a/b testing that matters

29
@commagere Blake Commagere A/B Testing Technically Split or Bucket Testing, but nobody calls it that.

Upload: blake-commagere

Post on 18-Aug-2015

85 views

Category:

Technology


2 download

TRANSCRIPT

Page 1: A/B Testing That Matters

@commagereBlake Commagere

A/B Testing

Technically Split or Bucket Testing, but nobody calls it that.

Page 2: A/B Testing That Matters

Blake Commagere @commagere

Who the hell am I? It’s a fair question

●Started seven companiesoFive of those have been bought

●Raised over 12M Angel / VC●I build viral things

oPlaxooCausesoVampires

●I’m obsessed with human behavior

2

Page 3: A/B Testing That Matters

Blake Commagere @commagere

My Fucking TalksAnalogies Rule

  Dave McClure Blake Commagere

●Makes sailors blush with his swearing

●Occasionally uses hyperbole in his talks

●Is funny

●Makes Dave McClure blush with his swearing

●The most egregious abuse of hyperbole

●Thinks he’s funny

3

Page 4: A/B Testing That Matters

Blake Commagere @commagere

This talk is for you.

Interrupt at any point. Don’t be shy.

If it sucks and you never say anything, I’m blaming you.

4

Page 5: A/B Testing That Matters

Blake Commagere @commagere

 What I’m covering

●What the hell is A/B testing?

●Why should you care?

●How should you A/B test?

●How can A/B testing fuck you over?

●A few ancillary benefits of A/B testing

5

Page 6: A/B Testing That Matters

Blake Commagere @commagere

 What I’m NOT covering AKA “Shit you should Google if you don’t already know”

●Familiarize yourself with concepts likeoConfidence IntervalsoComparative ErroroStatistical SignificanceoSample Size requirements

FYI, there are online calculators for all these things

●How to fucking do mathoYes there’s a lot of math in A/B TestingoChances are, the tools you use will do the math for you

If not, your code better do the math for you

6

Page 7: A/B Testing That Matters

Blake Commagere @commagere

 What the hell is A/B testing?

7

Page 8: A/B Testing That Matters

Blake Commagere @commagere

What the hell is A/B Testing? Will it hurt?

●Randomized ExperimentoControl & VariantoTests against a specific goaloTypically you want to establish Statistical Significance with a Confidence Level of 95%

●Bucket or Split Testing oTechnically, this is what you want most of the timeoMultiple variantsoPeople usually say A/B when they mean Bucket/Split oJust go with it

8

Page 9: A/B Testing That Matters

Blake Commagere @commagere

 Why should you care?

9

Page 10: A/B Testing That Matters

Blake Commagere @commagere

 Do I really need A/B Testing? Technically, you may not need it

●Do you have users?●If not, do you plan on having users?●If not… congrats! You don’t need A/B Testing because you don’t have a business!

●For everyone else, you need A/B Testing.

10

Page 11: A/B Testing That Matters

Blake Commagere @commagere

You are not your user Your users are much, much stupider

  You Your User

●Busy as hell

●Dedicated

●Very knowledgeable

●Eager to learn

●Not Stupid

●Lazy

●Impatient

●Stupid

●Stupid

●Real. Stupid.

11

Page 12: A/B Testing That Matters

Blake Commagere @commagere

 The Result? You’re building the product for you

●You are not your user

●Your instincts are probably wrong

●Your wants/needs are probably different

●You view the market differently

12

Page 13: A/B Testing That Matters

Blake Commagere @commagere

 Still not convinced?

●Create and Make are synonyms

●“Make an Ad” vs “Create an Ad”

●Software Prefers ‘Create’ because REST, CRUD

●Make outperformed by up to 20% for non-artists

●There is a mental barrier to Create - it sounds more involved/complex. Making is easier.

13

 Create vs Make - one word makes all the difference

Page 14: A/B Testing That Matters

Blake Commagere @commagere

 How should you A/B Test

14

Page 15: A/B Testing That Matters

Blake Commagere @commagere

 Blockers on your A/B testing If this is happening, all your tests will suck

●Shit is brokenoFIX IT NOW WHY ARE YOU FUCKING READING THIS

●Shit is slowoPageSpeed on frontend, oProfilers on backend / mobile

●Official definitions of slow*oA webpage that takes > ~2 seconds & no progress

indicatoroAn app that isn’t immediately responsive

*according to me

15

Page 16: A/B Testing That Matters

Blake Commagere @commagere

 Ensuring tests maximize impact AKA the most obvious advice ever

1)ABT. Always Be Testing 2)Win as frequently as possible3)In areas that matter

●Any moron can get #1 right●Most morons can’t do #2 or #3●Essentially, you need to know what to test and how to test.

16

Page 17: A/B Testing That Matters

Blake Commagere @commagere

 Only a few features matter That’s it

●Users are either engaged or not ●Data on your userbase tells you which 2-3 features matter

●Optimize these features●Make Onboarding focus on these features●Facebook example:oFriends, Photos, Status UpdatesoNew User Experience focused on theseoNetwork leverage to help new users

17

Page 18: A/B Testing That Matters

Blake Commagere @commagere

 Similarly, only a few flows matter●Is a flow critical?odoes it get a user engaged with a critical featureodoes it help another user engage with a critical feature

●Is every step necessary?oAlmost always, the answer is no

●Where are the leaks?oAlmost always, this is the result of bad messagingoFind the worst leak, fix it, then move to the next one

18

 That’s also it

Page 19: A/B Testing That Matters

Blake Commagere @commagere

 Your messaging is never perfect AKA test your fucking messaging

●In most cases it is better to:oUse fewer words - your users are lazy oBe Colloquial - your users are dumboTarget the user - your users are selfish

●You can always target betteroSome data is implicit to your existing dataoYour users only care about what they get

●Language is always evolvingoLanguage in every channel is evolving as welloChannels change & language in a channel changes!

19

Page 20: A/B Testing That Matters

Blake Commagere @commagere

 Tools of the trade

●Web:oGoogle Analytics (Content Experiments)oOptimizelyoMixPanel

●Mobile:oMixPaneloOptimizelyoSwrve

20

Page 21: A/B Testing That Matters

Blake Commagere @commagere

 Which tool should I use? 

●The best one for your company:oIt dependsoOn. so. many. things.

●You’ll end up doing some coding too probably.

●The worst one for your company?oNot having oneoCreating your own from scratch

21

Page 22: A/B Testing That Matters

Blake Commagere @commagere

 How A/B Testing can fuck you over

22

Page 23: A/B Testing That Matters

Blake Commagere @commagere

 Pitfalls Some very common mistakes

●Find statistical significance sample size in advanceoYou want 95% confidence leveloWith 20,000 users, this can be 377 people. NOT 50/50

●Do not take convenience sampleso2-4 weeks is ideal

●Test in 7, 14, 28 day intervalsoHuman behavior is day dependent, tied to weekoYour results will skew if you do 8 day tests

23

Page 24: A/B Testing That Matters

Blake Commagere @commagere

The Lies We Tell Ourselves Sometimes we like lying because we’re lazy

●You can make the numbers lie

●Avoiding a channel that you hateoEmail worksoSEO worksoTwitter worksoFacebook worksoIf a channel doesn’t work for you, you’re probably doing it wrong

24

Page 25: A/B Testing That Matters

Blake Commagere @commagere

Farmville How to optimize for suck

  Farmville on Launch

Farmville after ~2 yrs

25

Page 26: A/B Testing That Matters

Blake Commagere @commagere

 Ancillary Benefits

26

Page 27: A/B Testing That Matters

Blake Commagere @commagere

Designer vs Engineer Your users are much, much stupider

  Designer Engineer

●Better not suck at design

●Better have design exp

●Better respect design

●Will not always agree with engineers on design

●Not Stupid

●Probably sucks at design

●May not have design exp

●Should respect design

●Will not always agree with designers on design

●Not Stupid

27

Page 28: A/B Testing That Matters

Blake Commagere @commagere

 Using A/B Testing for Team Bonding●The loudest voice usually gets their way

●This can stifle good ideas

●Sometimes it’s ok to test a bad idea

●Becomes a teachable moment

●You could be wrong (GASP!)

28

Page 29: A/B Testing That Matters

Blake Commagere @commagere

Want more swearing and/or advice?

[email protected]

@commagere

29