YK●

Case study 03 · Sheypoor marketplace

Save it once.
Find it sooner.

Turning an underused Save Search feature into a visible, low-friction part of browsing—then measuring each change independently.

My roleProduct Designer
ScopeDiscovery, flow, rollout
MethodsFeedback, testing, analytics
ProductLive marketplace

01 · Context

A useful feature
nobody noticed.

Save Search lets people keep their criteria and receive notifications when matching listings appear. Yet awareness and usage were extremely low, weakening a natural path back into Sheypoor.

Regular marketplace users—especially people repeatedly looking for homes, vehicles, or rentals—were manually rebuilding the same searches instead.

1core entry point: search results
3success signals to monitor
4interventions released in stages

02 · Research

Trace the whole
return loop.

The question was bigger than whether the save control was usable. I examined discovery, saving, notification delivery, and the moment users returned.

Research mixUser feedback and surveys were paired with usability tests, Google Analytics event tracking, and Microsoft Clarity session recordings. The evidence showed that discovery and notification quality were connected parts of the same retention loop.

Search → Save → Notify → Return

ADOPTION IS A SYSTEM

03 · Key insights

Low adoption had
more than one cause.

Visibility was the first barrier, but the post-save experience also had to earn trust.

01

Discovery came too late

The option appeared only after users applied several filters, so many never encountered it.

02

The flow blocked intent

Restrictions before saving added friction at exactly the moment users wanted continuity.

03

Alerts did not always help

Notification delays, SMS issues, and irrelevant alerts reduced the value of saving.

04

Return behavior had a rhythm

Clear peak hours showed when people were most likely to revisit saved searches.

04 · Design decisions

Meet intent
in the moment.

The redesign treated Save Search as a journey rather than a single control.

Remove unnecessary restrictions

Let people save earlier instead of requiring multiple filters first.

Make the feature visible in results

Add a dedicated section where recurring search intent is already clear.

Prompt in context

Use a lightweight toast after users browse several listings, when saving becomes useful.

Repair the notification loop

Improve timing and relevance while reducing incorrect or excessive alerts.

Surface value
at the moment
of repeated intent.

PROMPT, SAVE, RETURN

05 · Rollout & evidence

Learn one change
at a time.

Each improvement was released separately with gaps between launches. That made it possible to isolate behavioral impact instead of guessing which variable moved adoption.

4 staged releasesVisibility, flow restrictions, contextual prompts, and notification logic were separated.
Honest outcome framingUsage, retention, and engagement were monitored; no unsupported uplift is claimed here.

The redesign established a stronger foundation for Save Search by making it easier to discover, access, and understand inside the browsing journey.

NEXT CASE STUDY

Hyper Family
loyalty direction

→