Meituan · Local Services · IM Consultation

Designing Trust Before the Bill

Turning uncertain local-service pricing into a guided, comparable, bookable decision.

+5%

Conversion lift · ~2k extra daily orders · −50% disputes.

Timeline: 4 weeks · 2025
Ownership: End-to-end · sole designer
Impact: +5% conversion · validated via user-level A/B

Live prototype · interact above

Context & Signal

Users wanted to ask first, but the platform was not earning trust.

User behavior signal

Users visit 10 merchants, consult 6, spend 30 minutescomparing — and still don't trust the price.

Behavior

Consult outranked book and call.

Conversion

In-app consult stayed low.

Experience

Disputes clustered in home repair.

Turning Point

The brief asked for price visibility. The evidence pointed deeper.

Price was not a number problem. It was a process trust problem.

Before · 4-step linear journey

grayscale

01
Problem occurs
User finds an issue like toilet repair and opens Meituan to search.
02
Many merchants appear
User sees options but cannot tell who can diagnose accurately.
03
One-by-one outreach
Repeats the same questions across shops; waits in fragmented threads.
04
Pick one for visit
Merchants only promise an arrival fee; final quote deferred to on-site.

→ Trust break

Quoted price ≠ actual bill. Scope, materials, and conditions widen the range. Users feel the system did not warn them.

After · 3-step trust loop

redesigned

01
Diagnose the problem
Certified experts surface from search to define the issue — remove ambiguity before comparison.
02
Structure the intent
Multi-turn chat yields a service-order card so quotes compare on equal terms.
03
Compare and commit
Vetted merchants quote live; the chosen price threads into checkout, locked.

→ Trust restored

Price is the output of a credible process. Range is explained by structured intent — what the user agreed to is what they pay.

How do we guide users through a credible path to a price they can trust before they commit?

System Design

One end-to-end flow. Trust compounds across every stage.

IM Experience

Three entry states, one interaction model.

01 · Standard repair

Live

Diagnose → service order → live quotes → booked.

02 · Self-serve fix

Live

Resolves in-thread when a visit isn't needed.

03 · After hours

Live

Closed state stays explicit, never a dead end.

Each phone is the live prototype booted into one flow — tap the suggested replies to play it through.

Quoting Engine

Conversation becomes a contract. Merchants quote against it.

Diagnosis in action

Diagnosis start in chat — Vague problem → structured intent.

Post diagnosis recommendation state — Recommendations only after confidence.

Competitive quoting

Show progress before price.

Live updates make waiting legible

Trust signals appear before price

Range, not a single number

Reset only what is unsafe to assume.

Expired quotes stay visible but disabled

Only the time slot resets

Hard expiry, soft continuity

Closing the loop

The trust loop closes where it began.

Return flow stays in the same thread

Re-engagement is one tap

Interactive Prototype

Try the full flow.

Framework Extensions

The same loop scales: education, banquet, maternity care.

Domain

01 · Diagnose

02 · Structure

03 · Commit

Home repair

Reference case

Certified expert defines the issue.

Chat → structured service-order card.

Live competitive quote → checkout.

Education consultation

Parents · advisors · plans

Advisor clarifies goals, grade, budget, schedule.

Constraints → structured learning brief.

Plan comparison with visible recommendation quality.

Banquet booking

Event · venue · quote lock-in

Capture event size, date flex, menu tier, must-haves.

Non-negotiables → comparable requirement card.

Venue offers on equal terms; explicit quote boundaries.

Maternity care

Family · caregiver · continuity

Triage need and risk; surface caregiver credentials first.

Care scope, boundaries, schedule → service brief.

Package selection in trust context; follow-up in same thread.

Home repair

Reference case

01 · Diagnose: Certified expert defines the issue.
02 · Structure: Chat → structured service-order card.
03 · Commit: Live competitive quote → checkout.

Education consultation

Parents · advisors · plans

01 · Diagnose: Advisor clarifies goals, grade, budget, schedule.
02 · Structure: Constraints → structured learning brief.
03 · Commit: Plan comparison with visible recommendation quality.

Banquet booking

Event · venue · quote lock-in

01 · Diagnose: Capture event size, date flex, menu tier, must-haves.
02 · Structure: Non-negotiables → comparable requirement card.
03 · Commit: Venue offers on equal terms; explicit quote boundaries.

Maternity care

Family · caregiver · continuity

01 · Diagnose: Triage need and risk; surface caregiver credentials first.
02 · Structure: Care scope, boundaries, schedule → service brief.
03 · Commit: Package selection in trust context; follow-up in same thread.

Impact & Validation

Trust-first won the A/B.

+0%

Conversion lift

Search-to-purchase, validated via user-level randomized A/B.

~0k

Additional daily orders

Incremental volume at projected rollout coverage.

−0%

Pricing disputes

Post-service complaints in this flow.

User-level randomized A/B · Meituan + Dianping

Reflection

Next time, I would push on three fronts.

Merchant experience deserves its own product pass.

Guide pricing should explain variability, not imply a promise.

Scale with AI triage, escalate to human experts.

Transparent process is often a stronger trust advantage than transparent pricing alone.

Designing trust before the bill · 2025