Good Friction vs Bad Friction
Adding friction that addresses the user's psyche can increase conversion, because not all friction is bad
- Difficulty
- Easy
- Time to result
- ~weeks to results
- Steps
- 3
- Confidence
- 92%
The common assumption is that all friction in a signup or onboarding flow hurts conversion. Schaffer's reframe: friction that speaks to what a first-time user is anxiously wondering ('does this support my case? can I even do this?') is reassuring and can lift conversion. The test for good friction is whether the added step alleviates a real psychological concern the user already has.
Origin
Laura Schaffer discovered this from Twilio's first growth experiment, where adding qualifying questions to a signup flow (against strong internal conviction that developers want zero friction) raised signups ~5% with no personalization.
Core principles
- 01Bad friction is bad and good friction is good — there is no rule that all friction is bad
- 02First-time users arrive scared, expecting difficulty and looking for the 'bogeyman'
- 03Questions that mirror what a user is organically wondering feel comforting, not obstructive
- 04The psyche of the user at each stage matters as much as understanding your product and market
How to run it
- 1
Identify the user's first-time anxieties
Work out what a new user is worried about at the front door: will it support my language/use case, am I in the right place, can I even do this.
- 2
Design friction that answers those anxieties
Add steps or questions that let the user confirm their situation is supported — e.g. 'what language are you coding in?', 'what's your use case?', 'what product do you want to use?'
Pro tip Well-chosen questions double as valuable data (developer vs non-developer split, language trends) for growth, docs, and product teams.
- 3
Test and confirm with qualitative feedback
Run the experiment and talk to users who went through the flow to verify why it worked — Schaffer learned the questions felt 'comforting' by interviewing customers.
Watch out Diagnosing why friction helped requires qualitative interviews; the quantitative lift alone won't explain the mechanism.
In the wild
Twilio's signup asked only for username, email, and password, backed by strong conviction that developers want no friction. Schaffer added ~four dropdown questions (coding language / are you a developer, use case, product of interest) and, with early employee Andre, quietly shipped an A/B test to a small group on a Friday evening, expecting conversion to drop.
→ Signups improved by about 5% with no personalization — just the questions. Interviews revealed the questions reassured anxious first-time users that they were in the right place, and the responses gave critical data on developer-vs-non-developer mix and languages.
Common mistakes
Treating all friction as harmful
Blanket friction-aversion removes reassuring signals users want, and it leaves you blind — collecting nothing about who signs up or what they want to do, hurting prioritization.
Assuming your audience wants zero questions
Twilio's high-conviction belief that developers want no friction was wrong for signup; untested assumptions about audience preferences can suppress conversion and data.
Is it for you?
Best for
Growth/PM teams optimizing signup or early-onboarding flows for products where new users feel uncertain or intimidated
Not ideal for
Adding friction blindly without tying each step to a real user anxiety — arbitrary friction still hurts
From the transcript
“it improved conversion by like five percent like just improve signups”
“the psyche of the user is so so critical”
“bad friction is bad and good friction is good right there's no such thing as being simple as just all friction is bad”
“it was actually giving folks something comforting”
From the episode
Career frameworks, A/B testing mistakes, counterintuitive onboarding tips, selling to developers
Laura Schaffer (VP of Growth at Amplitude)