catalume

PUBLIC.SURFACE · AI SHOPPING VISIBILITY · LOSS FRAME

See why AI shoppers pick competitors — then fix the gap.

Every lost recommendation has a moment behind it. Catalume shows what the buyer retrieved, what it compared, and what to change next. Catalume observes how synthetic buyers find, understand, compare, and try to buy from your products.

No login needed · read-only public pages · deterministic preview

Open the merchant app

GROUND TRUTH FIRST · WHY THIS SIGNAL IS DIFFERENT

Rank is not the same as chosen.

A free public scan recomputes the same five listing signals Shopify scores — from your public pages, with no install — then reports the crawler posture and storefront agent reachability it observed directly. Those are read, not modeled. The replay comes after them.

Cross-store context

See the product that won an intent, the evidence it exposed, and the gap your storefront left open.

Cross-model discipline

Advice tuned for one AI can cost you the others. Catalume tests the same fix across model legs before it touches your store.

THE GROWTH LOOP

From a missing signal to a verified win.

A complete path that makes the machine's decision legible without turning a simulation into a guarantee.

See how it works
01 · DETECT

Detect

Run synthetic buyer missions against public storefront surfaces.

02 · DIAGNOSE

Diagnose

Name the failure: missing facts, ambiguity, retrieval, or checkout friction.

03 · FIX

Fix

Review an evidence-grounded artifact before anything changes.

04 · VERIFY

Verify

Re-run the failed mission and check the neighboring intents and AI legs.

05 · ATTRIBUTE

Attribute

Connect verified wins to Share of Recommendation and influenced revenue when available.

MISSION MATRIX · FIVE CONTEXTS

One intent, five visible legs.

Every stage carries its own fidelity badge. The retrieval row reads the same commerce substrate the consumer products read; the choice row is modeled and says so.

same intentGPTGEMCLDPERCP
retrieval
choice~~

REPLAY THEATER · SAMPLE

See the moment the choice changed.

A replay shows the buyer's question, what came back, what was compared, and where the path broke. It is the modeled stage, and it is labeled as one.

querybest pour-over set under $100
retrievalNorthstar Goods surfaced
choicecompetitor won on heat-retention proof

PUBLIC METHOD SAMPLE

A grade with context, not a badge with a claim.

The homepage illustration is a sample. Published grades are opt-in, time-bounded, and grounded in observed missions; they never rank product quality, revenue, or popularity.

Open leaderboard

Live public placements

Available only when the API confirms at least 20 current, consented, provenance-backed merchant grades.

See the current cohort

PLANS THAT SCALE WITH THE LOOP

Start with signal. Add coverage when it earns its keep.

Verification stays part of the product. Plans scale by coverage, mission volume, automation depth, and store count.

Compare plans
Scan$0free
Starter$29/ month
Pro$199/ month
Agency$499/ month

QUESTIONS, ANSWERED PLAINLY

Trust is part of the interface.

Read all FAQs
Is this my exact ChatGPT ranking?

No. Catalume combines ground-truth retrieval where a platform exposes the same commerce substrate with high-fidelity provider simulations and clearly labeled modeled stages. It shows evidence and fidelity badges instead of pretending every replay is the consumer app.

What does a public scan read?

Only public storefront pages and publicly reachable commerce surfaces. It does not ask for credentials, bypass a storefront password, or read shopper data. Detailed diagnosis and private store data require install or verified control.

What are the four readiness pillars?

Discoverability measures whether an agent can find the store and product. Comprehension checks whether it describes the facts correctly. Choosability checks comparison and recommendation. Transactability walks the path to checkout without completing payment.

Does Catalume automatically change my store?

No. Fixes are presented as reviewable artifacts. Nothing publishes without explicit approval, applied changes are revertible, and verification re-runs the failed mission plus collateral checks.

CATALUME / READ-ONLY BY DESIGNOpen the scanner