Slide 1

Slide 1 text

Measuring and Improving Performance Insights from one of Cash App’s main screens

Slide 2

Slide 2 text

Hi, I’m Colin Marsch Android Engineer @ Cash App

Slide 3

Slide 3 text

Overview

Slide 4

Slide 4 text

Overview • Where we started

Slide 5

Slide 5 text

Overview • Where we started • What metrics we are tracking

Slide 6

Slide 6 text

Overview • Where we started • What metrics we are tracking • Collecting the data

Slide 7

Slide 7 text

Overview • Where we started • What metrics we are tracking • Collecting the data • Setting up monitoring

Slide 8

Slide 8 text

Overview • Where we started • What metrics we are tracking • Collecting the data • Setting up monitoring • Challenges we faced

Slide 9

Slide 9 text

Overview • Where we started • What metrics we are tracking • Collecting the data • Setting up monitoring • Challenges we faced • Discoveries

Slide 10

Slide 10 text

I work on Cash App’s Money Tab

Slide 11

Slide 11 text

Where we started

Slide 12

Slide 12 text

Only scroll performance tracked

Slide 13

Slide 13 text

Only scroll performance tracked

Slide 14

Slide 14 text

Why now?

Slide 15

Slide 15 text

Why now? • Compare data during a Kotlin Multiplatform migration

Slide 16

Slide 16 text

Why now? • Compare data during a Kotlin Multiplatform migration • Have visibility into the features on the screen

Slide 17

Slide 17 text

Why now? • Compare data during a Kotlin Multiplatform migration • Have visibility into the features on the screen • Ensure Cash App is performant for all customers

Slide 18

Slide 18 text

What metrics to track

Slide 19

Slide 19 text

What metrics are we tracking? Scroll performance Load times Stale Data ?? ?? ?? ??

Slide 20

Slide 20 text

What metrics are we tracking? Scroll performance Load times Stale Data ?? ?? ?? ??

Slide 21

Slide 21 text

What metrics are we tracking? Scroll performance Load times Stale Data ?? ?? ?? ??

Slide 22

Slide 22 text

What metrics are we tracking? Scroll performance Load times Stale Data ?? ?? ?? ??

Slide 23

Slide 23 text

Overall load time

Slide 24

Slide 24 text

Overall load time

Slide 25

Slide 25 text

Applet load times

Slide 26

Slide 26 text

No content

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

data class Applet( val id: AppletId, val state: AppletState, )

Slide 29

Slide 29 text

data class Applet( val id: AppletId, val state: AppletState, )

Slide 30

Slide 30 text

sealed interface AppletState { data class Loading(...) : AppletState data class Unadopted(...) : AppletState data class Adopted(...) : AppletState data class Unavailable(...) : AppletState data class Error(...) : AppletState }

Slide 31

Slide 31 text

sealed interface AppletState { data class Loading(...) : AppletState data class Unadopted(...) : AppletState data class Adopted(...) : AppletState data class Unavailable(...) : AppletState data class Error(...) : AppletState }

Slide 32

Slide 32 text

sealed interface AppletState { data class Loading(...) : AppletState data class Unadopted(...) : AppletState data class Adopted(...) : AppletState data class Unavailable(...) : AppletState data class Error(...) : AppletState }

Slide 33

Slide 33 text

Savings: Adopted ✅ Bitcoin: Loading
 Stocks: Loading Taxes: Loading Savings: Adopted ✅ Bitcoin: Adopted ✅
 Stocks: Loading Taxes: Unavailable ✅ Savings: Adopted ✅ Bitcoin: Adopted ✅
 Stocks: Adopted ✅ Taxes: Unavailable ✅

Slide 34

Slide 34 text

Savings: Adopted ✅ Bitcoin: Loading
 Stocks: Loading Taxes: Loading Savings: Adopted ✅ Bitcoin: Adopted ✅
 Stocks: Loading Taxes: Unavailable ✅ Savings: Adopted ✅ Bitcoin: Adopted ✅
 Stocks: Adopted ✅ Taxes: Unavailable ✅

Slide 35

Slide 35 text

Savings: Adopted ✅ Bitcoin: Loading
 Stocks: Loading Taxes: Loading Savings: Adopted ✅ Bitcoin: Adopted ✅
 Stocks: Loading Taxes: Unavailable ✅ Savings: Adopted ✅ Bitcoin: Adopted ✅
 Stocks: Adopted ✅ Taxes: Unavailable ✅

Slide 36

Slide 36 text

Collecting the data

Slide 37

Slide 37 text

In-house analytics

Slide 38

Slide 38 text

In-house analytics Pros:

Slide 39

Slide 39 text

In-house analytics Pros: • Infrastructure already set up in Cash App

Slide 40

Slide 40 text

In-house analytics Pros: • Infrastructure already set up in Cash App • Data fl ows to visualization tools our teams are familiar with

Slide 41

Slide 41 text

In-house analytics Pros: • Infrastructure already set up in Cash App • Data fl ows to visualization tools our teams are familiar with Cons:

Slide 42

Slide 42 text

In-house analytics Pros: • Infrastructure already set up in Cash App • Data fl ows to visualization tools our teams are familiar with Cons: • Large cost for high volume events

Slide 43

Slide 43 text

In-house analytics Pros: • Infrastructure already set up in Cash App • Data fl ows to visualization tools our teams are familiar with Cons: • Large cost for high volume events • Visualization tools are mainly business focused

Slide 44

Slide 44 text

Datadog

Slide 45

Slide 45 text

Datadog Pros:

Slide 46

Slide 46 text

Datadog Pros: • Focused on engineering metrics

Slide 47

Slide 47 text

Datadog Pros: • Focused on engineering metrics • Used in Cash App backend services

Slide 48

Slide 48 text

Datadog Pros: • Focused on engineering metrics • Used in Cash App backend services • Many supported default metrics

Slide 49

Slide 49 text

Datadog Pros: • Focused on engineering metrics • Used in Cash App backend services • Many supported default metrics Cons:

Slide 50

Slide 50 text

Datadog Pros: • Focused on engineering metrics • Used in Cash App backend services • Many supported default metrics Cons: • External dependency risks (APK size, app startup, etc)

Slide 51

Slide 51 text

Datadog Pros: • Focused on engineering metrics • Used in Cash App backend services • Many supported default metrics Cons: • External dependency risks (APK size, app startup, etc) • Additional e ff ort to implement and maintain

Slide 52

Slide 52 text

Setting up monitoring

Slide 53

Slide 53 text

Automated alerting

Slide 54

Slide 54 text

Automated alerting • Main focus on overall and applet load times

Slide 55

Slide 55 text

Automated alerting • Main focus on overall and applet load times • Possible alerts on secondary metrics (e.g. scroll performance)

Slide 56

Slide 56 text

Early rollout to beta groups

Slide 57

Slide 57 text

Clear dashboards Easy to understand for our team and other stakeholders

Slide 58

Slide 58 text

Challenges we faced

Slide 59

Slide 59 text

Challenges we faced

Slide 60

Slide 60 text

Challenges we faced • Architecture

Slide 61

Slide 61 text

Challenges we faced • Architecture • Customer-based measurement

Slide 62

Slide 62 text

Challenges we faced • Architecture • Customer-based measurement • Platform consistency

Slide 63

Slide 63 text

Discoveries

Slide 64

Slide 64 text

Features perform differently

Slide 65

Slide 65 text

Some staleness is inevitable

Slide 66

Slide 66 text

Lessons learned

Slide 67

Slide 67 text

Lessons learned • Architecture choices have long term impact

Slide 68

Slide 68 text

Lessons learned • Architecture choices have long term impact • Performance monitoring is worth the e ff ort

Slide 69

Slide 69 text

Lessons learned • Architecture choices have long term impact • Performance monitoring is worth the e ff ort • Performance will always need to be top of mind

Slide 70

Slide 70 text

Stay tuned! code.cash.app or a future Droidcon talk