[#500distro] measuring for impact: knowing when, what & how to a/b test
Embed Size (px)
TRANSCRIPT
![Page 1: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/1.jpg)
@mike_greenfield
Measuring For Impact: Knowing What, and How to A/B Test
@mike_greenfieldCEO/Co-Founder, Laserlike
2014-08-07
![Page 2: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/2.jpg)
@mike_greenfield
You know you should A/B test.
![Page 3: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/3.jpg)
@mike_greenfield
You also know you should exercise more eat less sugar spend less on coffee wear sunscreen etc., etc.
![Page 4: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/4.jpg)
@mike_greenfield
(Don’t worry, I’m not going to say anything else
about sugar or sunscreen.)
![Page 5: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/5.jpg)
@mike_greenfield
So, how do you create a culture in which people will constructively A/B
test?
Do six things.
![Page 6: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/6.jpg)
@mike_greenfield
1. Embrace “I don’t know”
We have 2+ ideas.
I don’t know which one will be more effective.
![Page 7: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/7.jpg)
@mike_greenfield
![Page 8: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/8.jpg)
@mike_greenfield
2. Have Data, Choose Metrics
To test, you need:• People using your product• (Approximate) agreement on the
metrics that matter
![Page 9: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/9.jpg)
@mike_greenfield
Not Many Users? Don’t A/B test!
• Laserlike, has ~60 users and has never run an A/B test
• We will run many, many tests when we have enough users
• A test should have at least a few hundred instances (and a lot more if effect sizes are likely to be small)
• Test iff you can have “business significance”
![Page 10: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/10.jpg)
@mike_greenfield
Know What You Want to Optimize
• If it’s important, you should be running tests to improve it
• If it’s not important, spend time on other things
• Most tests should be aimed at improving 1-2 specific variables
![Page 11: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/11.jpg)
@mike_greenfield
3. Have Clear Process, Tech for Testing
![Page 12: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/12.jpg)
@mike_greenfield
A/B Testing Process• New feature: if possible, roll out to a
small test subset first (10s or 100s of thousands)
• Version change: always test things that could (cumulatively) have business impact
• Everyone on the product team should be running and resolving tests
![Page 13: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/13.jpg)
@mike_greenfield
A/B Testing Tech• Using a third party testing service is
akin to building your site on Wordpress: great at some scales/competency levels
• No matter how you’re testing, a new test should be at most a few lines of code
• It should be easy to see how each side of a test compares across many variables
![Page 14: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/14.jpg)
@mike_greenfield
4. Understand the Math of What to Test
![Page 15: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/15.jpg)
@mike_greenfield
Process: Same vs. New Tweak
• What’s the probability your tweak will have a positive effect?
• What kind of effect might that have, and how might that effect change the company’s prospects?
• Will you be able to measure the change?
• Optimize on one variable, but look at others
![Page 16: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/16.jpg)
@mike_greenfield
Process: Same vs. Big Change
• What’s the probability that your change will have a negative impact?
• How big an impact might there be?• Will you be able to measure the
change?• Holistic approach
![Page 17: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/17.jpg)
@mike_greenfield
A/B Test for Quality
• Circle of Moms: test “warning” users when questions seemed short, low quality
• Resulting questions were graded for quality, without grader knowing test bucket
• End result: warning yielded ~5% fewer questions, but much higher quality
![Page 18: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/18.jpg)
@mike_greenfield
5. Understand the Math of Picking Winners
![Page 19: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/19.jpg)
@mike_greenfield
Resolving Too Soon vs. Resolving Too Late
• How big is the potential audience for this test?
• Example 1: end of year “most popular baby names” email that will never be sent again
• Example 2: Facebook signup flow
![Page 20: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/20.jpg)
@mike_greenfield
Longitudinal Tests vs. Immediate Tests
• Longitudinal: change home page, email frequency, product framing
• Need to examine effect over a long period
• Immediate: change button color, email subject
• Likely that long-term effects will be minimal
![Page 21: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/21.jpg)
@mike_greenfield
Automatically Resolve Tests?
• Longitudinal tests should not be automatically resolved
• Example: new home page design
• Immediate tests can be automatically resolved when speed is important and there is one clear objective function
• Example: Circle of Moms email subject optimization
![Page 22: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/22.jpg)
@mike_greenfield
Choose robust statistics• Bad: # of page views• Good: % of users viewing at least [5,
25, 100] pages• Potentially bad: # of sales (when
small)• Potentially good: # of people getting
through the second step of a sales funnel
![Page 23: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/23.jpg)
@mike_greenfield
6. Celebrate A/B Testing Successes
![Page 24: [#500Distro] Measuring for Impact: Knowing When, What & How to A/B Test](https://reader036.vdocument.in/reader036/viewer/2022062509/58f9b3b8760da3da068bd900/html5/thumbnails/24.jpg)
@mike_greenfield