🇵🇹 JNation 2025

Web Performance Meets Human Sciences

Website performance significantly impacts user experience, often leading to frustration and anger when pages are slow or unresponsive. Beyond psychological effects, poor performance can increase usage costs, and modern websites require a nuanced approach to capture continuous user interactions.

This session explores behaviors like rage clicks and cursor thrashing, emphasizing metrics that reflect user emotions and overall satisfaction. We’ll discuss the complexities of performance measurement, highlighting the need to look beyond single metrics. Ideal for those interested in understanding the business impact of site performance and the biases affecting user perceptions.

Transcript

1. WEB PERFORMANCE MEETS HUMAN SCIENCES • MAY 27, 2025. Olá, Coimbra! 👋 🇵🇹
2. WEB PERFORMANCE MEETS HUMAN SCIENCES MATHEUS ALBUQUERQUE • @ythecombinator
3. MATHEUS ALBUQUERQUE ↝ 👨💻 STAFF SWE @ MEDALLIA ↝ ⚛ CHAIR @ REACT SUMMIT NYC ↝ ⚡ GOOGLE DEVELOPER EXPERT ↝ 𝕏 YTHECOMBINATOR
4. Preamble WEB PERFORMANCE MEETS HUMAN SCIENCES
5. #question 🤔 What does it mean to be fast?
6. #question 🤔 What does it mean to be fast? If you were to summarize web performance in one metric, what’d be your pick?
7. WEB PERFORMANCE MEETS HUMAN SCIENCES If you were to summarize web performance in one metric, what’d be your pick?
8. WHAT DOES IT MEAN TO BE FAST?
9. WHAT DOES IT MEAN TO BE FAST?
10. WHAT DOES IT MEAN TO BE FAST?
11. #protip 💡 Speed metrics are not a single number.
12. “HACKABLE” METRICS…
13. WEB PERFORMANCE MEETS HUMAN SCIENCES
14. #protip 💡 Speed metrics can sometimes be hacked.
15. ↝ INDICATIONS OF ACTUAL USER BEHAVIOR: TIME SPENT ON A SITE, BOUNCE RATE, CONVERSIONS, AND REVENUE. ↝ UNFORTUNATELY, MOST OF THESE ARE ALSO PRIVATE CORPORATE DATA. “NON-HACKABLE” METRICS
16. “NON-HACKABLE” METRICS — STRENGTHENING THE LINK BETWEEN SITE SPEED AND BUSINESS OUTCOMES • ANDY DAVIES, 2018
17. “NON-HACKABLE” METRICS — STRENGTHENING THE LINK BETWEEN SITE SPEED AND BUSINESS OUTCOMES • ANDY DAVIES, 2018 Let’s talk about bias and frustration.
18. “NON-HACKABLE” METRICS — STRENGTHENING THE LINK BETWEEN SITE SPEED AND BUSINESS OUTCOMES • ANDY DAVIES, 2018 Let’s talk about measuring frustration.
19. Bias WEB PERFORMANCE MEETS HUMAN SCIENCES
20. “A cognitive bias is a systematic error in thinking that occurs when people process and interpret information in their surroundings, in fl uencing their decisions and judgments.” Judgment Under Uncertainty: Heuristics and Biases • 1982
21. COGNITIVE BIASES… — COGNITIVE BIAS CODEX • JOHN MANOOGIAN III, 2016
22. COGNITIVE BIASES… HOW WE SEE OUR DATA… OUTCOME BIAS • CONFIRMATION BIAS • CLUSTERING ILLUSION • SURVIVORSHIP BIAS HOW WE PERCEIVE TIME… INFORMATION PROCESSING • QUEUEING THEORY HOW USERS JUDGE APPS… SERIAL-POSITION EFFECT • PEAK-END RULE
23. HOW WE SEE OUR DATA… WEB PERFORMANCE MEETS HUMAN SCIENCES
24. ↝ OUTCOME BIAS: WHEN WE PICK DATA THAT CONFIRMS PAST OUTCOMES. ↝ CONFIRMATION BIAS: WHEN WE PICK DATA THAT CONFIRMS OUR PRE-EXISTING BELIEFS. ↝ CLUSTERING ILLUSION: WHEN WE SEE PATTERNS WHERE THEY DON'T EXIST. HOW WE SEE OUR DATA…
25. HOW WE SAMPLE OUR DATA…
26. SURVIVORSHIP BIAS
27. SURVIVORSHIP BIAS
28. SURVIVORSHIP BIAS
29. DEVICES & BANDWIDTH WEB PERFORMANCE MEETS HUMAN SCIENCES 💡 SURVIVORSHIP BIAS
30. DEVICES: ADOPTION
31. DEVICES: PARSE TIMES — JAVASCRIPT START-UP PERFORMANCE • 2017
32. BANDWIDTH: TECHNOLOGIES — GSM ASSOCIATION • 2022
33. 6.07 (MBPS) 44.31 (MBPS) — THE STATE OF LTE BY OPEN SIGNAL • 2018 BANDWIDTH: LTE SPEED
34. BANDWIDTH: DOWNLOAD SPEED — INTERNET SPEEDS BY COUNTRY • WORLD POPULATION REVIEW, 2025 MEAN DOWNLOAD SPEED 2024 (MBPS)
35. BANDWIDTH: AFFORDABILITY — FIXED-BROADBAND INTERNET BASKET • ITU DATAHUB, 2025
36. BANDWIDTH: AFFORDABILITY — FIXED-BROADBAND INTERNET BASKET • ITU DATAHUB, 2025
37. WEB PERFORMANCE MEETS HUMAN SCIENCES
38. BANDWIDTH MIX: AFFORDABILITY — USE TOOLS TO MEASURE PERFORMANCE • SAM DUTTON
39. POLITICAL CONTEXT 💡 SURVIVORSHIP BIAS WEB PERFORMANCE MEETS HUMAN SCIENCES
40. WEB PERFORMANCE MEETS HUMAN SCIENCES
41. POLITICAL CONTEXT: CHINA
42. POLITICAL CONTEXT: CHINA — DECODING THE WEB IN CHINA FOR PEAK WEB PERFORMANCE • JODIE CHAN, 2023
43. A11Y WEB PERFORMANCE MEETS HUMAN SCIENCES 💡 SURVIVORSHIP BIAS
44. ↝ GENERALLY, YOU CAN'T OPERATE A SCREEN READER WITHOUT A KEYBOARD. SO, ANY OF THE THINGS THAT IMPACT KEYBOARD USAGE WILL MOST LIKELY AFFECT A SCREEN READER. ↝ E.G. SLUGGISH USABILITY, DELAYED INPUT, AND FOCUS BEING DROPPED AS THE PAGE IS UPGRADED WITH JAVASCRIPT. A11Y ∩ PERFORMANCE
45. A11Y ∩ PERFORMANCE: DOM SIZE
46. A11Y ∩ PERFORMANCE: REFLOWS
47. WEB PERFORMANCE MEETS HUMAN SCIENCES
48. A11Y ∩ PERFORMANCE: METRICS?
49. A11Y ∩ PERFORMANCE: METRICS?
50. A11Y ∩ PERFORMANCE
51. WRAPPING UP… WEB PERFORMANCE MEETS HUMAN SCIENCES
52. EVEN PATIENCE IS A LOCAL THING… — UNDERSTANDING EMOTION FOR HAPPY USERS • PHILIP TELLIS, 2020
53. HOW USERS JUDGE APPS… WEB PERFORMANCE MEETS HUMAN SCIENCES
54. HOW USERS JUDGE APPS…
55. ↝ SERIAL-POSITION EFFECT: PEOPLE RECALL THE FIRST AND LAST ITEMS IN A SERIES AS BEST, AND THE MIDDLE ITEMS WORST. ↝ PEAK-END RULE: PEOPLE JUDGE AN EXPERIENCE LARGELY BASED ON HOW THEY FELT AT ITS PEAK & AT ITS END. HOW USERS JUDGE APPS…
56. ↝ RETENTION RATE COULD BE INFLUENCED BY THE BEST & WORST PERFORMANCE ON FIRST & LATEST PAGES. ↝ CONVERSION RATE: COULD DEPEND ON THE BEST & WORST PERFORMANCE ON THE PAGE BEFORE THE CONVERSION. HOW USERS JUDGE APPS…
57. HOW WE PERCEIVE TIME… WEB PERFORMANCE MEETS HUMAN SCIENCES
58. #question 🤔 Do you also feel like time speeds up as you age?
59. TIME PERCEPTION
60. TIME PERCEPTION: ELEVATORS
61. TIME PERCEPTION: AIRPORTS
62. TIME PERCEPTION: COMPUTERS IN 1968 — RESPONSE TIME IN MAN-COMPUTER CONVERSATIONAL TRANSACTIONS • ROBERT B. MILLER, 1968
63. — USABILITY ENGINEERING • JAKOB NIELSEN, 1993 TIME PERCEPTION: COMPUTERS IN 1993
64. — MEASURE PERFORMANCE WITH THE RAIL MODEL • GOOGLE TIME PERCEPTION: CHROME IN 2015
65. TIME PERCEPTION: CHROME IN 2015
66. TIME PERCEPTION: REACT
67. TIME PERCEPTION: REACT PRIORITY TIMEOUT WHEN I m m ediate SYNCHRONOUSLY TASKS THAT NEED TO RUN SYNCHRONOUSLY UserBlocking 250MS RESULTS OF A USER INTERACTION (E.G. A BUTTON CLICK) Normal 5S UPDATES THAT DON’T HAVE TO FEEL INSTANTANEOUS Low 10S TASKS THAT CAN BE DEFERRED BUT MUST STILL COMPLETE EVENTUALLY (E.G. AN ANALYTICS NOTIFICATION) Idle NO TIMEOUT TASKS THAT DO NOT HAVE TO RUN AT ALL (E.G. HIDDEN OFFSCREEN CONTENT)
68. Frustration WEB PERFORMANCE MEETS HUMAN SCIENCES
69. FRUSTRATION ↝ STRESS Brainwave analysis from the experiment revealed that participants had to concentrate up to 50% more when using websites via the slower connection. — Study conducted by Foviance on behalf of CA • 2011
70. FRUSTRATION ↝ ANXIETY Delayed web pages caused a 38% rise in mobile users' heart rates — equivalent to the anxiety of watching a horror movie alone. — Ericsson ConsumerLab • 2015
71. FRUSTRATION ↝ RAGE 40% of Brits reported that they had become physically violent toward their computers. — British Psychology Society • 2009
72. FRUSTRATION ↝ RAGE — TIME IS MONEY • TAMMY EVERTS, 2016
74. Measuring Frustration WEB PERFORMANCE MEETS HUMAN SCIENCES
75. ASKING THE USERS WEB PERFORMANCE MEETS HUMAN SCIENCES
76. WEB PERFORMANCE MEETS HUMAN SCIENCES
77. ↝ REQUIRES ACTIVE ENGAGEMENT FROM THE USER. ↝ IT’S HARD TO DESIGN THE POPOVER TO MAKE USERS CHOOSE TO ANSWER THE SURVEY (INCREASE SELF- SELECTION). ↝ PRONE TO SELECTION BIAS AND THE HAWTHORNE EFFECT. ASKING THE USERS
78. ASKING THE USERS
79. AFFECTIVE COMPUTING WEB PERFORMANCE MEETS HUMAN SCIENCES
80. WEB PERFORMANCE MEETS HUMAN SCIENCES
81. AFFECTIVE COMPUTING
82. AFFECTIVE COMPUTING
83. BEHAVIOR ANALYSIS WEB PERFORMANCE MEETS HUMAN SCIENCES
84. WEB PERFORMANCE MEETS HUMAN SCIENCES
85. WEB PERFORMANCE MEETS HUMAN SCIENCES RAGE CLICKS 💡 BEHAVIOR ANALYSIS
86. “Rage clicks occur when users rapidly click (or tap) on your site or app. Rage clicking is the digital equivalent of cursing to release frustration.” Severity and impact of computer user frustration • 2006
87. RAGE CLICKS
88. RANDOM SCROLLING WEB PERFORMANCE MEETS HUMAN SCIENCES 💡 BEHAVIOR ANALYSIS
89. ↝ RAPID SCROLLING, OFTEN THROUGH LARGE CHUNKS. ↝ INDICATES THAT A USER IS ON THE HUNT FOR THEIR NEXT STEP. ↝ PEOPLE HAVE LIMITED ATTENTION SPANS AND SCROLLING FEELS LIKE EXTRA WORK SO THIS CAN QUICKLY CAUSE USERS TO LOSE HOPE. RANDOM SCROLLING
90. RANDOM SCROLLING
91. WEB PERFORMANCE MEETS HUMAN SCIENCES
92. THRASHED CURSOR • WILD MOUSE WEB PERFORMANCE MEETS HUMAN SCIENCES 💡 BEHAVIOR ANALYSIS
93. THRASHED CURSOR
94. ↝ MOVEMENT COVERING A HIGH DISTANCE TRAVELLED IN A SHORT AMOUNT OF TIME, WITH NO INTENTIONALITY. ↝ INDICATES IMPATIENCE, DOUBT, DIFFICULTY, OR ANXIETY. THRASHED CURSOR
95. POSSIBLE ISSUES: ↝ PERFORMANCE: IF IT OCCURS WHILE USERS ARE WAITING. E.G., A VIDEO BUFFERING OR THE NEXT ROUTE LOADING. ↝ COGNITIVE LOAD: IF IT OCCURS WHILE THEY'RE ENGAGED IN AN ACT. E.G., FILLING OUT, CREATING, WRITING, OR OTHER CHALLENGING TASK. THRASHED CURSOR
96. THRASHED CURSOR
97. CORRELATIONS WEB PERFORMANCE MEETS HUMAN SCIENCES
98. CORE WEB VITALS WEB PERFORMANCE MEETS HUMAN SCIENCES 💡 CORRELATIONS
99. LCP ⁵ RETENTION RATE — CORE WEB VITALS AND THEIR EFFECT ON USER EXPERIENCE • PHILIP TELLIS, 2025
100. CLS ⁵ RELOAD RATE — CORE WEB VITALS AND THEIR EFFECT ON USER EXPERIENCE • PHILIP TELLIS, 2025
101. INP ⁵ RAGE CLICK RATE — CORE WEB VITALS AND THEIR EFFECT ON USER EXPERIENCE • PHILIP TELLIS, 2025
102. UX-IMPACTING METRICS… WEB PERFORMANCE MEETS HUMAN SCIENCES 💡 CORRELATIONS
103. ↝ SCROLL EFFECTS WORK. ↝ APP RESPONDS TO CLICKS. ↝ MOST MEANINGFUL VIDEOS/ANIMATIONS RUN. ↝ PRIMARY FEATURE SHOWS UP/IS INTERACTIVE. ↝ COOKIE BANNER SHOWS UP. TIME UNTIL THE…
104. TIME UNTIL THE…
105. BUSINESS OUTCOMES WEB PERFORMANCE MEETS HUMAN SCIENCES 💡 CORRELATIONS
106. WEB PERFORMANCE MEETS HUMAN SCIENCES
107. BUSINESS OUTCOMES…
108. BUSINESS OUTCOMES…
109. CORRELATIONS BIAS BEHAVIOR ANALYSIS FRUSTRATION
110. Closing Thoughts WEB PERFORMANCE MEETS HUMAN SCIENCES
111. You can’t fi x what you can’t measure. #1 of 5 WEB PERFORMANCE MEETS HUMAN SCIENCES
112. Always try to correlate behavior analysis with contextual metrics and business outcomes to tell the full story. #2 of 5 WEB PERFORMANCE MEETS HUMAN SCIENCES
113. Core Web Vitals are a better starting point than a fi nish line. #3 of 5 — IN THE BLINK OF AN EYE • TIM KADLEC , 2024 WEB PERFORMANCE MEETS HUMAN SCIENCES
114. Acknowledging when you didn't meet the user’s expectations can alleviate negative perception. #4 of 5 WEB PERFORMANCE MEETS HUMAN SCIENCES
116. WEB PERFORMANCE MEETS HUMAN SCIENCES
117. There's probably a business case for making your app faster. But web performance is about more than “just” business. #5 of 5 WEB PERFORMANCE MEETS HUMAN SCIENCES
118. MORE THAN “JUST” BUSINESS!
119. MORE THAN “JUST” BUSINESS!
120. THAT’S ALL, FOLKS! OBRIGADO! 👋 🇵🇹 QUESTIONS? MATHEUS ALBUQUERQUE • @ythecombinator ⬆ ALL THE LINKS!