Proprietary + Confidential
The App-ortunity
Optimizing customer value
across Web and App
Lea Wehbe
App Specialist
Year on Year
growth in time
spent per mobile
app category
in 2017
Source: Statista 2017
82
Smartphone users
consult their phone
on purchases they
% are about to make
in–store
Source: Google Internal Data - Global
The Romanian Landscape
Retail sales in Romania see a slow growth, but Retail apps download and Revenue
are growing YoY.
Source: business-review.eu, 2019. romania-insider.com, 2018. App Annie Market Size Report, 2018-2019.
The “App-ortunity”
For retailers who actively Worldwide YoY increase of Transactions
for retailers (app vs. mweb)
promote their shopping
apps mobile transactions
represent Mobile Web = 31%
65%
of all transactions YoY increase
Apps = 38%
Source: Criteo 2018
EMEA: 26% of retail transactions are in apps
26%
App
mweb 33%
Desktop 41%
Source: CRITEO | Q1 -2018 Europe Retail Report
Shopping apps generate higher conversion rates...
4%
Mobile Web
3x 13%
App
Source: CRITEO | Q1 -2018 Europe Retail Report
...With the highest average order value...
$92 $100 $102
Mobile Web Desktop App
Source: CRITEO | Q1 -2018 Europe Retail Report
…And the lowest cart abandonment rate
97% 68% 20%
Mobile Web Desktop Mobile App
Source: Forbes 2018 Global Data
Proprietary + Confidential
REMOVE FRICTION
WITH A GREAT USER EXPERIENCE
A quick poll...
Who is happy about their
web <> app experience?
Why optimize across web and app destinations?
Users Marketers Google
Users expect Customer-centric Optimize value
frictionless mobile approach across across web and app
experiences across web+app drives in one campaign
web & app growth & retention,
but is complex.
What if we use apps to help you
value mobile ads?
What if we do that, by using a
simple solution?
DEEP-LINKING
One link to site and app
This means more people opening the app from
organic searches...
...as well as from paid media
Look how great the user experience is
And no changes were made
to their current Shopping
campaign!
Bringing the power of Google Analytics to App
campaigns
Google
Analytics
Provide signals for for Firebase
Turn user insights into
Google Ads campaign Understand how users
marketing actions
performance (“feed the behave in your app
machine”)
What deep linked products do we offer?
Automatically brings users back to the app if it has been
App Deep Linking installed using search / display network ads. Not directly
influenced beyond setting up deep links.
Bring users, who look at specific products, back to specific in app
Dynamic RMKT landing pages, using feed based ads with specific products on
mGDN/AdMob.
ACe
One campaign, fully automated to remarket users with static ads
across all Google Networks
What deep linked products do we offer?
Automatically brings users back to the app if it has been
App Deep Linking installed using search / display network ads. Not directly
influenced beyond setting up deep links.
Bring users, who look at specific products, back to specific in app
Dynamic RMKT landing pages, using feed based ads with specific products on
mGDN/AdMob.
ACe
One campaign, fully automated to remarket users with static ads
across all Google Networks
Deeplink your mSearch & mShopping campaigns
for higher conversion rates
User has app User does not have app
Improved Conversion Rate
How to fix this leak, and capture full web+app
value?
3 key steps to optimize your web campaigns
+ +
Confirm deep links work, Set up app Optimize for both web
and are enabled conversion tracking and app conversions
App Links Universal Links with GA for
Firebase SDK
Understanding the true value of your campaigns
with deeplinking
Web campaign Total Spend Conversions Conversion Rate Cost per Conversion
Campaign 1 $1000 100 10% $10
Web campaigns with Total Spend Conversions Conversion Rate Cost per Conversion
deep links
Web App
Campaign 1 $1000 100 80 18% $5.56
Campaign 2 $1000 100 50 15% $6.67
Campaign 3 $1000 100 5 10.5% $9.52
Campaign 4 $1000 100 200 30% $3.33
If you successfully set up your campaign,
you will be able to claim this missing conversion credit
By leveraging Google’s app deep linking,
OTTO sees more than ¼ of overall revenue from app purchases
Business Objectives Results
• Shift to mobile first strategy • More than ¼ of overall revenue from app
• Deliver a seamless and personal mobile • 86% YoY sales growth on app compared to
experience to mobile customers mWeb
• Grow customer retention • 2.3x higher conversion rate with mApp
What deep linked products do we offer?
Automatically brings users back to the app if it has been
App Deep Linking installed using search / display network ads. Not directly
influenced beyond setting up deep links.
Bring users, who look at specific products, back to specific in app
Dynamic RMKT landing pages, using feed based ads with specific products on
mGDN/AdMob.
ACe
One campaign, fully automated to remarket users with static ads
across all Google Networks
App Dynamic Remarketing
Engage existing app users with relevant ads about the products or items they engaged with most recently in your app via
the mGDN and App Network. Like regular/web Dynamic RMKT, this is a feed based solution.
1 2
After seeing User clicks
a product on on the ad and
Lazada's app, completes
user navigates the purchase
on the web through the app
Trendyol improves ROI by 70% with Dynamic RMKT for Apps
Business Objectives Results
• Re-engage high-value users more effectively and at • 3X growth in app traffic from Google
scale
• 140% increase in click-through rate
• Increase volume of installs to the Trendyol app
• 20% growth in conversion rate
• Improve click-through and conversion rate
• 70% increase in ROI
What deep linked products do we offer?
Automatically brings users back to the app if it has been
App Deep Linking installed using search / display network ads. Not directly
influenced beyond setting up deep links.
Bring users, who look at specific products, back to specific in app
Dynamic RMKT landing pages, using feed based ads with specific products on
mGDN/AdMob.
ACe
One campaign, fully automated to remarket users with static ads
across all Google Networks
ACe targets users who already have the app to
come back and engage
When compared to other re -engagement networks,
Omio sees a 25-30% better ROI with Google’s App Campaigns for Engagement
Business Objectives Results
• Increase ROI from existing users • 25-30% better ROI when compared to other
• Expand reach for remarketing campaign re-engagement networks
• Diversify from other remarketing channels • 5x increase in conversion rate
MAKE EVERY TOUCHPOINT SHOPPEABLE
TO MAXIMIZE SALES
Boost your omnichannel experience with Apps
App & mWeb: Rich vs Reach
Awareness
Engage new customers
mWeb
with delightful mWeb
Consideration
experiences while
leveraging on App to
Intent
build loyalty among
App
high value customers
Loyalty
In order to be successful in the apps space,
you need...
1. Users to download your app.
2. Users to use your app.
APP EXPERIENCES
THAT STAND OUT
Shaping tomorrow’s omnichannel shoppers
experience
IKEA Place
3D live experience in the app Zalando’s image recognition Ikea’s augmented reality app helps
to drive in-store purchases app makes shopping easier customers see which furniture fits
best in their home
Retail Apps
Seasonal Peaks
The Holidays are a critical time to drive
downloads in EMEA
Black Friday & Cyber Monday Post-Christmas
300K app downloads app active users
peak app install period
200K
100K
200M
+27% +23%
Oct 2018 Nov 2018 Dec 2018 Jan 2017
Source: App Annie, Oct 2018-Jan 2019, Top 24 EMEA countries
Source: Analys, Flurry, BLS, AppAnnie
During the Holiday Season...
2 in 5 45 %
Mobile shoppers are Time spent in shopping
downloading a new apps during Black Friday
shopping/retail app Week
Source: Google and Verto Analytics Apps Study
In Romania last year, 22k phones were sold on
Black Friday
“Record sales in the mobile phones
category were registered for the Apple
iPhone X, 64GB, 4G, Space Grey - 500
such phones sold in 43 seconds, and for
Samsung Galaxy S8, 64GB – 1,000
phones in 64 seconds.”
Source: Romania-Insider.com, 2018
App Campaigns could be part of your paid
media orchestration for Seasonal Peaks
Competitive bids Set higher bids than your regular campaigns and re-adjust bids accordingly
1. Campaigns should be created (not running) at least 4 day before.
Burst Campaigns 2. Budgets Uncapped
3. Set your tCPI/tCPA at least 20%-30% or higher
4. Alternatively, use max conversion beta for burst effect
1. Leverage ad groups beta for your different creative themes
Ad groups &
2. Leverage your Google Merchant Center to empower machine learning
Feeds with your in-app feeds
Re-engage your user base with deep linking your Search campaigns, App
Re-engage DR and/or ACe
User Acquisition: Beta products (next gen)
Max conversions Feeds for App Campaigns
Alpha Alpha
Beta Beta
Released Released
Optimizes for spending your budget in Leverage your rich in-app content to
an allotted time for seasonal/holiday deliver better performance by surfacing
pushes relevant ads to the right user
Does not guarantee a CPI Does not impact YT inventory yet.
Timing is key!
Date Week of 11.11 Week of 18.11 Week of 25.11 BLACK FRIDAY: 29.11 CYBER MONDAY: 02.12
Same AC Create BF ad group in existing campaign, adjust bids and budgets as you go to capture all traffic
Burst Create new AC with BF creatives and a noticeably higher bid than evergreen ACs
Action
Plan Deep-link your mSearch campaigns (ongoing) Re-engage your user base with App DR/ACe+feeds to increase purchase volumes
Re-engage
Branding
Increase bids and Increase bids and
Bidding strategy Uncap bids and budgets
budgets by 20% budgets by 30%
Period Non-sale SALE 40% SALE 70% SALE 80%
Hepsiburada reaches 45X more shoppers and surges to the #1 spot on
iOS and Android during Black Friday using Universal App Campaigns
Business Objectives Implementation Strategy Winning Metrics
• Take advantage of high • Launch UAC campaign during • 45x growth in app user base
commerce period around Black specific 5 day period to: during Black Friday
Friday to drive more installs of ○ Reach new app users • 52K growth in new users overall
the Hepsiburada app
○Increase retention and • No.1 spot on iOS and Android (up
• Find new app customers likely to conversion volume from 200th position)
make a purchase in the
• Most successful UAC campaign in
Hepsiburada app
Turkey relative to competitors
Year on Year
growth in time
spent per mobile
app category
in 2017
Source: Statista 2017
Google App Products Mapping for Retail
Retail
● Installs / Actions: 20/80 split
Core products ● Appify
● App Dyn Rem
● Add to cart, purchase, high-value purchase, pBuyer
In-app events
● Separate campaigns for priority categories w/ higher bids
Setup specifics ● Conv window for purchases = 14 days
● Feeds, Ad Groups, SDK features incl. deferred deep links
Betas ● Max Conv for Flash sales
● Goal: Purchase
ACe ● Audience: cart abandoners, item viewers, lapsed buyers
App Campaigns leverage user signals and Machine Learning to
find the right customers at the right time
1
User Signal 2
Machine Learning 3
Distribution to
Data Algorithm high value users
Action
Install, In-app events,
Google Play Data
Context
User attributes, e.g. device,
time of day, location, search
queries, apps installed
Your KPIs
Intent
Search, Display,
Play Store, YouTube
Confused why Machine Learning matters? Check this resource on Machine Learning.
User Acquisition: bid according to your objective
UAC for Installs UAC for Installs Advanced UAC for Actions
tCPI tCPI tCPA
Easy and quick setup Easy and quick setup Needs less data points than tROAS
Fastest learning phase Fast learning phase Less fluctuations than tROAS
Maximum scale under tCPI High scale at tCPI Faster learning phase than tROAS
Takes into account post install events
Longer learning phase
Needs more data than UAC Installs/Advanced
tCPI is the only level Doesn't capture high IAP value variances vs
No post install considerations
No possibility to pilot post install performance tROAS
