Gaël has developed an app named GreenPower. After two years work and a lot of effort, the app has now been downloaded over 1.2 million times from the Google Play Store.
During this time Android development went from being a hobby activity with a few daily downloads, to quitting his regular job and start working full time on the project.
Gaël has learned a lot of do's and don'ts about how to make an app successful and climb up the charts. In this presentation, he will provide a humble feedback on this journey, and try to bring advices that may be useful for other developers.
2. ”Ordinary things done consistently produce
extraordinary results”
Keith Cunningham
23.01.2013 - GDG Oslo - 2/35
3. Why this presentation?
Share tips of lessons learned, challenges
faced
Exchange information, both ways
23.01.2013 - GDG Oslo - 3/35
4. Background
About me
Software engineer
16 years in various large companies (Dev, PL)
August 2012: Resigns from Canal Digital
Sept 2012: Goes indie, starts AS
Jan 2013: Starts at StartupLab
23.01.2013 - GDG Oslo - 4/35
5. Background
History of GreenPower app
2010: My first HTC hero
March 2010: First Free version published
October 2010: First Paid version published
Jan 2013:
1.3M downloads Free (>2500/day)
200.000 active users
23.01.2013 - GDG Oslo - 5/35
8. Freemium: Free app
Ads
Don’t overuse (no intersticials, push notifications etc) or
negative feedback
Acts as reminder, incentive to go Premium. No big income.
Avoid 3rd party ads SDK (time to integrate, allways
crashes somewhere). Use mediation instead.
Limited features
Put enough to make people happy
Subtile cursor in what to make premium (ask / poll users)
Focus on the Free version, conversion later
23.01.2013 - GDG Oslo - 8/35
9. Freemium: Paid app
Paid version
All features, no ads
How to set the price
Look at competitors price range
Don’t get greedy too soon
Start low, gradually increase, see impact on revenues
A higher price can decrease downloads, but higher revenus
Fewer paid downloads no big deal if Free downloads stay
high to stay high in rankings
But! A higher price doesnt mean fewer downloads! ☺
23.01.2013 - GDG Oslo - 9/35
10. Freemium: Conversion rate
Indonesia
Phillipines
Chile Free / Paid ratio
Thailand
Argentina
Turkey Average 5%
Ukraine
Brazil
Romania
Huge variations per country
Portugal
United States
Greece
Austria
South Africa
Netherlands
Czech Republic
Sweden
Canada
Germany
United…
Finland
Switzerland
Norway
0% 2% 4% 6% 8% 10 % 12 % 14 % 16 % 18 %
Ratio
23.01.2013 - GDG Oslo - 10/35
11. Freemium: Alternatives
In app payments
Single version
Free
Pay per feature
Pros
Higher chance to sell individual features at cheaper
price
Cons
Several apps (Free & Premium) increase visibility on
Market and search engines
23.01.2013 - GDG Oslo - 11/35
12. ”The important is not what you do but how you
do it”
Keith Cunnigham
23.01.2013 - GDG Oslo - 12/35
13. Ratings
Higher ratings mean more downloads
More visible stars, user more prone to download
Higher in search results
Thresholds
3.75/5
4.25/5
Going from 3.74 to 3.75
Major increase in downloads
Now 4.22 aiming at 4.25
300 more 5 stars needed ☺ (10.000 ratings)
23.01.2013 - GDG Oslo - 13/35
14. Ratings
Ask users for ratings
Try to ask only happy users
Message only on upgraded
versions
Popup after a few days use
Don’t ask too much
Avoid 1 star ratings
Answer comments (when
possible)
Try to identify user and answer
23.01.2013 - GDG Oslo - 14/35
15. Quality matters: Software
Avoid crashes and freezes
Hard: Hundreds of devices / OS combinations
Cautious with 3rd party SDK
Try/catch all, use asynchronous calls
Fix bugs quickly
Testing is vital, but tricky
Buy a Nexus 10 for testing all screen sizes
Automate testing (monkeyrunner etc)
Beta testers among users
Testing companies
Crowdtesting (bugfinders etc)
Testing on remote physical hardware
23.01.2013 - GDG Oslo - 15/35
16. Quality Matters: App Description
Make attracting visuals
Sexy logo, screenshots and banner mean more
downloads
20%+ when improving GUI
Video
Textual description
Focus on benefits rather than features
Gently stuff it with right keywords
AdWords, AppDevHQ: Find out relevant keywords
23.01.2013 - GDG Oslo - 16/35
17. Quality matters: Support
Provide best support
Respond promptly to all issues
Be nice and supportive, it can pay
An initially unhappy user can turn out to promote
your ap (ex Sprint tech support)
Accept all refund requests
Support tools? (Zendesk etc)
Delegate support?
23.01.2013 - GDG Oslo - 17/35
20. Promoting
Don’t focus too much on making the product
Marketing and promotion is as important, if
not more
23.01.2013 - GDG Oslo - 20/35
21. Promoting: Ads
Buy ads AdWords
Rather expensive ads
Understand what you Very feature rich. High
learning curve
pay for: CPC vs CPI
AdMob
High variability in Cumbersome approval
prices process
Ok prices
Promotions AppBrain
(AndroidPit, Amazon Basic but quite low CPI
etc) Facebook
Target precisely where ads
Advertising: Time appear
consuming. Delegate? Cheap
23.01.2013 - GDG Oslo - 21/35
22. Promoting: ASO & SEO
75% of app downloads comes from search
Source: www.fiksu.com
App store optimization (ASO)
Go up in Google Play search results
Ensure high ratings, high ranking
Right keywords in desciption
Search Engine Optimization (SEO)
Google Play search uses regular web search as well
Improve ranking of your Google Play page in Google’s
search
Forums
PR companies
23.01.2013 - GDG Oslo - 22/35
24. Promoting: Share
Share from the app
Share button
Google+, Facebook
, email...
Buzzdoes
Make your app viral
Reward users for
sharing your app
23.01.2013 - GDG Oslo - 24/35
26. Localize: Why?
Localize the app
USA: From 50% to 12% of
downloads
Make people happy
Cons: Increase apk
size, increase complexity
Localize on Google Play
App description
Localize screenshots: +30%
downloads
Banner
23.01.2013 - GDG Oslo - 26/35
27. Localize: How to translate
Translators
Good-will users: Ok at start. Cumbersome when
many languages.
Translation company (Universally Speaking LTD).
Single point of contact.
Translation platforms: Crowdin etc
23.01.2013 - GDG Oslo - 27/35
28. Localize: Pick languages
Chose countries
Top languages in Google Play per category
Big variations in Free vs Premium
23.01.2013 - GDG Oslo - 28/35
30. Publish
Google Play
Obvious. No pain publishing.
Nice new console
A few shortcoming: Would need API for publishing
updates description (pain if many languages)
Consider alternative markets
Increase visibility and downloads even on Google
Play
Reach other countries
Cons: Several builds
23.01.2013 - GDG Oslo - 30/35
31. Publish: Alternative markets
AndroidPit (Free & Paid)
Great to reach visibility in
Germany (#1 country for
GreenPower now). Several
promotion possibilites.
Jesta Digital GMBH (Paid only)
20% revenues of GooglePlay
Samsung apps (Free & Paid)
10% volume of GooglePlay
Weird and annoying console
Amazon (Paid only)
Only a few dozens sales
monthly. But increases
visibility. Free app of the day.
23.01.2013 - GDG Oslo - 31/35
32. Publish: Alternative markets
GetJar (Free only)
Several dozens Free
downloads daily. Worth it.
Korea
Tstore and Olleh: Carriers
specific markets. Several
hundred Free downloads.
China
No reliable alternative
market for paid apps
Other
Opera, Mobiroo, Appland...
23.01.2013 - GDG Oslo - 32/35
33. Publish: Code tips
One Android library project
One Android project per Market and flavor
(Free/Paid)
Single build script
Don’t put flags in properties
23.01.2013 - GDG Oslo - 33/35
34. Protect your app
When it gets popular it gets pirated
A few hours after publishing on Google Play, pirated and available on
download websites
Android license check
Useless if used alone
Proguard
Obfuscate code, doesnt prevent piracy
Dexguard
The winning tool: Code obfuscation, tamper detection, class encryption
etc.
No more piracy
DMCA Guardian
USA’s Digital Millenium Copyright Act (DMCA) law
Automated web search and takedown service
23.01.2013 - GDG Oslo - 34/35
35. Thank you
Questions?
Feedback & suggestions: gael.pouzerate@gmail.com
Liked it? !
23.01.2013 - GDG Oslo - 35/35