Skip to content

Store locator SEO: give every location a crawlable page

Store locator SEO: crawlers can't be relied on to type postcodes, so give each branch its own URL, link it from a static index, and add markup and a sitemap entry.

By Crawlito · 7 min read

In short

  • Google's lazy-loading guide states that Google Search does not interact with your page, so branches that appear only after someone types a postcode and submits a search cannot be reached through the locator.
  • Google can generally crawl a link only if it is an a element with an href, so a locator that draws pins from an API response leaves crawlers with no branch URLs to follow.
  • The fix is a plain HTML index of regions that links to one URL per location, with address, hours and phone in the server-rendered HTML of each page.
  • Google's LocalBusiness documentation defines url as the fully qualified URL of the specific business location, and it must be a working link.
  • A single sitemap can list at most 50,000 URLs or 50 MB uncompressed, so large dealer networks split location URLs across several files under a sitemap index.
On this page

Store locator SEO comes down to one rule: every location needs its own URL that a crawler can reach by following ordinary links. A locator that waits for a visitor to type a postcode, then fetches results from an API and draws pins on a map, gives crawlers nothing to follow, because you cannot rely on a crawler to type into a search box or submit a form. Keep the widget for people, and add a static index of regions that links to one page per branch with the address, hours and phone in the HTML.

This post covers the locator pattern specifically. For the markup itself, see the LocalBusiness schema guide, and for the general rule about click handlers and real href values, see JavaScript links and crawler discovery.

Why a search-box locator hides your branches

A typical locator page looks like this in the server's response:

html
<h1>Find a store</h1>
<form id="locator">
  <input name="postcode" placeholder="Enter your postcode">
  <button type="submit">Search</button>
</form>
<div id="map"></div>
<ul id="results"></ul>
<script src="/js/locator.js"></script>

Everything a customer cares about arrives later. The script listens for the submit, calls something like /api/stores?lat=...&lng=...&radius=25, and writes the results into #results and the map. Three facts from Google's own documentation explain why that leaves branches undiscovered:

  1. Google does not interact with the page. Google's lazy-loading guide says plainly that "Google Search does not interact with your page", and warns against loading content only on user actions such as scrolling or clicking. Typing a postcode and pressing Search is a user action.
  2. Links need to be <a href> elements. Google's crawlable links guide says Google can generally only crawl a link if it is an <a> element, and cannot reliably extract URLs from <a> elements without an href. Map pins, list items with click handlers and API responses are not links.
  3. Not every crawler runs JavaScript. Google's JavaScript SEO basics notes that not all bots can run JavaScript, and calls server-side rendering or pre-rendering "a great idea" for that reason. The major AI crawlers covered in do AI crawlers run JavaScript, such as GPTBot, ClaudeBot and PerplexityBot, did not render it in Vercel and MERJ's December 2024 analysis, so for them the empty <ul id="results"> is the whole page.

The result is a site where the locator works perfectly for people and the crawler sees one page with a form on it. Even if a branch page exists somewhere, nothing links to it except a script.

The "detail panel" variant

Some locators do have a page per branch, but only as a panel that opens over the map, with a URL like /stores#branch-1427 or a route that only exists inside the app. A fragment is not a separate page: Google's JavaScript SEO basics tells sites not to use fragments to load different page content and to use the History API instead. A route that returns the generic locator shell when requested directly is not a branch page either. Test each branch URL in a fresh session, as described below.

The crawlable structure: index, region, location

The fix keeps the widget and adds a plain link structure underneath it. Every branch gets reached in two or three clicks from the locator page, without any typing:

LevelExample URLWhat the HTML contains
Locator/stores/The search widget, plus links to every region
Region/stores/scotland/ or /stores/ca/A heading, a short list of branches with name, town and a link to each
Location/stores/scotland/glasgow-buchanan-street/Name, full address, phone, opening hours, services, directions, LocalBusiness markup

Choose region levels that match how people describe places in your market: states, provinces, counties or cities. A network of 40 branches may need only the locator and the location pages. A dealer network of several thousand usually needs a country, region and city layer so that no index page carries an unmanageable list.

The region links belong in the server-rendered HTML of the locator page, below or beside the widget:

html
<section aria-labelledby="all-regions">
  <h2 id="all-regions">All stores by region</h2>
  <ul>
    <li><a href="/stores/scotland/">Scotland</a> (14 stores)</li>
    <li><a href="/stores/north-west-england/">North West England</a> (22 stores)</li>
    <li><a href="/stores/wales/">Wales</a> (9 stores)</li>
  </ul>
</section>

A region page then links each branch with real anchors:

html
<h1>Stores in Scotland</h1>
<ul>
  <li>
    <a href="/stores/scotland/glasgow-buchanan-street/">Glasgow, Buchanan Street</a>
    <p>145 Buchanan Street, Glasgow G1 2JX</p>
  </li>
  <li>
    <a href="/stores/scotland/edinburgh-princes-street/">Edinburgh, Princes Street</a>
    <p>88 Princes Street, Edinburgh EH2 2ER</p>
  </li>
</ul>

The widget can still enhance these pages. A map can be drawn from the same data, and the search box can jump straight to a branch. What matters is that the links exist in the response before any script runs. The search-results URLs the widget produces, such as /stores?postcode=G1, do not need to be public landing pages: the internal search results guide covers how to treat them.

What each location page needs in its HTML

A location page is useful to a person and a crawler for the same reason: it answers the practical questions about one place. Put these in the HTML text, not in an image, an iframe or a map tooltip:

  • Name as customers know it, with the branch qualifier: "Northfield Hardware, Glasgow Buchanan Street".
  • Full postal address in one block of text.
  • Phone number for that branch, not only the head office.
  • Opening hours as a readable table, including how holiday changes are shown. Keeping those in step with your Business Profile is covered in holiday hours on location pages.
  • Services or departments at that branch, since branches often differ: click and collect, a trade counter, servicing, parking.
  • Links back to the region page and to nearby branches, so a crawler that lands on one location can reach the others.

A minimal server-rendered location block:

html
<h1>Northfield Hardware, Glasgow Buchanan Street</h1>
<address>
  145 Buchanan Street<br>
  Glasgow G1 2JX<br>
  <a href="tel:+441410000000">0141 000 0000</a>
</address>
<h2>Opening hours</h2>
<table>
  <tr><th scope="row">Monday to Friday</th><td>08:00 to 18:00</td></tr>
  <tr><th scope="row">Saturday</th><td>09:00 to 17:00</td></tr>
  <tr><th scope="row">Sunday</th><td>Closed</td></tr>
</table>
<p><a href="/stores/scotland/">All stores in Scotland</a></p>

Avoid thin, template-only pages. If every branch page differs only by its address, it still helps a customer find the right door, but write what is genuinely specific to each branch where you have it. Businesses that serve areas rather than run storefronts should describe coverage instead of inventing branch addresses.

LocalBusiness markup and sitemap entries per location

Google's LocalBusiness documentation says to define each location as a LocalBusiness type, and defines url as "the fully-qualified URL of the specific business location", adding that it must be a working link. That only works when each branch has a URL of its own, which is what the structure above gives you. Google also asks for the most specific LocalBusiness subtype possible, so pick one from schema.org/LocalBusiness, such as Store or one of its children like HardwareStore, and point url at the location page itself:

json
{
  "@context": "https://schema.org",
  "@type": "HardwareStore",
  "name": "Northfield Hardware, Glasgow Buchanan Street",
  "url": "https://www.example.com/stores/scotland/glasgow-buchanan-street/",
  "telephone": "+44 141 000 0000",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "145 Buchanan Street",
    "addressLocality": "Glasgow",
    "postalCode": "G1 2JX",
    "addressCountry": "GB"
  },
  "parentOrganization": {
    "@type": "Organization",
    "name": "Northfield Hardware",
    "url": "https://www.example.com/"
  }
}

schema.org's parentOrganization links the branch to the brand, and schema.org lists it as superseding the older branchOf. Google's LocalBusiness page does not mention either property, so treat it as plain schema.org vocabulary rather than something Google documents for local results. Generate the markup from the same store record that renders the visible page, so the address and phone in JSON-LD cannot drift from the text.

Then list every location page in an XML sitemap. Google's sitemap guide asks for fully qualified, absolute URLs and limits one file to 50,000 URLs or 50 MB uncompressed, so large networks split by country or region and can submit a single sitemap index file that lists the parts:

xml
<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
  <url>
    <loc>https://www.example.com/stores/scotland/glasgow-buchanan-street/</loc>
    <lastmod>2026-11-02</lastmod>
  </url>
</urlset>

Google says it uses lastmod if it is consistently and verifiably accurate, so set it from the date the store record last changed, not the build date. The sitemap is a discovery aid on top of the links, not a replacement for them: the XML sitemaps guide covers submission and what not to list.

How to test your locator with curl

You can check the whole chain from a terminal. Each step looks at the raw HTML a crawler that does not run JavaScript would receive.

  1. Does the locator page link to regions? Count real anchors that point into the store section:
bash
curl -s https://www.example.com/stores/ \
  | grep -oE 'href="/stores/[^"]+"' | sort -u | head -20

If this prints nothing, the only way in is the search box.

  1. Does a branch page work on a direct request? Fetch one location URL in isolation and look for its address:
bash
curl -s -o /tmp/branch.html -w 'status=%{http_code} final=%{url_effective}\n' -L \
  https://www.example.com/stores/scotland/glasgow-buchanan-street/
grep -c '145 Buchanan Street' /tmp/branch.html

You want a 200, the same final URL, and a count above zero. A 200 with a count of zero usually means the server returns an empty app shell and fills the address in with JavaScript.

  1. Is the markup in the response? Check for the JSON-LD block and that its url matches the page:
bash
grep -o '"url": *"[^"]*"' /tmp/branch.html

The first match should be the branch URL; a second one is the brand's parentOrganization URL. If nothing prints, the markup is either missing or injected by a script.

  1. Is every branch in the sitemap? Compare the number of <loc> entries in the store sitemap with the number of branches in your store database. A mismatch usually means closed branches still listed or new ones missing.

Run the same requests with an AI crawler's user agent too (curl -A), since a firewall can treat those differently from a plain request.

Frequently asked questions

Not by searching. Google's documentation states that Google Search does not interact with your page, so results that appear only after typing and submitting a form are not reached unless the branch URLs are also linked or listed in a sitemap.

Do I need to remove the map widget?

No. Keep it for visitors. Add plain HTML links to regions and branches alongside it, so the widget becomes an enhancement rather than the only route to each location.

Should postcode search results be indexed?

Usually not. The maintained region and location pages are the public landing pages; the result URLs the widget generates for each query are a tool for visitors, not pages to promote.

Is one page listing every store enough?

For a small network it can be, as long as each store still links to its own page with full details. One long page cannot carry a separate URL, hours and LocalBusiness markup for each branch the way a page per location does.

Check a location page

Once the index and branch pages are live, run the free check on one location URL to see the HTML and markup a crawler receives without running your locator script. For a network with hundreds of branches, an audit reviews the locator, region and location templates together and delivers prioritized fixes with evidence.

Filed under Rendering and content. Topics: local-seo, javascript, internal-links, structured-data