What is llms.txt — and does your Shopify store need one?
llms.txt is a proposed convention: a Markdown file at the root of your site that gives language models a curated, token-friendly summary of what the site is and where its important content lives. Think of it as a sitemap written for a reader with a small attention budget. For a store, that typically means: what you sell, your key collections, shipping/returns basics, and links to the pages that matter.
It's cheap to add, harmless to have, and heavily marketed right now as the key to “AI visibility”. So let's be precise about what it does and doesn't do.
The honest part: adoption ≠ impact
The uncomfortable data point, reported in a 2026 Ahrefs analysis of sites that deployed the file: the overwhelming majority of llms.txt files get essentially zero crawls from AI bots. Major AI providers haven't committed to reading it, and none document it as a ranking or recommendation input. That doesn't make the file useless — conventions sometimes win later — but it does make one thing clear: llms.txt is a lottery ticket, not a strategy.
Any app whose headline feature is generating llms.txt is selling you the cheapest-to-build item on the checklist. Have the file — it costs nothing — but judge tools by what else they do.
What actually feeds AI answers about your products
- Your catalog data itself. Since Shopify syndicates eligible catalogs to AI shopping surfaces by default, the titles, descriptions, categories, identifiers and images in your product data are the primary input — presence is solved; representation is the game.
- Your product pages + structured data. Search-grounded assistants read your pages like a fast, literal visitor: schema.org Product markup, clean SEO titles and metas.
- Third-party corroboration. Reviews, comparisons and mentions elsewhere let a model verify that your product and brand are real and current.
So should you add it?
Yes — as a five-minute, zero-cost item, alongside an AI-aware sitemap and a machine-readable product feed. Preferd generates all three for your store automatically, and then treats them the way they deserve to be treated: as table stakes in the corner of the dashboard, not as the product. The features that earn their keep are the ones that change what an assistant can say about your products — audited, enriched product data and dated measurement that shows whether you're actually being recommended.