over MCP · works inside Claude & Codex

Launch with the store page that already won.

A weak icon at launch makes every install cost more, for months. StoreLab splits real ad traffic across pixel-accurate App Store and Google Play pages and tells you which variant to ship, before launch day. And it lives in your agent: ask Claude for a test, approve the publish, ask who's winning.

€300/mo
where enterprise listing tests start. $0 where yours does
~4 min
to your first live test
0
SDKs, store reviews, or sales calls
+23%
install-intent lift in the demo's icon test
claude · StoreLab MCP
youS
a scripted session · every tool shown ships today
01the problem

Launch day is a blind bet.

The native tools arrive too late.

Google's Store Listing Experiments and Apple's Product Page Optimization only work on a live app. Your launch traffic, the most expensive traffic you'll ever buy, is spent before they can tell you anything.

The testing vendors price you out.

Geeklab starts at €300/mo for one test a month, with consultants bundled whether you want them or not. Others don't publish prices at all; you get a sales call.

So most studios guess.

Icon picked in Slack, screenshots from the last build, validated by real UA budget. StoreLab exists so the guess never ships.

02work from your agent

The dashboard is optional.

Every StoreLab feature is a tool your agent can call over MCP. Fourteen tools cover the whole run: draft, creative, publish, results, decision.

The whole demo above is real tool calls. Nothing shown is a mockup.

connect an mcp client
claude mcp add --transport http StoreLab https://mcp.7play.ai/mcp
/plugin marketplace add 7play-ai/storelab-plugin then /plugin install storelab@storelab-pluginor the plugin,github.com/7play-ai/storelab-plugin
Codex, Cursor or another client? Full setup at storelab.7play.ai/mcp
Ship a test from chat
Your agent creates the experiment, uploads the creative, and publishes. It shows you the preflight and asks before anything goes live.
Results without the UI
Ask who is winning, or let your agent check on a schedule and report back. The answer is the decision object: winner, lift, probability, and the warnings that say when not to call it.
Clone in one prompt
Assets are reusable. Replicate your iOS test for Android without re-uploading a single file.
03live simulation

Watch a test decide.

This is how a StoreLab test runs. Each ad click routes a new visitor to a different variant, so the store page changes as traffic splits. Tap around, hit install, and watch every tap roll up into the decision. Or press play and let it run itself.

01 · your ad
candychase
Sponsored
Candy Chase
Install now
8,412 likes
candychase catch every candy raccoon before they grab the loot
any network · meta, tiktok, google
02 · look-alike store page
Candy Chase!
Candy Chase! icon
Candy Chase!
Catchem all
In-App
Purchases
12K Ratings
4.8
Age Rating
4+
Years
Chart
No.4
Puzzle
Developer
Meadowbyte Games
Language
EN
+ 3 More
Size
214.8
MB
Candy Chase! screenshot 1
Candy Chase! screenshot 2
Candy Chase! screenshot 3
What's New
Version History
Version 3.4.1
2d ago
New Spring Meadow season with 20 rare blooms to collect. Smoother merges, faster loading, and a cozy new soundtrack.
Chase the candy raccoons through the gingerbread market. Match, catch, and collect every sweet in this cozy puzzle. No timers, just a growing candy stash.
Meadowbyte Games
Developer
Ratings & Reviews
4.8
out of 5
12K Ratings
My comfort game
2d ago
I open this every night to unwind. Zero pressure, endlessly satisfying merges.
meadowmaker
Beautiful and relaxing
5d ago
Gorgeous art and the seasonal gardens keep it fresh.
lila.grows
Information
Seller
Meadowbyte Games
Size
214.8 MB
Category
Games
Languages
English & 11 more
Age Rating
4+
In-App Purchases
Yes
This preview page is used for market research and may redirect to an official app store when available.
waiting for traffic
pixel-accurate app store & google play
03 · decision
waiting for first visitor…
click the ad to route traffic
install-intent · lift · confidence
Run this on your app
every tap is tracked and rolled straight into the decision · no SDK, no store review
04surface area

Every element is a variable.

Test one thing or the whole listing. Pick an element to see the change that moved install-intent, and by how much.

App name
the highest-impact ASO lever
A punchier verb in the title moved install-intent more than any other single change we tested.
+34%
install-intent
Screenshots & order
the first frame decides
Leading with gameplay instead of the title card kept far more visitors scrolling past the fold.
+27%
install-intent
Icon
art direction, side by side
A warmer background and a clearer hero character read better at thumbnail size in the feed.
+18%
install-intent
Subtitle
promise vs. proof
A cheeky hook outperformed the literal description on cold paid traffic.
+12%
install-intent
Description
the first line is the hook
Opening on action beat opening on setup, even in the lines shown above the fold.
+9%
install-intent
Category & metadata
positioning signals
Casual positioning matched the paid audience better than a puzzle label.
+6%
install-intent
Candy Chase icon
Candy Chase!
Catchem all
GetIn-App
Purchases
12K ratings
4.8
age
4+
chart
No.4
Chase the candy raccoons through the gingerbread market. Match, catch, and collect every sweet in this cozy puzzle. No timers, just a growing candy stash.
market-research preview · storelab.7play.ai
install-intent · this combo
8.0%control
click the tags · every element is a variable
05workflow

One link. Four steps.

01
Create variants
Duplicate your listing and change what you want to test: name, icon, screenshots, or all of it.
02
Publish one link
One public URL. StoreLab assigns each visitor a variant and keeps it stable across visits.
storelab.7play.ai/x/your-app
03
Point your ads at it
Use the link as your campaign destination on Meta, TikTok, or Google. UTM captured automatically.
?utm_source=meta&utm_campaign=q3
04
Read the decision
Install-intent per variant, lift vs control, and a plain-language call: ship it or keep testing.
06distribution

Share once. Track every channel.

One experiment, a tagged link for every platform. StoreLab bakes the UTM source into each one, keeps the same variant split everywhere, and breaks install-intent down by channel, so you know exactly where your budget converts.

pick a channel
Instagram
utm_source=instagram · utm_medium=paid_social
storelab.7play.ai/x/petalquest?utm_source=instagram&utm_medium=paid_social&utm_campaign=petalquest_q3
paste as your campaign destination
install-intent by channel
last 14 days
channel
install-intent
cvr
best
Every channel keeps the same variant split, so numbers stay comparable. Google skews to Variant A, paid social to Variant B, an insight you would miss with one blended number.
07the decision

Plain language, not p-values.

Every experiment resolves to a decision card a founder can act on. Behind it: honest measurement you can defend.

A quiet market-research disclosure on every public page
Neutral install buttons. No official badges, no store impersonation
We count install-intent clicks, never claimed installs
likely winnerName & icon test · Q3
Variant B, "Petal Quest: Garden Merge", is ahead by +34% on install-intent. Ship it, or let confidence tighten for two more days.
install-intent
10.7%
prob. best
87%
sessions
9,610
08pricing

A real test, free. No card, no sales call.

Your agent's access follows your plan. Upgrade mid-session and the new tools work on the next call, no reconnect.

Free
$0 /mo
1 experiment at a time
2 variants per experiment
runs 30 days or 1,000 sessions
agent access over MCP: browse projects, read results
Start free
recommended
Pro
$99 /mo
5 active experiments
25,000 sessions per month included
your agent does the work: create, upload, publish (asks you first)
UTM share links and per-channel results
decision reports, plus polls that ask converters why
Start Pro with a 7-day trial
Studio
$299 /mo
25 active experiments
250,000 sessions per month included
no experiment time limit
custom domains and PDF reports
agent automations, soon: weekly digest, auto-decide guardian
Start Studio with a 7-day trial
Geeklab charges €300/mo for one test a month. Your first StoreLab test costs nothing, and your agent can run it.
full limits inside the app · enterprise · talk to us
Your next launch decision could be data.
Create a project, build two variants, publish one link. Your first install-intent numbers can be in tonight's report.
storelab.7play.ai · part of the 7play.ai suite
StoreLab
store page experiments · lives in your agent over MCP · part of the 7play.ai suite
© 2026 7play.aistore preview pages are used for market research and may redirect to an official app store when available · not affiliated with apple or google