Build a Zero-Party Preference Center From Scratch
Objective: Design and draft the actual field structure, copy, and value exchange for a preference center that a subscriber would willingly complete.
You're the lifecycle marketer at Nubank building the first preference center for the credit card waitlist list. Right now the only option subscribers have is 'unsubscribe.'
Draft a preference center form (fields, copy, value exchange) that captures zero-party data instead of just letting people leave.
Before you start
What you'll need
Free path (everything below is enough to finish)
Free, no account setup, fast to iterate on copy
Free tier supports flows and custom properties up to 250 contacts
The process
2 steps
Step 01 of 02
A preference center replaces the blunt unsubscribe link with granular choices: frequency and topic, so a subscriber who is mildly annoyed can self-correct instead of leaving entirely.
Nubank's waitlist list currently has one button: unsubscribe. Which fields turn that into a preference center a subscriber would actually use?
Procedure
- List every distinct email type Nubank currently sends (waitlist updates, product education, promotions, referral nudges)
- Turn that list into a topic-interest multi-select field, not a single dropdown
- Add a frequency field with three real options: weekly, monthly, only major updates
- Add one open value-exchange line above the form explaining what changes for the subscriber immediately after they submit
PREFERENCE CENTER DRAFT Topics (multi-select): [ ] Card launch updates [ ] Money tips [ ] Referral rewards Frequency (single-select): Weekly / Monthly / Major updates only Value-exchange line: 'Tell us what matters and we'll cut everything else.'
Healthy
Every field maps directly to a segment or send-frequency rule the ESP can act on the same day.
Unhealthy
A generic 'update your preferences' field with no options, which collects nothing usable and still reads as a stalling tactic before unsubscribe.
What this means
A preference center only earns its place if every field it collects changes what the subscriber receives next.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Unsubscribe rate rising faster than list growth | Ship the preference center as a link inside the unsubscribe confirmation page, not just the footer | 30 min |
Step 02 of 02
Progressive profiling spreads data collection across multiple touchpoints instead of asking everything at signup, one question in the welcome series, another after a first action.
The preference center only captured topic and frequency. What's the next single question to ask, and where does it go?
Procedure
- Pick the single highest-value unanswered question (e.g. 'What are you saving for?')
- Schedule it 5-7 days after the preference center submission, not immediately
- Map the answer to a Klaviyo custom property that a segment can filter on
FLOW: Preference Center Submitted -> wait 5 days -> Email: 'One quick thing' -> single reply-to-answer question -> tags contact with savings_goal property
Healthy
The follow-up question arrives after trust is already established and answers a gap the first form left open.
Unhealthy
Sending the follow-up question the same day, which reads as more friction stacked onto friction the subscriber just tried to reduce.
What this means
Progressive profiling works because it's paced, not because it asks fewer questions overall.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Preference center completion is high but profile data stays thin | Add one progressive-profiling touchpoint per lifecycle stage instead of expanding the original form | 30 min |
Final deliverable
A drafted preference center form plus a one-question progressive-profiling follow-up flow, both mapped to ESP properties.
See a reference example
Wise, Preference Center Draft (excerpt) Topics: [ ] Transfer rate alerts [ ] New country launches [ ] Multi-currency account tips Frequency: Weekly / Monthly / Major updates only Value-exchange line: 'Pick your topics, we'll skip the rest.' Follow-up flow (Day 5): 'Which currency do you send most often?' -> tags contact with primary_currency property
Success criteria
You're done when you can:
- Every preference field maps to a real segment or send rule, not a cosmetic dropdown
- The follow-up question is scheduled after the initial form, not bundled into it
- A clear value-exchange line appears before the fields, not after