mobile attribution & marketing analytics - appsflyer · mobile attribution & marketing...

Mobile Attribution & Marketing Analytics for eCommerce

Mobile Attribution& Marketing

Analyticsfor eCommerce

A Practical Guide on How to Let Data Work for You


Mobile commerce is on a tear.There’s no other way to put it.To name

just one study from the many out there — PayPal-Ipsus research

paper covering 22 countries found that mCommerce is growing at

three times the rate of eCommerce.

Between 2013 and 2016 the compound annual growth rate for

m-commerce is projected at 42 percent vs. 13% for eCommerce. To

put growth in actual sales, mCommerce sales are expected to eclipse

$290 billion worldwide by 2016.

With such numbers, it’s no wonder that usage of shopping apps is

the fastest-growing app category, with sessions on shopping apps

on iOS and Android devices increasing by 174% year-over-year,

including 220% on Android alone.

That’s up from 77% last year as more and more retailers are building

apps on top of their mobile-optimized sites despite the hefty price

tag of developing and maintaining a commerce app.

Introduction


2www.appsflyer.com

http://evigo.com/18826-paypal-m-commerce-spendigs-290-billion-2016/

http://flurrymobile.tumblr.com/post/115194992530/shopping-productivity-and-messaging-give-mobile


Mobile Browser vs. App Funnel Metrics

Mobile Browser

Conversion Rate Conversion Rate

App

Products viewed per user

Basket

Purchase Purchase

Basket

Products viewed per userx3.3

x3.0

x1.2

x0.8

Apps also generated almost 50% of mobile transactions for retailers

who have made their app experience a priority. In fact, mobile apps

perform better than any other channel, including desktop.

Why are apps becoming a must-have touchpoint for retailers?

Beyond the fact that mobile is dominated by apps (86% of time spent

on mobile is on apps), there are several factors at play here: speed,

optimal native mobile experience, ability to use mobile payments

and push notifications, not to mention the impact on loyalty and

lifetime value from a user who actually made a commitment to the

brand by downloading its app.

Source: Criteo, State of Mobile Commerce, Q2 2015


3www.appsflyer.com

www.criteo.com/media/2512/criteo-state-of-mobile-commerce-report-q2-2015-letter-digital.pdf

40.media.tumblr.com/9a670fd39f5b9f3ac38bde28b3f27d2f/tumblr_inline_nm47j4qNAh1tpd7xq_500.jpg


Getting Started with Mobile Advertising Analytics?We recommend reading this first:

THE BEGINNERS GUIDE TO

MOBILE ADVERTISING ANALYTICS

Today’s savviest retailers (online & offline) are embracing big data

as a major facilitator of business growth. Consumer information from

anonymous shopping patterns to cross-channel loyalty programs is

collected, sliced and diced to provide a more relevant and seamless

consumer shopping experience - whether from a brick & mortar

location, the website, mobile site or app.

A key communication channel merchants have with consumers

on mobile is advertising, where relevancy and personalization play

a big part. However, unlike the web, the ability to track, measure

and ultimately optimize app install and retargeting campaigns is

a challenge because of the mobile world’s inherent fragmentation

(different OSs, different environments, different devices).

The good news is that advertising attribution and analytics platforms

utilize sophisticated technology to connect the dots and provide retail

app marketers with rich data to enable smart marketing decisions

across the omni-channel user journey. This guide will cover the latest

of what is possible in this area and how it applies to retail marketing.

GET YOUR FREE COPY


4www.appsflyer.com

http://get.appsflyer.com/white-paper-1/http://

http://get.appsflyer.com/white-paper-1/

http://get.appsflyer.com/white-paper-1/


A View From Above

Setting Goals

Defining Key In-AppEvents to Track

Relevant/Personalized Consumer Communication

Smart Consumer Acquisition

Retention & Cohort Analysis

Retention UpliftLTV UpliftROI Uplift

The Bottom Line


5www.appsflyer.com


Let’s Dive In

Defining Rich In-App Events that Matter to Your Business

A retail app, as you’re well aware, is one of the

most complex apps to develop. The diversity

of in-app actions that can be taken by

consumers is vast. The ability to track user

behavior on a granular level and then view

aggregated analytics is the basis of making

smart marketing and advertising decisions

across all acquisition and engagement

channels.

It is therefore important that you clearly define

your goals and pinpoint the in-app events that are directly

related to these goals.

Here’s a list of key in-app events in retail apps:

• Login • Wishlist

• Registration • Add to Cart

• Content Viewed (product/category screens)

• Initiated Checkout

• Search String • Purchase

ORDER NOW


6www.appsflyer.com


RevenueContent TypeContent IDQuantity Currency

Search StringPriceCurrencyContent IdQuantifyCustomer User ID

Purchase Content Viewed

Diving deeper into the data involves one more crucial step: defining

the right parameters of each in-app event - thus making it “rich”.

For example:

So how can rich in-app events lead to the ability to meet specific

business objectives? Well, if one of your goals is to increase the

number of consumers who bought at least two electrical products,

you’ll want to track the purchase event “enriched” with the quantity

and content type parameters. If you want more women shoppers as

active consumers, you can track the content viewed event in addition

to content type, content ID, quantity and / or search string parameters.


7www.appsflyer.com


1

2

Expand your knowledge about your users through deep

analytics

Set the path towards granular segmentation and enhanced

targeting with networks that are at the forefront of mobile

targeting capabilities (those that are able to create advanced

audience targeting campaigns after receiving rich in-app event

data in real time from you or your tracking provider)

Ultimately… by tracking these rich in-app events, you’ll be able to


8www.appsflyer.com


Smart Customer Acquisition

So we’ve laid out goals, defined the in-app events that matter to

your business and understood why this data is so valuable.

Now let’s take a step back and recall what advertising attribution

and analytics is all about from a user acquisition standpoint: The

idea is to continue to measure what users are doing inside the app

after they’ve installed it — particularly if these are rich in-app events.

The data is then aggregated to get an in-depth picture of your audience

and which channels (paid or owned), ad networks, campaigns and /

or creatives have led users you want - not just any users - to install

the app.

Measuring Paid Acquisition

Here’s a table that shows which networks did well and which did

not when aggregating rich in-app event data of users who added

women fashion items to the cart.

Networks 4 and 5 reign supreme, while network 2 should be

dropped.

Key takeaway:


9www.appsflyer.com


Enabling Lookalike Targeting

By measuring in-app events you can get a clear picture of the types

of consumers that generate the highest value to your brand. You

can then use this knowledge to find users that “look like them” from

a variety of sources by:

1) Increasing spend in a top performing network to maximize reach

of its quality audience (make sure you don’t go overboard and

end up targeting the same users)

2) Replicating the same targeting criteria used by the performing

network with other networks

3) Utilizing the capabilities of the most advanced networks that

support real time data transfer of rich in-app events from a tracking

provider and automatically turn it into a lookalike campaign.

An interesting insight comes from network 5: although it delivered

the lowest number of users, in terms of the install to in-app

action conversion rate, 8.2% is a great figure so an even bigger

investment in this network makes sense.

Another example has network 6 acquiring the largest volume of

users who logged-in and made at least two purchases within their

first two weeks of activity.

Key takeaway:


10www.appsflyer.com


Key takeaway:

Measuring Activities on Owned Channels

Leveraging your own assets on owned channels — particularly emails

and phone numbers of your existing segmented consumers who

have yet to install your app — is something you’ll definitely want to

measure, and compare to other channels — including paid ones.

The advertiser's email campaign is doing very well compared to

other channels with the highest conversion rate and the second

highest average revenue per user.


11www.appsflyer.com


Key takeaway:

Measuring the Consumer Journey with Multi-Touch Attribution

Another important layer to track in order to optimize user acquisition

campaigns runs higher up in the funnel, and it involves the ability

to go beyond last click attribution. Flawed as giving all the credit to

the network that delivered the last click is, it’s the industry standard.

Which is fine if we look at it from a pure billing perspective.

However, if a marketer is able to see his most common conversion

paths that led users to install an app, he will continue to invest in

these contributing networks. After all, these networks drove your

users down the funnel and made them “sales-ready.”

Network A view

Network A view

Network B click

Network B click

Network C click

Network Dlast click

Network Dlast click

1,500 installs

900 installs

Let’s explore the following scenarios:

Network C plays a key role in driving installs


12www.appsflyer.com


Measuring (True) Omni-Channel LTV

The proliferation of mobile devices, on top

of web and brick & mortar touchpoints

have created an amazing opportunity for

omni-channel retailers. However, it has also

generated a massive challenge of connecting

the dots to provide a seamless and consistent

user experience.

By using a server-to-server integration you

can measure the actions acquired app users

take across touchpoints (assuming they

are logged in to enable matching via their

consumer ID), giving you the true value

generated by the acquiring ad network.

U.S. Cross-Device Share of eCommerce Transactions

% of users who used multiple devices in path to purchase

41%

37%

43%

User completed purchase on:

Smartphone

Desktop

Tablet

Source: Criteo, State of Mobile Commerce, Q2 2015


13www.appsflyer.com


Let’s look at the following example to illustrate this point:

Disconnect

"I clicked on an ad that led me

to install an app after which I've made 3

purchases"

"Sadly, my LTV is only

$50

"When it's all

connected I'm worth $350!

Connected

Key takeaway:The value an ad network provides omni-channel retailers is often

much higher than they think. Best of all, you can track it to see it.


14www.appsflyer.com


Measuring the Impact of TV Ads

As app marketers drive user acquisition through DSPs, SSPs, Google

AdMob, Facebook, and a whole slew of mobile channels, they’re

realizing that mobile as a channel is becoming saturated, overly

competitive and even limited in reach. One channel they are turning

to for help is television.

The fact is TV is still very much a part of the fabric of our lives. According

to Nielsen’s 2014 Q4 Report, there were 285 million TV viewers in the US

so running a TV ad campaign guarantees access to a mass audience.

Digital video platforms have a long way to go to catch up.

The good news is that even while watching TV, a viewer’s smart

device is only an arm’s length away. In fact, Nielsen found that 84% of

US viewers are watching TV with a second screen in hand. Sixty-four

percent of viewers are using their smart device simultaneously. It’s a

great opportunity to capture their attention to download an app on

the spot or incentivize inactive users to take action.

View ad for 'Famous Footwear' mobile app


15www.appsflyer.com

http://www.nielsen.com/content/dam/corporate/us/en/reports-downloads/2015-reports/total-audience-report-q4-2014.pdf

http://www.nielsen.com/us/en/insights/news/2014/fast-friends-how-brands-are-capturing-the-attention-of-social-tv-audiences.html

comingnext.tv/blog/how-smartphone-penetration-drives-second-screen-usage/

comingnext.tv/blog/how-smartphone-penetration-drives-second-screen-usage/

http://techcrunch.com/2015/07/01/why-you-should-advertise-your-app-on-television/

https://www.youtube.com/watch?v=KGgrWyetTyc

https://www.youtube.com/watch?v=KGgrWyetTyc


How does it work? Basically TV syncing providers offer detailed

information on the exact time a TV ad was aired. A tracking provider

matches that data which is provided in real time with installs and

attributes credit to the TV network, assuming it occurred within a

predetermined time frame after the ad ran (usually within minutes

to a couple of hours at most).


16www.appsflyer.com


Measuring Shopping via Social Networks

A new form of online shopping that’s creating a lot of buzz as of late

has to do with ’Buy’ buttons, as social platforms — including Facebook,

Instagram, Pinterest, Twitter and Google — have all announced plans

to offer this service.

How does it work? A user enters his credit card number into the social

app, along with his shipping details. When he later sees something he

wants to buy, he can click on the “Buy” button, which then informs

integrated retailer partners where to ship it.

This form of one-click shopping has great potential and offers another

way in which retailers can acquire new customers — although probably

not new app users as it doesn’t involve any app install. However, the

merchant can always encourage them to install the app once they

are in his database.

In theory, when and if this functionality takes off, measurement is

rather straightforward. The social app can send an in-app event to

the analytics company (usually requires a special integration of its

own to select providers), which is actually a deeplink with metadata

on the actual purchase.


17www.appsflyer.com

http://www.nytimes.com/2015/07/06/technology/new-simple-buy-buttons-aim-to-entice-mobile-shoppers.html?_r=1


Relevant & Personalized Consumer Communication

The digital age consumer expects nothing short of a seamless

experience as she constantly moves back and forth across online

and offline brand touchpoints. Any glitch and you risk losing her to

a sea of competitors eager to take your place.

The good news is that this consumer also perceives behavior-

based communication as a service rather than an annoyance and is

comfortable with sharing information in exchange for personalization,

according to a newly-released and comprehensive mybuys report.

60%

60%

58%

58%

50%

I'm comfortable having my shopping interests and behaviors used

by retailers...

In order to expedite my shopping experience on site

So they can deliver relevant offers throughout the year

In return for a more personalized experience

As I believe I will receive a better overall experience

To coordinate a better shopping experience across the channels I use

The report also found that 52% of consumers spend more money

when they receive a truly personalized and consistent shopping

experience.


18www.appsflyer.com

http://www.mybuys.com/wp-content/uploads/7th-Annual-Personalization-eBook-MyBuys-and-the-e-tailing-group.pdf


Mobile, which is already playing a huge role across the consumer

journey (both in the research and purchase stages), is no exception.

Whether through email, coupons, push notifications or advertising,

personalized and relevant communication on mobile devices is

paramount to success.


19www.appsflyer.com

http://www.ninthdecimal.com/wp-content/uploads/2014/09/ND_Insights_Q2_2014.pdf

http://www.emarketer.com/Article/Smartphones-Closing-Retail-Ecommerce-Device-Gap/1012467


Retargeting Enablement, Attribution and Analytics

First things first: a must-have element in a retargeting campaign on

mobile is deeplinking as you’ll want to direct users to specific app

screens. It’s a no-brainer.

Second, there’s the issue of transferring data to a retargeting network

based on the level of integration you or your analytics provider have

with the network, as follows:

Data on standard in-app events can be transferred ‘manually’ in lists

(for example, a list of IDFAs of users who viewed products but didn’t

buy). Once a list is created, however, it cannot be changed.

Here we have two types:

1 Basic integration: Sending standard in-app events in real time to

enable segmented retargeting. For example, users who added a

product to cart but didn’t buy. In most cases, these integrations

can only support a name parameter (but at least a user in that

segment is ‘dynamic’ and will automatically be part of a different

segment once he purchases).

2 Advanced integration: Sending rich in-app events in real time to

enable personalized retargeting on a ‘segment of one’ (e.g. a

user with IDFA 1234 who viewed product 777 can be shown a

creative with product 777 and relevant messaging; you can even bid

higher for that user if, for instance, he has a high LTV). Currently, the

number of networks that can support such an advanced integration

is still limited but more and more networks are following suit.

Static:

Dynamic:


20www.appsflyer.com


When it comes to retargeting attribution, the mechanism is no different

than an install. It simply means an advertiser defines a desired in-

app event (e.g. app open for a dormant user, in-app purchase for an

existing user) for which the attribution provider would credit a media

source that delivered the last click that led to the desired action.

Retargeting attribution is often divided between “re-attributions”

and “re-engagements.” The former represents users who re-installed

the app after interacting with a retargeting campaign (assuming the

IDFA/GAID remains the same), while the latter describes users who

have the app installed and engaged with a retargeting campaign.

Once tracked, aggregated data of all users will help you pinpoint the

best performing retargeting networks, campaigns and creatives in

the same way that was illustrated in the acquisition campaigns above.


21www.appsflyer.com


Enabling & Measuring Facebook Custom Audiences Campaigns

Back in May this year, Facebook announced that as of November 4, its

advertising clients will no longer be able to access device-level data

from their Facebook mobile app install campaigns.

The good news is that the social network’s Mobile Measurement

Partners will still have access to this data, which means you’ll still be

able to use it in your Custom Audiences campaigns. This is done by

configuring the campaign in Facebook which will run automatically

when it receives rich in-app event data from the measurement partner.

Measuring Location-Driven Retargeting Campaigns

According to a Google-Ipsos report, 4 in 5 consumers want ads

customized to their city, zip code or immediate surroundings. The

report also found that 50% of consumers who conducted a local

search on their smartphone visited a store within a day while 18%

actually bought a product. That’s why targeting consumers while

on-the-go has tons of value.

Tracking location is possible

by Lat/Long parameter. The

advertiser can add these

params to each in-app

event which can be sent

to the networks that support

rich in-app events. You’ll then

be able to understand where

consumers take specific types

of actions.


22www.appsflyer.com

https://facebookmarketingpartners.com/marketing-partners/#ad=&co=&in=&la=&qss=&sp%5B%5D=Measurement

https://facebookmarketingpartners.com/marketing-partners/#ad=&co=&in=&la=&qss=&sp%5B%5D=Measurement

https://think.storage.googleapis.com/docs/how-advertisers-can-extend-their-relevance-with-search_research-studies.pdf


Measuring Bluetooth-Based Beacon Technology

With beacon technology, retailers can pinpoint a user’s location

and send relevant content based on where consumers are. Since

bluetooth connections actually work indoors unlike GPS, it captures

an audience that’s highly likely to engage with the messaging while

in the immediate vicinity of a store or, unlike GPS, in store.

Sending a push notification is ideal if the user has the app installed.

If not, an SMS will also work well. Sending an ad while in-store can

work by offering a user an immediate discount for installing the app

(requires sending of real time attribution data to enable opening of

the specific campaign app screen after deferred deeplinking).

Measuring a beacon-driven campaign - whether using Apple’s iBeacon

or Google’s brand new platform-agnostic Eddystone platform and

recording events in your server - involves defining in-app events and

reporting them to your analytics provider.

You can then use cohort reports t o

measure which group of users from

which campaign had the highest

engagement with beacon messages.


23www.appsflyer.com


Measuring Direct Marketing Efforts

Direct communication with consumers mainly over owned channels

is the key to a strong consumer-brand relationship. All you have to

do to measure these campaigns is set up a tracking link and then

compare your different channels — paid or owned.

Email Campaigns: According to research by Forrester, 42%

of retailers' email messages were opened by consumers on

their smartphones and 17% were opened on tablets. So you’ll

definitely want to measure email engagement and more

importantly what follows. With proper deeplinking in place,

you can direct users to specific app screens from the mobile

web or email app.

Push Notifications: Push has gone mainstream. That’s the key

takeaway from the following highlights of a study by Urban

Airship:

• Retailers sent 34% more push notifications in 2014 compared

to 2013

• Consumers’ engagement rate with push notifications in the

2014 holiday season doubled compared to 2013

• More than one-third of all retail app opens were the result

of a push notification

Coupons: The number of adults who redeem coupons via a

mobile device is expected to jump from 78.7 million to 104.1

million between 2014 and 2016, according to eMarketer. With

such volume, it is important that retailers offer the best overall

experience to optimize usage.


24www.appsflyer.com

https://www.forrester.com/Search+Marketing+Tops+Online+Retail+Customer+Acquisition+Tactics+According+to+Shoporg+Forrester+Research+Study/-/E-PRE7204

http://marketingland.com/view-mobile-retail-holidays-past-present-112419

http://marketingland.com/view-mobile-retail-holidays-past-present-112419

http://www.emarketer.com/Article/Shoppers-Turn-Smartphones-Tablets-Cash-Coupons/1011582


Creating a Unified User Experience for Optimal Results -

Ideal for Direct Messaging

Overcoming the challenge of a fragmented mobile space with multiple

channels to create this elusive seamless communication with prospects

and existing consumers involves five components:

1) Smart Links: Advanced technology can make your life much

easier by allowing you to configure a single tracking link per

campaign for all sources. This smart link automatically detects

where the user is and where he needs to be directed as he

engages with the campaign. So instead of configuring at least

five tracking links for iOS, Android, Windows, Amazon and

web, all you need is one. That means there’s a lot less room

for error - be it broken links or links that end up taking your

users to the wrong place.

2) Deeplinking: Connecting the web-to-app and app-to-app

environments, while supporting the opening of a specific app

screen.

3) Deferred deeplinking: The ability to identify whether the

user has the app installed on his device or not and if not, to

take that user to the app store where he can download the app

4) Real-time attribution data: Enables the opening of a specific

landing page after deferred deeplinking and upon first app

launch.

5) Utilizing customer user ID to re-engage with the consumer

in way that reflects her specific actions across the different

channels.


25www.appsflyer.com


Let’s explore two examples:

1) Email offer to a consumer who has the app installed:

NEWCOLLECTION!

2) SMS to an existing consumer who does not have retailer’s

app installed:

Email promotion for new iPhone on sale

SMS with deal on new collection

Redirect to the right app store

Specific landing page of new collection deals

SmartphoneGmail app iPhone product

page in smartphone app

iPhone product page in tablet app

iPhone product page in desktop app

Tablet web

Desktop web

Download App Launch

Single Smart Link


26www.appsflyer.com


How to Keep Consumers Coming Back for More

Retaining consumers to foster a lasting consumer-brand relationship

and ultimately maximize their lifetime value is perhaps the most

important goal for a retailer. The problem is that it’s also the most

challenging… So what can you do to get retention right?

There are two main strategies here: first, product-related. It goes

without saying that your product must provide a top notch user

experience. Any unnecessary friction and you risk losing the valuable

consumers you worked so hard to acquire.

Second, advertising-related, which is what we’ll focus on. Specifically

we’re talking about retention and cohort analysis.


27www.appsflyer.com


Retention Analysis

The more straightforward analysis involves the retention report. It

is calculated as the unique number of users who were active on a

specific day/week out of the total number of unique users who first

launched the app in the selected timeframe.

Depending on how deep the granularity offered by the analytics

provider is, you can check the retention of users from a media source,

campaign, ad group, country, city, OS version, device brand, device

model etc.

Let’s explore the following case study of a leading retailer, broken

down by campaign:

Key takeaway:• Campaigns 1 and 4 performed well across the board, retaining

12-14% of users on day 12.

• Campaigns 2 and 3 underperformed from day 1, retaining only

5-7% of users on day 12.


28www.appsflyer.com


Optimization performed on campaigns 2 and 3:

• Budgets reduced

• Targeting & messaging modified to better reflect campaign 4

• Low performing ad sets within the campaigns were shut down

The data from a couple of weeks later clearly shows the advertiser’s

efforts paid off:


29www.appsflyer.com


Cohort Analysis

Retention only tells us how many users who launched the app during

a set timeframe reopened the app in the days that followed. Cohort

goes much deeper. Let’s take a step back to explore this strategy.

A cohort report enables you to group users with common characteristics

and measure specific KPIs over different timeframes. It offers a

really good indication of the quality of the average customer and

whether it’s increasing or decreasing over time. The metric - whether

sessions, revenue, or any other defined in-app event - is calculated

per different timeframes, which represent the first X activity days per

user, and then accumulated among all users (that’s why the graph

never drops).

Here’s an example of how a cohort analysis can

boost retention and lifetime value by increasing

the average number of sessions per user. In this

case, the cohort was grouped by the media

source that acquired the user (other examples

include campaign, publisher,

ad group, geo etc.)


30www.appsflyer.com


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

A

B

C

D

Cohort graph as seen on May 1:

Average Sessions Per User

Days

Key takeaways:

• Network A starts off with a bang, producing high quality users

that are highly active in the first 14 days after which the growth

rate begins to plateau

• Network B is also delivering quality users showing a modest yet

steady climb throughout the period

• Network C is slowly growing until day 7 but then flattens

• Network D performed poorly - not only did it acquire low quality

users that barely engaged with the app, it also did not show

any improvement over time.


31www.appsflyer.com


Optimization performed:

• Having noticed the drop on day 14, the retailer began retargeting

its users at that time to encourage them to continue engaging

with the app

• The budget of Network B was increased hoping it would generate

more users who would only increase their engagement over time

• Network C was broken down to the campaign and ad level and

those that underperformed were removed

• Network D was removed

Cohort graph as seen on June 1:

Key takeaways:

• Network A retargeting worked well as consumers continued

engaging with the app at a growing rate beyond day 14

• Network B continued to deliver quality users (the increase is not

seen in this graph as it shows the average sessions per user, but

it can easily be viewed in the report that measures quantity)

• Network C optimization worked well as the engagement growth

rate improved compared to the previous month

A

B

C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Days


32www.appsflyer.com


As more and more retailers are embracing mobile as a central

if not the most important component of their strategy, they

are moving beyond mobile-optimized sites to develop native

apps. Despite the fragmentation and the challenges it creates, the

mobile world is rapidly evolving (and that’s an understatement),

catching up with the web when it comes to measurement, analytics

and ad optimization capabilities. Retailers who are ready to take on

this brave new world will reap the rewards.

To Sum Up


33www.appsflyer.com

mobile attribution & marketing analytics - appsflyer · mobile attribution & marketing...

Documents