of IT, I can ensure strong collaboration between the Renault Group and Renault Digital. Also, the executive committee monitors the hub’s ac-tivities in its monthly meetings to make sure that key Renault depart-ment heads are heavily involved.

Do your IT people feel that there is a split, with some teams working on legacy IT systems and others working on digital?

That’s a risk, but all of the projects we run continue to be steered by

the IT department. When projects require newer digital skills, we take these projects and incubate them within Renault Digital, keep-ing the team intact. So we take a team of business and IT employ-ees and we physically move them into Renault Digital. This allows the team to work with people who can provide a unique mix of digital skills, such as scrum masters, agile coaches, and experts in UI [user interface], UX [user experience], data, and design thinking. And we employ what we call training by doing: that is, the team is trained in a specific methodology by exe-cuting its project. So it’s really an incubation system in which Re-nault Digital provides methodolog-ical and technical skills to Re-nault’s teams.

Can you give an example of an executed digital project that has had an impact?

Yes, I can give two examples.

The first is around vehicle manu-facturing. Today, when you look at a plant, our teams are organized into what we call elementary working units. They are small teams of 10 to 12 people who gen-erally work on the mounting or as-sembly chain, and they are headed by a unit leader. This unit leader should be close to his teams to en-sure that the work is being done in good conditions and that the quali-ty is high. However, he regularly had to go back to his office, be-cause all the information he need-ed for his work was stored on his computer, which was generally far from the production chain. So he was traveling about 12 kilometers a day, going back and forth be-tween his office and his teams.

We decided to make his life easier by moving all the information he needs onto a tablet. The principle

to accelerate the transformation, we created a subsidiary to provide the skills and environment.

32 | BCG Technology Advantage—Q&A

is quite simple. We wanted to re-duce the amount of time he spends walking back and forth, be-cause he can use that time to inter-act with his teams and check quali-ty. So we equipped the plants with Wi-Fi, configured our legacy sys-tems to display information on a tablet in real time, and trained the teams to use the new system.

The gains are easily measurable. We know that the unit leaders save time, and we know the quali-ty of their work is higher. The unit leaders are very satisfied, because it enables them to work more closely with their teams. It took only a few months to put in place. We are experimenting with this at an industrial site in Valladolid, Spain, and we plan to scale it to the whole company.

The second example is on the com-mercial side. On our websites, we have a small module that enables consumers purchasing a new vehi-cle to obtain a price quote that is based on their turning in a used vehicle. It used to be that we sim-ply published a guide that quoted average prices for used cars, but we totally redesigned this experi-

ence for our customers. If you have a used vehicle, you simply enter the registration plate on the web-site. From there, we ask you to compute its mileage and the gener-al condition of the vehicle (if it has scratches, for example). All this is highly visual; it can be done in a few clicks. You enter your name and address, and within seconds, you receive a proposal, which is contractual. The closest dealer contacts you, and you go to the dealership. The dealer checks the vehicle’s condition and buys it from you at the agreed-upon price.

We put this module into produc-tion in France recently, and it mul-tiplied touchpoints by a factor of 20. Obviously, these touchpoints are important because they gener-ate traffic in the network for fu-ture sales. So just by changing this module and allowing for a contrac-tual commitment from the dealer-ship, we dramatically multiplied the number of contacts initiated by our customers.

With just a few months’ work, we can develop these new products that are highly adapted to the business. The business can judge

in a few weeks whether or not the product is efficient, and then we can correct, adapt, and put it into production very quickly. And these products will obviously contribute to the value I was talking about before—the annual €600 million in value from new digital initiatives.

And where does Renault stand today in terms of value?

We started this in January 2017, so we are just through the first year. But we are in line with our fore-casts—and even beyond that for the sales and marketing team. So we have revised our ambitions. Now we expect €1 billion by 2020. The value of digital transforma-tion is here, and we measure it ev-ery day.

Frédéric, thank you very much. Thank you, Antoine.

Antoine Gourévitch is a senior part-ner and managing director in the Paris office of The Boston Consulting Group and the global leader of data and digital platforms. You may con-tact him by email at gourevitch [email protected]



noTE To THE rEaDEr

AcknowledgmentsThe authors of “The Big Leap Toward AI at Scale” thank Sebastian Steinhäuser, Henri Humpert, Johannes Knebel, and Lukas Rapp for support. They also thank Tilman Buchner, Sylvain Duranton, Christoph Gauger, Antoine Gourévitch, Andreas Gräf, Christoph Hartmann, Derek Kennedy, Rolf Kilian, Ryoji Kimura, Andreas Lundmark, Roger Premo, Arun Ravindran, Martin Reeves, Thomas Reichert, Michael Ruhl, and Daniel Schlecht for their invaluable comments and contributions, as well as Ralf Herbrich, director of machine learning at Amazon.

The authors of “The $1 Trillion Opportunity in Digital Support Functions” thank Laura Muller for her help in pulling together the article content.

The authors of our agile coverage thank BCG agile practitioners around the globe for contributing their experiences.

The BCG Technology Advantage authors also thank Astrid Blumstengel, Meghan Huff, and Stuart Scantlebury for their contributions to this publication and Katherine Andrews, Gary Callahan, Alan Cohen, Catherine Cuddihee, David Duffy, Kim Friedman, Jeff Garigliano, Abby Garland, Trudy Neuhaus, Mike Sisk, Amy Strong, and Mark voorhees for their help in writing, editing, design, and production.

For Further ContactNicholas ClarkPrincipal BCG London +44 20 7753 [email protected]

Martin Danoesastro Senior Partner and Managing Director BCG Amsterdam +31 20 548 4000 [email protected]

Marc Roman Franke Project LeaderBCG Berlin+49 30 28 87 [email protected]

Grant FreelandSenior Partner and Managing Director BCG Boston+1 617 973 [email protected]

Philipp Gerbert Senior Partner and Managing DirectorBCG Munich+49 89 231 740 [email protected]

Antoine GourévitchSenior Partner and Managing DirectorBCG Paris+33 1 40 17 10 10 [email protected]

Michael GrebeSenior Partner and Managing DirectorBCG Munich+49 89 231 [email protected]

Yann LetourneuxDirector BCG Paris office +33 1 40 17 10 [email protected]

Michael LeyhLead Knowledge AnalystBCG Düsseldorf +49 2 11 30 11 [email protected]

Vanessa LyonPartner and Managing DirectorBCG Paris+33 1 40 17 10 [email protected]

Jan-Hinnerk MohrPrincipalBCG Berlin+49 30 28 87 [email protected]

Sukand RamachandranPartner and Managing DirectorBCG London+44 20 7753 [email protected]

Benjamin RehbergPartner and Managing DirectorBCG New York+1 212 446 [email protected]

Fabrice RoghéSenior Partner and Managing DirectorBCG Düsseldorf+49 2 11 30 11 [email protected]

Anne-Françoise RuaudAssociate DirectorBCG Paris+33 1 40 17 10 [email protected]

Michael RüßmannSenior Partner and Managing DirectorBCG Munich+49 89 231 [email protected]

Michael SpiraProject LeaderBCG Munich+49 89 231 [email protected]


© The Boston Consulting Group, Inc. 2018. All rights reserved.

For information or permission to reprint, please contact BCG at [email protected].

To find the latest BCG content and register to receive e-alerts on this topic or others, please visit bcg.com. Follow The Boston Consulting Group on Facebook and Twitter.11/18

BCg Tech

nology a

dvantage

novem

ber 2018

bcg.com

bcg technology advantage november 2018 · 5% before the start of the transformation. so how,...

Documents