Docs / for Shopify / Troubleshooting

AEO Ultimate for Shopify

When something looks wrong

Grouped by what you are seeing. The messages quoted here are the real ones from the app, so if you have an exact error in front of you, search this page for it.

IndexNow rejected my URLs

IndexNow tells you why by status code, and each one has a different fix. The app surfaces the code rather than hiding it, because the difference matters.

403 is the common one. Use the Verify action first — it fetches your key file the way the engine does. A password-protected or staging-locked storefront fails here, because the search engine cannot read the file either.

422 means a host mismatch. The key file lives on one host and you submitted URLs on another — usually a myshopify.com URL submitted against a custom-domain key file, or the reverse. Submit URLs on the same host the key file is served from.

The app cannot read or edit robots.txt

On Shopify, robots.txt is generated from a template in your published theme. If your store has no published theme — every theme is a draft, or you are mid-migration — there is nothing to edit.

Publish a theme, then retry. If you see “Couldn’t read the theme’s robots.txt template”, the theme has no robots.txt.liquid yet — Shopify creates it on first customisation, and the app can add it for you.

The schema audit says my page is not a full page

People usually paste what they see in Inspect Element rather than View Page Source. Those differ: Inspect shows the DOM after JavaScript has run; View Source shows what was actually delivered. Schema has to be in the delivered HTML, because that is all a non-JavaScript crawler ever sees.

Use Ctrl+U (or ⌘+⌥+U on a Mac), select everything, paste it all.

Other audit responses and what they mean:

• “That page returned 404” — the URL is wrong, or the product is unpublished.
• “That page redirected to [address], a private address this app won’t read from” — the URL resolved to a local or internal network address. The app refuses those deliberately; it will not be used to probe private networks.

Image sourcing cannot find a manufacturer image

The app verifies a brand’s official website through Wikidata before fetching anything, rather than guessing at a domain. That is the safeguard, and it is also why it sometimes declines:

“Couldn’t reach Wikidata to check [brand] just now, so we don’t know whether it has a verified official site — this is our lookup failing.” That wording is deliberate: it is telling you the lookup failed, not that the brand is unverified. Retry shortly.

If the brand genuinely has no verified official site in Wikidata, paste the manufacturer’s address yourself — brand.com or the full product page URL. If you get “doesn’t look like a website address”, you have probably pasted a search result or a marketplace listing rather than the manufacturer’s own site.

“That doesn’t look like a full address — copy it with right-click → Copy image address” means you pasted the page the image sits on, not the image itself. Right-click the image directly.

“That search is no longer open”

Full message: “That search is no longer open. Start a new one — nothing was lost.”

An image search session expired, usually because the tab sat idle. The reassurance is accurate — no product data was changed. Start a new search.

Search Console or redirect tools failed

“Search Console fetch failed” is nearly always the connection rather than the data: an expired OAuth grant, or a property you no longer have access to. Reconnect the account and confirm the property is still verified to you.

“Could not read that batch of URLs” or “Could not read the redirects to create” means the pasted input could not be parsed — usually a spreadsheet copied with extra columns. One URL per line, nothing else.

Nothing here matches my problem

Send the exact message and what you were doing. Questions that reach me become documentation, so it genuinely helps the next person.