making data analysts self-sufficient: a case study of amaysim

36
© 2014 Alteryx, Inc. | Confidential Making Data Analysts Self- Sufficient at Amaysim Brian Dirking Director of Product Marketing, Alteryx 1 Adrian Loong Business Intelligence Manager, Amaysim Brandon Chavis, Solutions Architect, AWS Network Dustin Smith, Product Marketing Manager, Tableau Julian Dell IT Director, Amaysim

Upload: alteryx

Post on 15-Jul-2015

4.849 views

Category:

Data & Analytics


1 download

TRANSCRIPT

Page 1: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

© 2014 Alteryx, Inc. | Confidential

Making Data Analysts Self-

Sufficient at Amaysim

Brian Dirking

Director of Product

Marketing, Alteryx1

Adrian Loong

Business Intelligence

Manager, Amaysim

Brandon Chavis,

Solutions Architect,

AWS Network

Dustin Smith, Product

Marketing Manager,

Tableau

Julian Dell

IT Director,

Amaysim

Page 2: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

ENABLING DATA DRIVEN DECISIONS WITH REALTIMEINTELLIGENCE

Presented by: Julian Dell

Page 3: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

3

amaysim is Australia award winning low cost mobile provider dedicated to delivering simplicity, fairness and low prices.

In 4.5 years, Amaysim has gone from a startup to become Australia’s largest MVNO with over 600k customers

Award winning MVNO• Largest MVNO with over 600k customers• Market leading NPS • Money magazine‘s best of the best awards for

2012 and 2013. • Roy Morgan: #1 Mobile phone service

provider 2013 , 2014

About

1. Size of Data• Over 10b call data records to be analyzed• Over 20-30m call data records added

daily

2. Velocity & Complexity of data with multiple data sources

• Livechat• Zendesk• Call data records• Google analytics / Website data• Point of sale data• Exact target

Size of analytics challenge

Introductions

Page 4: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Team Experience and Expertise

Julian Dell – IT Director

• Worked in the role since before launch of amaysim in 2010

• Has been working in IT industry for over 25 years with over 10+ years in senior management roles. Worked in Financial markets, Financial planning, Advertising, Media and Telco industries. Worked in some of the world’s largest companies including Citibank, Deutsche Bank and News Corp

• Has been part of two previous start-ups (Truelocal and Adstream) including building initial launch systems and transforming their IT systems from start-up to mature business IT setups

Expertise:

• IT leadership

• IT strategy

• IT Transformation of startups

• Programme management

• E-commerce and systems development

4

Page 5: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

© 2014 Alteryx, Inc. | Confidential

Introducing Alteryx

Brian Dirking

Director of Product Marketing

5

Page 6: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

© 2014 Alteryx, Inc. | Confidential

Customer Success

Customers across the world

700+

Strong Foundation

Renewal rate with funds

to support innovation

95%+

Corporate Info.

6

Leader in

Data Blending and

Advanced Analytics

Page 7: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

© 2014 Alteryx, Inc. | Confidential

Share

The Alteryx Solution For Analyst Enablement

7

All Relevant Data

Access, cleanse, and blend

data with unique packaged

data options

Rapid design of predictive

analytics with unique

spatial understanding

Packaged Market &

Customer Data

Consumerize the use of

sophisticated analytics

Enrich

Analy

ze

Ble

nd

Te

xt

Page 8: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Fast, simple, petabyte-scale data warehousing for less than $1,000/TB/Year

Amazon Redshift

Brandon Chavis,

Solutions Architect

Page 9: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Amazon Redshift

Easy to Use

• Provision in minutes

• Monitor query

performance

• Point and click resize

• Built in security

• Automatic backups

Priced to let you analyze

all your data

Page 10: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Common Customer Use Cases

• Reduce costs by

extending DW rather

than adding HW

• Migrate completely

from existing DW

systems

• Respond faster to

business

• Improve performance

by an order of

magnitude

• Make more data

available for analysis

• Access business data

via standard reporting

tools

• Add analytic

functionality to

applications

• Scale DW capacity as

demand grows

• Reduce HW & SW costs

by an order of

magnitude

Traditional Enterprise DW Companies with Big Data SaaS Companies

Page 11: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Dustin SmithProduct Marketing Manager

Page 12: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Help People

See and Understand

their Data

Page 13: Making Data Analysts Self-Sufficient: A Case Study of Amaysim
Page 14: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

MAKING DATA ANALYSTS SELF-SUFFICIENT AT AMAYSIM

Presented by: Julian Dell & Adrian Loong

Date: 14/04/2015

Page 15: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

ENABLING DATA DRIVEN DECISIONS WITH REALTIMEINTELLIGENCE

Presented by: Julian Dell

Page 16: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Strategy - Data driven decisions in real time

1. Knowledge is power

2. Smart management =

happy customers & happy

shareholders

3. Get a real competitive

edge - stop looking back

and start predicting the

future

4. A 3 year rolling forecast in

real time – generate real

company wealth by being

reliable

Page 17: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

To make data driven decisions with realtime intelligence

17

Be the business enabler: Transform Data into Insight– Using analytics to drive business performance

Visual and Easy: Make it easy for business users to see trends and ask questions through self-service

Data Visualizations

Recommendations

to improve

business

performance

Management

consulting

Page 18: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Source: http://timoelliott.com/blog/2013/02/gartnerbi-emea-2013-part-1-analytics-moves-to-the-core.htmlhttp://blogs.gartner.com/michael_blechar/2011/10/01/the-heart-of-information-infrastructure-is-the-information-capabilities-framework/

Shifting from reactive to proactive analytics

Page 19: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

How Amaysim has benefited from their analytics program

Workforce Productivity

• We have a 3 person analytics team covering wide span of functions (Finance, Marketing – Customer retention & acqusition, Sales – Retail and Online, Data warehousing, HR)

• By enabling line of business users to quickly build on a baseline of analytics, they can easily solve their own specific business problems quickly and do not have to wait on Business intelligence teams

Reduced time to insight

• We are able to get the data we need faster. Projects that would have taken 2-3 weeks are now down to a day

Data driven decision making

• Line of business users are able to get direct access to their own data in an easy visualization enabling them to able to solve problems faster.

• People can look at the data, do some discovery, and then arrive at an answer

Page 20: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

AGENDA

20

Executive Buy in & Funding the Journey

Cultural change and enabling business leaders to succeed

Technical perspective – Using Alteryx and Redshift in conjunction with Tableau to

bring the best of breeds together

Page 21: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Data Warehouse / Business IntelligenceThe Stack

Source DataIT

ETLData

WarehouseBusiness

ETLVisualizer

& Reporting

• ECC

• CRM

• CSC

• Google analytics

• Sales POS

• The list goes on…

• Alteryx desktop selected

and deployed

• Gives our business the

agility we need

• Business driven rather than

IT driven

• Ability to do predictive

analytics

• Tableau desktop & Server

selected and deployed

• Self serve BI

• Business dashboarding

• Ability to do predictive

analytics

Time to load

Reliability of Data

Query performance

Flexibility for slice/dice

Visualization performance

Data definitions / governance

Data exploration

• CDC used for real

time data replication

• Slower to build

• More reliable

• Redshift selected for

reliability, processing

power and scalability

21

Page 22: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Best of Breed solutions

Redshift gives us the speed and robustness to store and analyse vast volumes of data Alteryx fuels Tableau visualisation, allowing us to quickly iterate gain insights in Tableau

Source

system

1

Livechat

Source

system

2

Zendesk

Amazon Redshift Datamart

External data / Marketing data

Blend & enrich Visualization

• Clean the data

• Apply business rules

• Validate business rules in

teams

3

4

2

1

Comments

Tableau can be used to directly visualise and analyse big data directly from Redshift

1

Alteryx can be used to blend data which is not in the Redshift database

2

Alteryx can apply and validate more complex business rules before visualizing outputs in Tableau

Continuously iterate between Alteryx and Tableau when discovering more data

4

3

Page 23: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Best of Breed solutions

Alteryx is an intuitive workflow for data blending and advanced analytics, including spatial and predictive analytics in a simple and easy to use drag and drop interface

Page 24: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

24

Best of Breed solutionsTableau is excellent for collaboration – We use Tableau server to share dashboards & insights and Tableau desktop is used extensively for slice and dicing

Page 25: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

AGENDA

25

Buy in & Funding the Journey

Cultural change and enabling business leaders to succeed

Technical perspective – Using Alteryx and Redshift in conjunction with Tableau to

bring the best of breeds together

Page 26: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

How do you get buy-in and engage the business?

26Source: The Age http://www.theage.com.au/executive-style/style/stitched-

up/homedelivered-style-for-timepoor-men-20150122-12vua1.html

Page 27: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Address current business painpoints via doing short proof of concepts with new tools

Use cases

Business

description

Current

Painpoints

Exec dashboardHigh level dashboard for management enable

consistent understanding of KPIs across

Amaysim

• No “single” source of truth for Execs

• No current and consistent data definitions

Finance Audit

Wholesale

invoice rec

External data

delivery

Commissions

processing

Need to get read balances for each individual

mobile phone number

• Unable to extract raw XML from Call data files

easily to establish a balance sheet position

Wholesale invoice reconillation with Amaysim

internal data

• Unable to extract raw XML from Call data files

easily

• Different data definitions between Optus and

Amaysim

Data is delivered to external targets with

downstream impact to data marts, other

application systems (Monte) etc.• Exact target (CRM)

• Datamarts

• Datarama (marketing)

• Monte (Scheduling)

General

analytics

• Ad hoc deep dive

• Slice and Dice

General analytics include ad hoc analytics on

portfolio, churn propensity, CLV and other areas

• Business users do not have a wholistic end to

end view of the entire Amaysim business.

• Certain users (eg: Marketing) have resorted to

using their own analytics software

Commissions is calculated to be paid across our

distribution network.

• Identify opportunities to improve commissions

processes

How do you engage the business?

1

Page 28: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

28

Start small, iterate quickly and deliver business value

2 Start small, deliver continuously

• Start small ( 1 desktop license) , prove value and extend outwards

• Iterate quickly & deliver continuously to get management buy-in

• It helps to speak the language…

Source: http://diginomica.com/2013/08/05/tapping-creative-soul-cfo//

How do you engage the business?

Page 29: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

29

Do things that drive top line or bottom line performance

Link it back to Financial performance

CASE STUDY: REVENUE ASSURANCE

How do you engage the business?

• Historically amaysim receives Call data record files from Optus which gets pre-processed before loading to internal IT applications.

• Alteryx was used to perform independent revenue assurance on XML files to verify integrity of revenue accounts

3

610_.gz

620_.gz

Optus

XML

610_1.xml

610_2.xml

610_3.xml

610_4.xml

610_5.xmlunzip

CDR610_1

CDR610_2

CDR610_3

1Combine all

into 1 dataset

2 Eliminate

duplicate CDRs

3 Identify linked

Data CDRs

Re-calculate data

rounding

4

Load Redshift5

Business Rules Data stores

Source systems

Pre-Processing

Visualisation

Optus XML

Page 30: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

30

Visualize and engage business users by keeping the message simple and direct4

Senior execs are often short on time and want to be given a straight answer

• Top left detail should always contain the most important metrics

Give insights not just metrics

How do you engage the business?

Page 31: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

AGENDA

31

Executive Buy in & Funding the Journey

Cultural change and enabling business leaders to succeed

Technical perspective – Using Alteryx and Redshift in conjunction with Tableau to

bring the best of breeds together

Page 32: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Cultural change & enabling business users to self serve

Democratise the Opportunity1

Source: http://www.heynataliejean.com/2009/01/end-of-nerd-era.html

Most business leaders want to use data to make better decisions..

• You don’t have to be an IT specialist to use Alteryx

• Give users access to the tools, use Alteryx & Tableau to show them how much faster you can spot a trend

• 1:1 Training sessions for different stakeholders tailored to their needs

Page 33: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Cultural change & enabling business leaders to succeed

Make it relevant to different stakeholders2

Senior management

• More interested in insights, dashboards and outcomes

• Spend time showing them dashboards built and how opportunities to improve business performance (eg: revenue generation, expense reduction).

Functional specialists & leaders

• Alteryx relevant to functional leaders

• Alteryx to “clean” the data before visualizing in Tableau

• How can it be used to streamline processes

Source: http://www.timeshighereducation.co.uk/news/business-schools-not-first-port-of-call-for-managerial-recruits/2013863.article

Page 34: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Cultural change & enabling business leaders to succeed

Celebrate success: Keep building and iterating3

Existing tools require analysts to have “coding” experience

Few disparate analysts using SQL to deliver analytical solutions

Where we are now

Using ‘Best’ of Breed products• Visualization : Tableau • Data blending: Alteryx• Warehousing : Redshift (WIP)

Using Alteryx & Tableau to deliver :• Revenue assurance• Sim Sales• Customer disconnections &

churn• Port-outs by carrier• CDR analysis

Where we will be

Self serve analytics for all business users• Real time P&L with Slice and

dice with visualization• Assurance over logic

processing (Audit, Commissions)

Advanced predictive analytics • Churn propensity• Customer behaviour

Where we started

Page 35: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Enabling Business success

3 Key Areas

BEST OF BREED PRODUCTS3

EXEC BUY-IN & FUNDING 1

CULTURAL CHANGE & ENABLING BUSINESS LEADERS TO SUCCEED

2

Key Observations

ENABLING

BUSINESS

SUCCESS

• Answer an important business question - Establish the value of Tableau early on by showcasing critical business issues in dashboard

• Keep the message simple and direct• Go beyond dash boarding to engage business leaders on

portfolio observations

• Democratize the Opportunity - Gather people of interest together from Sales, product, marketing

• Make it relevant to different stakeholders - Include everyone & break down silos

• Celebrate success - Keep people involved and excited

• Combine Redshift (Database), Alteryx (ETL) and Tableau for Visualization

• Tableau is excellent for collaboration – We use Tableau server to share dashboards & insights and Tableau desktop is used extensively for slice and dicing

3 ingredients to enable business success

Page 36: Making Data Analysts Self-Sufficient: A Case Study of Amaysim

Q&A and Next Steps

Download the Visual

Analytics Kit:

Sample analytics workflows

Corresponding Tableau

Visualizations

www.alteryx.com/kit

36

Get the Amazon Redshift Free Trial: http://aws.amazon.com/redshift/free-trial

Download the Amazon Redshift Getting Started Guide: http://docs.aws.amazon.com/redshift/latest/gsg/getting-started.html

Download a Free Trial of

Tableau:

www.tableau.com/products/

trial

Our partner in Australia:MIPPhone: +61 2 9260 0700Email: [email protected]