Machine Readability of DTC Product Pages Is Effectively Binary
PlatformDTC Research
Abstract
Twenty-three direct-to-consumer product pages scored for how well a machine reading raw HTML can understand them. Median score 100 out of 100, with effectively no middle of the distribution: pages either emit schema.org Product markup and land at the top, or emit none and sit in a narrow band below.
Key Findings
Median score 100 of 100, with no populated middle of the distribution
A quarter of the sample emits no Product markup at all, including widely recognised brands
Everything is measured against raw HTML with no JavaScript executed — a price that only materialises after client-side rendering is a price an agent cannot read
Adobe measured retail product pages at 66% machine readability in Q1 2026, the lowest of any major page type, while AI-sourced traffic to those sites grew 393% year over year
Corrections
An early version of the checker passed a 113-character bot-challenge page as healthy. A page that never reached real content scored as though it had; the checker now distinguishes a readable page from one that was never served.
The Rule That Makes the Measurement Worth Anything
Everything is measured against the raw HTML with no JavaScript executed. That is not a limitation of the method, it is the method. A checker that spins up a headless browser first will cheerfully report that a client-rendered page is fine, when an agent reading the same URL sees nothing. The test a merchant can run in a terminal — grepping the raw response for a price — is the same test the checker runs, which is also the honest way to verify anything reported here.
A Distribution With No Middle
The result is bimodal rather than continuous. Pages that emit correct structured product data — name, price, availability, identifiers — cluster at the top of the scale. Pages that emit none sit in a narrow band below. There is almost nothing between the two groups.
Why Bimodal Is Good News
A smooth distribution would imply a long programme of incremental tuning. A binary one implies a switch. Most of the distance between the bottom cluster and the top is covered by a single decision rather than a hundred small ones, which makes this one of the few page-quality problems with a short path to done.
Conclusion
Demand from agent-mediated shopping has arrived faster than the ability of product pages to serve it. The encouraging part of this measurement is its shape: the fix is discrete, verifiable from a terminal, and largely a matter of emitting structured data that the page already has in its database.
Published In Full
Read the complete study, method and datasetReferences
- [1]PlatformDTC Research (2026). We Scored 23 DTC Product Pages. Machine Readability Is Almost Binary. https://platformdtc.com/blog/product-page-readability-study
- [2]PlatformDTC Research (2026). What an AI Agent Sees on Your Product Page — and Two Ways We Got It Wrong. https://platformdtc.com/blog/what-an-agent-sees
- [3]Adobe (2026). Retail product page machine readability and AI-sourced traffic, Q1 2026.