Implementation worksheet · 5 min read
A Progressive Profiling Question-Selection Worksheet
Admit a question into onboarding only if it passes three tests, in writing: it changes the user's next experience (branches the flow, configures the workspace, personalizes genuinely), it can't be inferred from data you already have (email domain, invite context, SDK usage), and it's asked at the moment its answer is used — not stockpiled at signup. Everything else moves to later prompts, enrichment, or the bin. The worksheet scores each candidate question against all three; most teams' signup forms fail their own audit by half.
Progressive profiling degenerated from 'ask at the moment of use' into 'spread the interrogation across more screens'. The worksheet restores the original discipline: questions are purchases, paid in drop-off, and each must show what it bought.
Put it into practice
1. List every question you currently ask
Signup form, onboarding steps, first-run modals. Include the ones marketing added in 2024 that nobody remembers defending.
2. Score each against the three tests
Changes next experience? Uninferable? Asked at use-moment? One point each. Threes stay, twos move to their use-moment, ones and zeros die or route to enrichment.
3. Attach each survivor to its consumer
Name the system that consumes the answer: the branch fork, the template picker, the integration suggester. An answer with no named consumer is being collected, not used — that's a data request, and it goes through a different door with different consent.
4. Sequence by use-moment
Role at the fork. Team size when suggesting invites. Goal when the template gallery opens. The question arriving with its purpose visible converts better than the same question in a signup wall — the user can see what it buys them.
5. Re-audit quarterly
Questions accrete; consumers get deprecated. The report field whose dashboard died in Q2 is still taxing signups in Q4 unless someone owns this audit.
Question-selection worksheet
Copy this structure into your review document and record your observed result for each row.
| Question | Changes experience? | Uninferable? | At use-moment? | Verdict |
|---|---|---|---|---|
| Role | ||||
| Company size | ||||
| Use-case/goal | ||||
| Phone number | ||||
| Referral source |
A failure worth checking
The sales-wanted-it failure: 'company size' and 'phone number' in the signup wall because the CRM has columns for them. They fail all three tests — nothing in the product changes, both are enrichable, neither has a use-moment — and they tax every signup to pre-fill a call script for the 2% who get called. Route sales enrichment through enrichment; onboarding questions serve the user answering them.
Common questions
Doesn't asking more up front improve lead scoring?
It improves the score's inputs while shrinking the population being scored — drop-off compounds per field. Enrichment services and behavioral signals feed the same models without taxing conversion; make sales prove the phone field's win-rate delta beats its signup cost.
Is it ever right to ask something inferable?
When inference is unreliable AND the answer visibly personalizes — asking a gmail.com signup their company name because domain inference failed is fine at the moment the workspace gets named. The test is the experience change, not purity.
Basis and scope
This is a proposed implementation method using illustrative examples, not a measured benchmark or a customer case study. Prepared with AI assistance. Validate product-specific behavior against current documentation and your own test environment.