samplepets.example

The strongest true thing here is real: samplepets.example holds five top-10 organic rankings for genuinely valuable local searches, and Google's own real-visitor speed data passes on every page tested.

None of that visibility survives contact with where local pet owners increasingly start looking. A real ChatGPT test named Sample Pet Co. in only 2 of 15 priority questions, and the Google local map pack showed a Sample Pet Co. brand in just 1 of 14 packs that appeared.

1. 01Six stores, one brand, no shared identity online. Every store still runs its own separate domain, has no page of its own on samplepets.example, and nothing links them together in a way search engines or AI models can read.
2. 021,226 category pages say nothing. The pages built to win searches like "brand a" and "cat litter [city]" carry no description and no text at all, on exactly the queries where a competitor already outranks the site with a weaker page.

Sample Pet Co. is genuinely ahead of the market on one thing: it is one of very few local pet retailers with a working AI-shopping-agent protocol built in. A single-location competitor, Competitor A, is winning the map pack and the AI answers that infrastructure was supposed to earn instead. Closing the identity and content gaps below is what keeps that advantage from being wasted.

Sample Pet Co. operates as the shared e-commerce storefront for six neighbourhood pet shops across the city and its neighborhoods: Store One, Store Six, Store Two, Store Five, Store Three and Store Four. The catalogue runs to roughly 3,500 products and 1,200 product categories on Shopify, spanning premium brands such as Brand A, Brand B, Brand C and Brand D, with free same-day delivery across the city on orders over $69.99.

Competitors were selected algorithmically from real ranking overlap, then checked by hand: national-chain-one.example and national-chain-two.example are both genuine national pet-retail chains, not platform or social accounts that sometimes surface as false positives in this kind of analysis. A third competitor, Competitor A, surfaced repeatedly across local search and AI testing below and is referenced throughout, even though it fell outside the original algorithmic set, because the live data made its relevance impossible to ignore.

The site has exactly nine pages outside the product catalogue, six static pages and three blog articles, carrying 597 words between them describing what the business is. The About page opens directly into customer reviews without a single sentence introducing the company. The FAQ page, by contrast, is genuinely strong: 1,554 words of specific, useful policy detail, the model the rest of the site should follow.

The blog is the clearest sign of unfinished work. All three articles are good, correctly structured, the city-specific, opinionated where it helps, 980 to 1,330 words each. All three were published the same day, March 24, 2026, and nothing has followed in five months. Neither of the two articles targeting a specific search term ranks in the top 100 for its own title phrase.

The 1,226 product-category pages are the largest content gap on the site. Every brand collection checked by hand, Brand A, Brand E, Brand F, Brand C, Brand G, Brand D, Brand I, Brand B, has no text beyond the product grid: a bare title, no meta description, nothing a search engine or AI assistant can quote.

The clearest actionable finding in this audit: 170 keyword gaps where a curated competitor ranks in the top 10 and Sample Pet Co. ranks below position 20 or not at all, most against product categories the site already sells.

The difficulty column is what matters most: several of these carry a difficulty score under 10 out of 100, meaning very little competition to outrank, and the site still sits 30 to 70 places behind. This lines up exactly with the content gap above: these are category and brand pages that exist, are indexed, and say nothing worth ranking for.

Backlink profile, for scale: samplepets.example has 162 referring domains at a spam-risk score of 23 of 100 (moderate, worth monitoring). National Chain One has 3,018 referring domains at spam score 10; National Chain Two has 4,569 at spam score 16. This is a longer project than the content fixes above, and explains part of why national chains outrank a well-run local business on head terms like "pet store."

"Competitor D — Best bet for speed. Same-day delivery across the metro area, generally free over $39, with a 2pm local cutoff. Sample Pet Co. — Same-day delivery, with free delivery over $69.99 and orders before 12 PM eligible for same-day service."ChatGPT, real web search, asked "same day pet food delivery [city]"

Asked "brand a dog food [city]," ChatGPT named six local retailers including Store Two, Store Three and Store Four, three of Sample Pet Co.'s own stores, as three separate, unrelated businesses. Sample Pet Co. was not named.ChatGPT, real web search

Competitive AI landscape

Sample Pet Co. scores well ahead of the market here. Its robots.txt explicitly welcomes AI crawlers, links a plain-language agents.md file, and implements the Universal Commerce Protocol (UCP), a live endpoint letting AI shopping agents search the catalogue, build a cart and start checkout (a human still approves any actual payment). No competitor sampled in this audit has this. This is our own structural readiness read, checked against real AI behaviour above, not blended into it: it removes crawler access as an explanation for the low AI-mention rate. The gap is what AI systems find once they arrive, covered in Content Quality and the appendix.

The foundations are sound: HTTPS, HSTS, and every domain variant correctly redirects to one canonical address; a deliberately non-existent URL returns a proper 404. Category pages, the 1,226 pages that should be the site's strongest search asset, fall short at the page-type level.

Page Speed & Images

The number that matters, real visitor data from Google's own measurement tool, passes everywhere tested, comfortably below the 2,500 ms "good" threshold on two of three pages and right at it on the homepage. A separate lab-only diagnostic flags heavy JavaScript and several oversized promotional images (one banner alone was 6.27 MB uncompressed) as the reason a synthetic lab score reads low even though real visitors are not experiencing that slowness today. Shopify's image delivery already auto-converts images to modern, much smaller formats for real browsers, so no image-format work is needed here.

Across roughly 1,448 tracked search terms, the site holds 102 top-10 positions (7%), a further 588 in positions 11–30, and the remaining 758 below position 30. In the most recent measurement period, 778 keywords were newly gained and 529 were lost, real churn worth watching rather than a one-time snapshot. Search intent on the 15 priority terms audited skews transactional and navigational (9 of 15), meaning most are genuine buying searches, which raises the stakes on the content and local-visibility gaps above.

The sitemap itself is valid, complete for what it declares, and kept current automatically. The gap is what it leaves out: 169 pages the rendered crawl reached that appear in no sitemap, almost entirely a /collections/<category>/<brand-or-tag> pattern, for example /collections/cat-food/brand a, /collections/cat-food/brand b and roughly 140 more of the same shape. These pages are live, return a normal 200 response, are not blocked from search engines, and some already rank (/collections/cat/brand-h-fresh holds position 6 for "brand h cat food"). None are managed through the sitemap meant to be the map of the site.

Everything above is something we measured on samplepets.example. This section is not — it is the standard platform layer for a local multi-location retailer. The top four rows are not generic, though: they are platforms that showed up in this audit's own search and AI data as places where your customers are landing instead of on your site.

On the Status column. This audit measured your backlink profile in summary (162 referring domains, spam-risk score 23) but did not pull the list of which domains those are, so presence on a directory cannot be confirmed from the audit data. The two confirmed rows are confirmed from links on your own homepage. Everything else reads "Worth checking" rather than a claim we cannot support.

Link building. The highest-value opportunities for this business are not in the table above — they are the 170 keyword gaps and the competitor referring-domain comparison already in Competitor Analysis. Suppliers and premium brands you stock (Brand A, Brand B, Brand C, Brand E) commonly run "where to buy" or stockist pages; those are relevant, followed links from domains already ranking for your target terms, and the relationship exists. We recommend against paid link placements, guest-post networks and directory-submission packages — your spam-risk score is already 23 of 100 against competitors at 10 and 16.