Built into every store
SEO and AI search are built into every TeraCart store
Updated · Written by the TeraCart team
Every TeraCart store comes with the technical SEO that most platforms sell as apps: its own sitemap and robots rules, Product, Offer and Review structured data, canonical addresses on product and collection pages, an llms.txt file for AI assistants, and product feeds for Google, Meta and OpenAI. None of it is tied to a paid plan, so it is on for Free stores too. Stores moved over from Shopify keep their product, collection and page addresses, and get permanent redirects for blog posts and for the redirects they set up in Shopify.
What every store gets
These are served by the store itself, with nothing to install. The address is the store's own: its teracart.co subdomain, or its custom domain once you add one on Plus or higher.
| Feature | What it does | Where it is |
|---|---|---|
| Sitemap | Lists the home page, products, collections, pages, policy pages and blog posts, with last-updated dates. | /sitemap.xml |
| Robots rules | Lets search crawlers in, keeps /api/ out, and marks the cart, checkout, account and order pages noindex. | /robots.txt |
| Structured data | Product with an Offer for every variant, ratings and reviews, breadcrumbs, answered questions, collections and blog posts. | In each page |
| Canonical addresses | Product and collection pages name one address, so variant and page-number versions do not compete. | In the page head |
| llms.txt | A plain-text map of the store for AI assistants. | /llms.txt |
| Product feeds | Google Merchant, Google product ratings, Meta and OpenAI. | /products-feed/google.xml and others |
| Blog | A blog with its own post markup, listed in the sitemap and llms.txt. | /blog |
A store that is not open yet is kept out of search on purpose. It shows a coming-soon page marked noindex, its robots.txt blocks crawlers, its sitemap is empty and its feeds answer with an error until you open it.
Structured data: what the markup says
Search engines and AI shopping tools read structured data to understand a product without guessing. Each product page carries a Product with its name, description, images, SKU, brand and category. It has one Offer for every variant, with the price, the currency, whether it is in stock and its condition.
When a variant has a compare-at price above its price, the Offer also says the higher figure is the struck-through price. When reviews are switched on and shown on the page, the product adds an aggregate rating and the reviews on the page, at most ten, with each reviewer's name, star rating, text and date. Answered customer questions are marked up as a question page, and every product and collection has a breadcrumb trail.
Collections are marked as collection pages with a list of the products on the page. Blog posts carry headline, author, dates, image and publisher. The home page names the store as an organisation and describes its search box.
llms.txt, and what it can and cannot do
Each store's llms.txt is built from its live catalogue and cached for about fifteen minutes. It names the store, links its sitemap and, when feeds are on, its Google and OpenAI feeds, and counts its products, collections and reviews. It lists up to 30 collections, up to 50 products ordered by sales with their details, and up to 20 pages and 20 blog posts. It ends with a short note telling assistants not to invent discounts or reviews.
The store's robots.txt also speaks to AI crawlers. It lets OpenAI's search crawler, OAI-SearchBot, in, and turns away GPTBot, which OpenAI uses to train models. Those two rules are set for the whole platform, not store by store.
No AI search product has promised to read llms.txt. Treat it as a cheap, accurate index for any assistant that does, and rely on the sitemap, structured data and feeds for the crawlers that matter today.
Product feeds for Google, Meta and OpenAI
- Google Merchant Center: an RSS product feed at /products-feed/google.xml, with a price, sale price, availability, brand, images and shipping line for every variant.
- Google product ratings: a reviews feed at /products-feed/google-reviews.xml, for stores with reviews switched on.
- Meta (Facebook and Instagram): a catalogue feed as XML, CSV or TSV. Once you connect Meta and choose a business, TeraCart creates the catalogue and schedules a daily pull at 3 am India time.
- OpenAI: a JSON feed at /products-feed/openai.json, in TeraCart's own layout, with canonical links, prices, shipping and return policy. OpenAI runs its own application process for merchants, which you apply for separately.
Feeds are off until you switch them on. The switch is Product feeds, in Settings, then Tracking and analytics, inside the Advanced manual fallback section. Connecting Meta, or choosing a Merchant account in the Google connection, turns them on too. For Google you add the feed address in Merchant Center yourself; TeraCart does not upload the products for you.
Titles, descriptions, alt text and the blog
Each product has a meta title of up to 70 characters and a description of up to 160, with a preview of how they look in Google. The form stops you at those lengths. If you leave them empty, TeraCart builds a title and description from the product's own details. Every product image has an alt text field, and the SEO page in Settings counts the images missing one on your active products.
Collections take their own titles, descriptions, intro text and FAQs, and pages and blog posts have SEO fields too. The blog is part of every plan, Free included.
Moving from Shopify: redirects
Shopify and TeraCart use the same address shapes for products, collections and pages, and the import keeps each handle, so most links work without any redirect. The import also brings across the redirects you set up in Shopify that point to a page on your own store, and it redirects blog post addresses to TeraCart's blog. Shopify's /collections/name/products/name form has no matching page, so a link written that way will not load.
Redirects match the exact path and keep the query string. There is no screen yet for adding redirects by hand. The migration guide lists everything else that moves.
What is not included yet
- The sitemap is one file. That suits a catalogue well under Google's 50,000-URL limit per sitemap, and a store beyond that needs splitting.
- Store pages do not carry FAQ markup, and there is no multi-language or hreflang support.
- There is no RSS feed for the blog.
- Canonical tags are on product and collection pages, not on every page type.
- Google's product category is only filled in for a few product types, so set categories in Merchant Center if you need them.
Questions people ask
Do I need an SEO app on TeraCart?
For the technical basics, no. The sitemap, structured data, canonical addresses, feeds and llms.txt are part of every store. You still need good product pages and content, because no setting writes those for you.
Is built-in SEO available on the Free plan?
Yes. None of these features depends on the plan. A custom domain needs Plus or higher, so Free stores use their teracart.co address.
Will llms.txt make ChatGPT recommend my store?
No. It gives an assistant an accurate index of your catalogue, but no AI search product has said it follows llms.txt. Your structured data, your feeds and the quality of your pages do most of the work.
How do I switch the product feeds on?
Open Settings, then Tracking and analytics, and tick Product feeds in the Advanced manual fallback section, or connect Meta, or choose a Merchant account in the Google connection. The feeds answer with an error until one of those is done.
Will my Shopify links keep working after I move?
Links with the same product, collection and page handles work as they were, and blog posts and your own Shopify redirects are redirected. Links in the /collections/name/products/name form will not load. Check a sample of your best-ranking pages after the move to be sure.
Does TeraCart submit my site to Google?
No. You verify the store in Google Search Console and submit its sitemap yourself. TeraCart supports Google and Facebook domain verification, but it has no Bing verification field yet.