Skip to main content
Recommendation Audit

Why Don’t My Shopify Products Show Up in Perplexity Recommendations?

A readiness score cannot tell you whether an answer engine recommends your product. Here is a bounded way to measure the miss, fix one evidence gap, and rerun the same buyer question.

Colter Team·

A technically complete Shopify product page can still be absent from an AI shopping answer.

We tested 15 public Shopify products on August 13, 2026. Each product received one prewritten, unbranded buyer question in a logged-out Perplexity session. Thirteen products were absent from the rendered answer. None of the 15 merchant domains was cited.

This was a small, selected cohort and one observation per prompt. It is not a market-wide miss rate. It does show why a readiness score is only the start of the diagnosis.

What did the test actually measure?

The question for each product described a buyer job without naming the brand. For example:

What olive oil should I use to finish roasted vegetables?

Graza Drizzle’s public product page passed every deterministic product check in our audit. Perplexity did not mention Graza and did not cite its domain. The answer recommended choosing fresh extra-virgin olive oil by flavor strength.

That result says two separate things:

  • MEASURED: the product page exposed the product evidence our deterministic audit expected.
  • OBSERVED: this Perplexity answer did not mention or cite the product.

It does not prove why the product was omitted. The engine may have relied on third-party authority, comparison coverage, freshness, geography, its own retrieval system, or evidence we did not inspect.

Why didn’t a 100/A page earn a recommendation?

Structured data helps an answer engine identify a product. Recommendation questions also require comparative meaning.

For the olive-oil prompt, useful evidence would explain when Drizzle is the right choice, how it differs from an oil intended for cooking, and why a shopper should choose it over another finishing oil. That evidence may need to exist on the merchant site and in credible sources elsewhere on the web.

Our 15-product run showed the same gap in several categories. Eleven products that passed every deterministic product check were still omitted from their buyer answer. Two products were mentioned, but neither merchant domain received a citation.

Technical readiness and answer-engine placement are different measurements.

How should I audit a missing Shopify product?

Use one product, one buyer question, and one named engine. Keep them fixed throughout the test.

  1. Baseline the product evidence. Confirm that the engine can retrieve accurate product identity, price, availability, identifiers, policies, and the facts needed to answer the buyer question.
  2. Record the answer. Save the exact prompt, engine, timestamp, products mentioned, merchant citations, and a hash of the answer.
  3. Diagnose one evidence gap. Separate facts that are missing from the product page from comparison or authority evidence the answer appears to use.
  4. Make one bounded change. Add or correct truthful evidence. Do not rewrite the entire site and then guess which change mattered.
  5. Rerun the unchanged question. Record whether the product gained or lost a mention, gained a citation, moved in the comparison set, or remained absent.

A changed readiness score is useful evidence about the page. A changed answer is the result this test is designed to observe.

What would count as a successful fix?

Success depends on the buyer job. For the Graza example, a useful result could be an accurate mention of Drizzle for finishing roasted vegetables, a correct Drizzle-versus-Sizzle distinction, or a relevant Graza citation.

One successful rerun would still be an observation, not stable placement. Answer engines change their sources and outputs. Monitor the same small prompt set over time and report each engine separately.

FAQ

Does Product JSON-LD make a Shopify product appear in Perplexity?

Product JSON-LD makes facts easier to retrieve. It does not guarantee a mention, recommendation, ranking, or citation.

Start with the questions tied to a real product and buyer job. A useful page answers the question directly, exposes accurate machine-readable facts, and cites evidence. Volume without proof creates more pages to maintain.

Can the same audit be used for ChatGPT or Google AI Mode?

Yes, but each engine needs its own observation. Do not blend separate engines into one “AI visibility” score.

What does Colter Recommendation Audit do?

It records the product baseline, identifies one evidence gap, preserves a named-engine answer, and compares the unchanged rerun after a fix. Measured page evidence stays separate from observed mentions and citations.

Run a Recommendation Audit

Test scope and evidence

  • Date: August 13, 2026
  • Engine: Perplexity Search, logged out, public default experience
  • Cohort: 15 selected public Shopify products
  • Method: one prewritten unbranded category prompt per product; one observation per prompt
  • Observed: 13 products omitted; two mentioned; zero target merchant-domain citations
  • Limitation: this does not establish stable placement, cross-engine behavior, causation, traffic, conversion, or willingness to pay

Example public observations: Misen braiser, Graza Drizzle, Azuna, and PerTronix.

← All posts