GIAF 3 OCTOBER 2013
A/B TESTS

andy.toben@productmadness.com
What is an A/B test ?
“A/B
testing
is
a
methodology in advertising of
using
randomized
experiments with two variants,
A and B, which are the control
and treatment in the controlled
experiment.”
- wikipedia.org
What Can We A/B Test?
Basically anything:
• Page Colour
• Layout
• Call to action
• Images
• If you can change it, you can test it!
Step by Step Optimization
Testing one variable at the time works, but
can create issues:

- Time consuming:
Needs a lot of traffic or data
- Local verses Global Optimum:
Can lead to accepting the best variant for
that test, but it is not optimised against all
possible variants
Multi Variant Testing
This allows you to test all combinations at once however:
- Requires massive data sets – much more than Step by Step

- Requires mathematical tools
- No intuitive / building insights
How do we know who’s winning ?
How do we know who’s winning ?
Advertising is a game, so how about games ?
Strategies
What can we use A/B testing to discover?
•
•
•
•

Search - Option A / Option B
Optimize - Option A / Option A+ / Option AChange management - treatment verses control
Measure - No treatment verses treatment
A/B Testing User Experience
When A/B testing user experience you can
test out the following and create a big
impact… for better or worse!
- Order of game levels / features
- Bonus system
- Available content
- P2P assignment
- Pricing and Economy
Challenges We Face
Some A/B tests are very expensive
- You may need to commit development or art resource for
something that may not work.
Variant assignment needs to persistent
- In some cases you can’t just change a player’s track to the
“winning” branch
What to measure:
• Responsiveness (CTR)
• Engagement
• Retention
• Monetization
Branching user experience
Understanding Results
Which metric is the best one to look at, and which result should we take
action on?

Feature cost

1st day retention

7th day retention User value

10,000 coins

37%

10%

10 USD

25,000 coins

36%

6%

12 USD
But after all, It’s just a tool
Is there an
active
Experiment?

No

No
Change
No

Yes

User

Should we test
the User?
Allocate Variant

Yes
Questions ?

andy.toben@productmadness.com

Product Madness - A/B Testing

  • 1.
    GIAF 3 OCTOBER2013 A/B TESTS andy.toben@productmadness.com
  • 2.
    What is anA/B test ? “A/B testing is a methodology in advertising of using randomized experiments with two variants, A and B, which are the control and treatment in the controlled experiment.” - wikipedia.org
  • 3.
    What Can WeA/B Test? Basically anything: • Page Colour • Layout • Call to action • Images • If you can change it, you can test it!
  • 4.
    Step by StepOptimization Testing one variable at the time works, but can create issues: - Time consuming: Needs a lot of traffic or data - Local verses Global Optimum: Can lead to accepting the best variant for that test, but it is not optimised against all possible variants
  • 5.
    Multi Variant Testing Thisallows you to test all combinations at once however: - Requires massive data sets – much more than Step by Step - Requires mathematical tools - No intuitive / building insights
  • 6.
    How do weknow who’s winning ?
  • 7.
    How do weknow who’s winning ?
  • 8.
    Advertising is agame, so how about games ?
  • 9.
    Strategies What can weuse A/B testing to discover? • • • • Search - Option A / Option B Optimize - Option A / Option A+ / Option AChange management - treatment verses control Measure - No treatment verses treatment
  • 10.
    A/B Testing UserExperience When A/B testing user experience you can test out the following and create a big impact… for better or worse! - Order of game levels / features - Bonus system - Available content - P2P assignment - Pricing and Economy
  • 11.
    Challenges We Face SomeA/B tests are very expensive - You may need to commit development or art resource for something that may not work. Variant assignment needs to persistent - In some cases you can’t just change a player’s track to the “winning” branch What to measure: • Responsiveness (CTR) • Engagement • Retention • Monetization
  • 12.
  • 13.
    Understanding Results Which metricis the best one to look at, and which result should we take action on? Feature cost 1st day retention 7th day retention User value 10,000 coins 37% 10% 10 USD 25,000 coins 36% 6% 12 USD
  • 14.
    But after all,It’s just a tool Is there an active Experiment? No No Change No Yes User Should we test the User? Allocate Variant Yes
  • 15.