Start free trial
SKU-level AI visibility

The unit of revenue is the SKU. That's how we track it.

eCommerce Insights tracks every SKU in your Shopify catalog across ChatGPT, Perplexity, Google AI Overviews, Gemini, Claude, and Copilot. Variant-aware. Engine-by-engine. Weekly or daily refresh. Brand-level share of voice doesn't tell you which product is losing the sale.

Start free trial Try the free grader

Variant-aware · Six AI engines · Weekly or daily refresh

SKU · MERINO-BL-MD-FG
74 / 100
ChatGPT cited Perplexity cited AIO missing Gemini missing
+14 if you ship this week's diff

Brand-level AI tracking tells you the brand is doing fine. eCommerce Insights tells you which SKU isn't. That's the difference. Every variant gets its own score, its own citation record, and its own diff queue.

Why SKU-level

Four reasons to track the variant, not the brand.

01

Brand mentions don't convert. SKUs do.

A ChatGPT answer that namechecks your brand is a soft signal. A ChatGPT answer that recommends a specific SKU, with a price, an image, and a link, is the sale. Brand-level trackers stop at the first layer. Revenue lives at the second.

02

PDP fixes happen per SKU.

A merchandiser doesn't "improve the brand's AI readiness." They improve specific PDPs, one at a time, in a queue. If your tracking data doesn't resolve to a specific PDP, it can't tell a merchandiser what to do with their Tuesday morning.

03

AI engines cite specific products.

ChatGPT Shopping, Perplexity Shopping, and Google AI Overviews cite product-level URLs with product-level data — not brand homepages. Brand-level tracking misses the granularity the engines themselves work at.

04

Variants matter. So do seasonality and stock.

The medium forest-green merino base layer ranks; the large in the same color doesn't. The seasonal variant was cited for two weeks and then dropped off. The out-of-stock SKU is still showing in ChatGPT's answer. eCommerce Insights sees all of it.

How it works

Four steps from catalog connect to scored queue.

A HowTo schema version of this is embedded in the page's JSON-LD so AI engines can ingest the steps directly.

  1. Step 01

    Connect the catalog

    Install eCommerce Insights from the Shopify admin. The app resolves every variant as a distinct SKU and maps collections, metafields, and tags into the scoring model.

  2. Step 02

    Build the query bank

    eCommerce Insights generates a buyer-shaped query bank from your collections and best-sellers, rotated each scan. Category-tuned by default; editable if you have opinions.

  3. Step 03

    Scan six engines

    Every scan runs across ChatGPT, Perplexity, Google AI Overviews, Gemini, Claude, and Copilot. Each SKU gets a citation record per engine per query.

  4. Step 04

    Score and surface deltas

    Every SKU gets a 0–100 score plus week-over-week deltas per engine. Sort by revenue impact to triage the catalog top-down.

What eCommerce Insights scores per SKU

Five signals, one composite.

Every SKU's 0–100 eCommerce Insights Score is composed of five sub-scores, each weighted equally. Every sub-score is visible so a merchandiser can see why a SKU scores the way it does.

Signal 01

Structured data

Product schema coverage across name, brand, GTIN, SKU, description, image, offers, material, color, size.

Signal 02

Citation surface

How often the SKU is cited, relative to peers and competitors, on the query bank eCommerce Insights scans.

Signal 03

Entity clarity

How well the PDP disambiguates the product: canonical URL, brand consistency, title specificity.

Signal 04

Answer coverage

Whether the PDP body answers the top buyer questions for the category.

Signal 05

Review signal

Volume, recency, and structured-data surfacing of user reviews.

Composite

eCommerce Insights Score

The 0–100 composite. Read the methodology →

Example report row

What a single SKU looks like in eCommerce Insights.

SKU · MERINO-BL-MD-FG · Brand X Merino Base Layer, Men's, Medium, Forest Green
74 / 100
Structured data
18 / 20
Citation surface
12 / 20
Entity clarity
16 / 20
Answer coverage
14 / 20
Review signal
14 / 20
Engine citations · this week
ChatGPT cited Perplexity cited AI Overviews missing Gemini missing Claude cited Copilot cited
Co-cited with: rei.com, outdoorgearlab.com, reddit.com/r/ultralight
→ 1 diff ready to approve (+14 projected)

Illustrative record for a hypothetical SKU. Real scores are drawn from live scans.

Every SKU, every engine, every week.

Related

Keep reading.

Guide

SKU-level AEO

The pillar guide to optimizing for Answer Engine Optimization at the variant level.

 Guide

Product AI visibility

eCommerce Insights's category-defining wedge term and the discipline behind it.

 Feature

PDP optimization

The other half of the product: turning scores into shippable PDP diffs.

How SKU-level tracking compares to Profound's brand-level approach

Further reading: schema.org/Product — the canonical structured-data vocabulary eCommerce Insights scores PDPs against.

Ask AI about SKU-level tracking

Have your favorite AI engine summarize this page for your specific use case.

Frequently asked questions

What counts as a SKU for pricing?
A SKU is a tracked product variant in your catalog. A base product with five sizes across three colors counts as 15 SKUs. Plan caps are set against this variant-level count. Archived and draft products are excluded automatically; out-of-stock SKUs still count and are tracked with an availability flag.
Do you track variant-level or parent-product level?
Variant-level by default, with a parent-product roll-up view for triage. AI engines frequently cite specific variants — a particular size, color, or material — so tracking at the variant level is the only way to see whether the citation resolved to a SKU you actually sell in stock.
How often is tracking refreshed?
Weekly on the Seed plan and daily on the Shelf and Warehouse plans. Refreshes run overnight in US-Eastern by default; Warehouse-tier customers can request a custom schedule. A manual rescan is available on every plan for when a site change shipped and you want the score refresh immediately.
Can I add non-Shopify SKUs?
Yes. eCommerce Insights accepts a URL-and-feed ingestion for non-Shopify PDPs, and for BigCommerce, WooCommerce, and headless stacks with a catalog feed. Metafield write-back is limited to Shopify; tracking, scoring, and CSV-export of recommendations work on any platform.
What if I have 50,000 SKUs?
The Warehouse tier handles catalogs above 10,000 SKUs. For very large catalogs, eCommerce Insights batches scans by collection and lets you prioritize tracking cadence per collection (e.g., daily on the 2,000 revenue-weighted SKUs, weekly on the long tail). Contact sales for a quote against your exact catalog size.

Every SKU. Every engine. Every week.

14-day free trial. No credit card. Connect your Shopify store in under five minutes.

Start free trial Try the free grader