How I Learned to Stop Worrying and Love UX Metrics [PerfNow 2018]

Slide 1

Slide 1 text

@tameverts #PerfNow How I learned to stop worrying and love UX metrics

Slide 2

Slide 2 text

@tameverts WPOstats.com

Slide 3

Slide 3 text

What is the best UX metric? How fast should it be? How can we stay on track?

Slide 4

Slide 4 text

No content

Slide 5

Slide 5 text

https://www.cnet.com/news/appliance-science-the-well-done-physics-chemistry-of-the-toaster/

Slide 6

Slide 6 text

No content

Slide 7

Slide 7 text

No content

Slide 8

Slide 8 text

No content

Slide 9

Slide 9 text

TTFB DNS TCP Start render DOM loading DOM ready Page load Fully loaded TTI Resource timing Number of requests Bytes in Speed Index PageSpeed Navigation timing DOM elements DOM size Visually complete TTFP TTFMP

Slide 10

Slide 10 text

“The real thing we are after is to create a user experience that people love and they feel is fast… and so we might be front-end engineers, we might be dev, we might be ops, but what we really are is perception brokers.” Steve Souders

Slide 11

Slide 11 text

How do we measure perception… at scale?

Slide 12

Slide 12 text

Is it happening? Is it useful? Is it usable? Is it delightful? https://developers.google.com/web/fundamentals/ performance/user-centric-performance-metrics

Slide 13

Slide 13 text

No content

Slide 14

Slide 14 text

TTFB DNS TCP TTI FCP FMP FID OMG WTF

Slide 15

Slide 15 text

❑ Correlates to what users actually see in the browser ❑ Is easy to use and accessible right out of the box ❑ Recognizes that not all pixels and page elements are equal ❑ Allows us to customize what we measure on specific pages The best UX metric… #PerfNow @tameverts

Slide 16

Slide 16 text

No content

Slide 17

Slide 17 text

A brief history of performance metrics

Slide 18

Slide 18 text

Load Time the time from the start of the initial navigation until the beginning of the window load event

Slide 19

Slide 19 text

BBC loses 10% of users for every additional second it takes their site to load Ancestry.com saw a 7% conversion rate increase after improving load time by 64% FT.com reduced desktop load times to 1.5s and mobile to 2.1s and increased user engagement by 30% @tameverts #PerfNow

Slide 20

Slide 20 text

Start Render the time from the start of the initial navigation until the first non-white content is painted

Slide 21

Slide 21 text

start render repeat visits

Slide 22

Slide 22 text

wow!

Slide 23

Slide 23 text

@tameverts #PerfNow

Slide 24

Slide 24 text

Speed Index average time at which visible parts of the page are in the viewport

Slide 25

Slide 25 text

When good metrics go bad…

Slide 26

Slide 26 text

No content

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

No content

Slide 29

Slide 29 text

No content

Slide 30

Slide 30 text

First Paint First Contentful Paint First Meaningful Paint First CPU Idle First Interactive

Slide 31

Slide 31 text

First Paint (FP) Pixels first start to render

Slide 32

Slide 32 text

First Contentful Paint (FCP) Text and graphics start to render… BUT often catches non-meaningful paints (e.g. headers, nav bars) https://tinyurl.com/paint-metrics

Slide 33

Slide 33 text

First Meaningful Paint (FMP) The paint after which the biggest ATF layout change has happened and web fonts have loaded https://tinyurl.com/paint-metrics

Slide 34

Slide 34 text

No content

Slide 35

Slide 35 text

https://speedcurve.com/blog/an-analysis-of-chromiums-paint-timing-metrics/

Slide 36

Slide 36 text

“The rendering pipeline is very complex, and the timestamp should be the latest timestamp the browser is able to note in the pipeline (best effort). Typically the time at which the frame is submitted to the OS for display is recommended for this API.” https://w3c.github.io/paint-timing/#sec-terminology

Slide 37

Slide 37 text

Analysis of 40 top Alexa-ranked sites 95% of FP events occur before Start Render 85% of FCP events occur before Start Render 50% of FMP events occur before Start Render https://speedcurve.com/blog/ an-analysis-of-chromiums-paint-timing-metrics/

Slide 38

Slide 38 text

Slide 39

Slide 39 text

First CPU Idle (formerly known as First Interactive) Page is minimally interactive Most visible UI elements are interactive Responds to user input reasonably quickly

Slide 40

Slide 40 text

Time to Interactive (TTI) (formerly known as Time to Consistently Interactive) Displays useful content Event handlers are registered for most visible elements Page responds to user interaction within 50ms

Slide 41

Slide 41 text

No content

Slide 42

Slide 42 text

Slide 43

Slide 43 text

Okay… so then what?

Slide 44

Slide 44 text

Hero Rendering Times H1 Largest Image Largest Background Image First Painted Hero Last Painted Hero https://speedcurve.com/blog/web-performance-monitoring-hero-times/

Slide 45

Slide 45 text

No content

Slide 46

Slide 46 text

No content

Slide 47

Slide 47 text

No content

Slide 48

Slide 48 text

One-size-fits-most (not all) Dependent on how page is built Don’t work for image carousels and popups Hero times gotchas

Slide 49

Slide 49 text

So which metric is best?

Slide 50

Slide 50 text

http://lab.speedcurve.com/rendering/picker.php

Slide 51

Slide 51 text

And the winner is…

Slide 52

Slide 52 text

It depends! (You probably saw that coming.)

Slide 53

Slide 53 text

Deliver any content? Start Render Deliver significant amount of content? Speed Index, FMP Deliver critical content? Hero Rendering Times https://speedcurve.com/blog/rendering-metrics/

Slide 54

Slide 54 text

Slide 55

Slide 55 text

Custom metrics Measure performance with high-precision timestamps Available in both synthetic and RUM (yay!) https://www.w3.org/TR/user-timing/ https://speedcurve.com/blog/user-timing-and-custom-metrics/

Slide 56

Slide 56 text

how long does it take to display the main product image on my site?

Slide 57

Slide 57 text

Time to First Tweet The time from clicking the link to viewing the first tweet on each page’s timeline Pinner Wait Time (PWT) The time from initiating an action (e.g., tapping a pin) until the action is complete (pin close-up view is loaded) Time to Interact (TTI) @tameverts #PerfNow

Slide 58

Slide 58 text

@tameverts #PerfNow

Slide 59

Slide 59 text

Slide 60

Slide 60 text

No content

Slide 61

Slide 61 text

What else can we track?

Slide 62

Slide 62 text

First Click First Scroll First Key First Input Delay (only available with real user monitoring)

Slide 63

Slide 63 text

No content

Slide 64

Slide 64 text

No content

Slide 65

Slide 65 text

Great! We have metrics. Now… what to do with them?

Slide 66

Slide 66 text

Performance budgets FTW!

Slide 67

Slide 67 text

Thresholds YOU create for metrics that are meaningful for YOUR site https://addyosmani.com/blog/performance-budgets/ Milestone timings (e.g. start render) Quantity-based (e.g. image weight) Rules-based (e.g. Lighthouse scores)

Slide 68

Slide 68 text

A good performance budget should show you… What your budget is When you go out of bounds How long you’re out of bounds When you’re back within budget

Slide 69

Slide 69 text

No content

Slide 70

Slide 70 text

Recommended UX perf budgets* Metric How to track it Budget Start render Synthetic & RUM 2s First Painted Hero Synthetic 2s Speed Index Synthetic 4s First Interactive Synthetic 4s First Input Delay RUM 10ms @tameverts *See super important stuff on the next slide

Slide 71

Slide 71 text

Super important! Look at your own data Monitor your competitors No sandbagging allowed Take a step-by-step approach if necessary Use synthetic and RUM (numbers may will vary)

Slide 72

Slide 72 text

Pro tips Create budgets for your popular and regularly changing pages Review violations early and always Compare before and after releases Update budgets accordingly https://www.zillow.com/engineering/bigger-faster-more-engaging-budget/

Slide 73

Slide 73 text

Takeaways

Slide 74

Slide 74 text

1. There is no unicorn metric 2. Use your eyes 3. Don’t just look at averages and medians 4. Validate your metrics by correlating them with business and engagement KPIs 5. Set targets (e.g., performance budgets) 6. Make yourself accountable (e.g., alerts) 7. Don’t be afraid to investigate new metrics! @tameverts #PerfNow

Slide 75

Slide 75 text

thanks! @tameverts speedcurve.com/blog