Link partners and the badge program

The version we did not build

The original plan was ordinary enough. A business links to its LocalsKnow listing, an automated check confirms the link is live, and the outbound link on the listing flips to dofollow in return. Every directory in the country has done some version of it.

It went in the bin, because Google’s spam policies describe that exact arrangement twice.

Excessive link exchanges ("link to me and I’ll link to you") or partner pages exclusively for the sake of cross linking.

Requiring a link as part of a terms of service, contract, or similar arrangement without allowing a third party content owner the choice of qualifying the outbound link.

An automated system that grants a dofollow link on condition of receiving one is the textbook example of the first, and gating it behind our terms would have been the second. It also leaves a footprint that is not subtle: pairs of links appearing in both directions within minutes of each other, at scale, from one directory. Detecting that pattern is precisely what SpamBrain is for. At five hundred partners it would be visible from space.

Both sides lose in that trade. The partner’s link gets discounted and our outbound profile looks engineered, which is a bad position to be in when the whole product depends on being the directory that can be trusted.

What we built instead

The reward is decoupled from the reciprocity. Same benefit for the business, same reason to complete a profile, no risk to either party. Dofollow is earned by verification.

Dofollow switches on at a quality score of 70 on a verified listing. That is an editorial decision in the ordinary sense of the word: we are vouching for businesses whose credentials we actually checked, and declining to vouch for the ones we have not. It is defensible in a reconsideration request, it survives a manual review, and it gives owners a reason to finish their profile, which is what we wanted from the reciprocal scheme in the first place.

Sample records never get a dofollow link, whatever their state. The claim page covers how to move up the table.

The badge program

The badge runs alongside all of this and is deliberately not a PageRank play. It is a trust and traffic feature. A visitor on your own website sees that a licence was checked, and the ones who want the evidence click through to the listing where the register link and the check date are published.

• Verified listings get an embed snippet in the dashboard.
• The badge renders the live verification status and rating from our API.
• The link in the snippet carries rel="nofollow" by default, and this is the note explaining why.
• A weekly crawl confirms the badge is still live and shows the result in your dashboard. Nothing depends on it.

Here is the snippet, exactly as it ships. The rel is not hidden in a redirect.

<!-- LocalsKnow verified badge. Paste anywhere on your site. -->
<a href="https://localsknow.com.au/nsw/albury/plumbers/riverside-plumbing-co/"
   rel="nofollow noopener"
   title="Licence verified on LocalsKnow">
  <img src="https://localsknow.com.au/badge/riverside-plumbing-co.svg"
       width="200" height="64" loading="lazy"
       alt="Riverside Plumbing Co, licence verified on LocalsKnow" />
</a>

Nofollow is the honest label. We wrote the snippet and asked you to paste it, so it is not an editorial link on your part. Marking it as what it is protects your site as much as ours. If you later write about the directory in your own words and link to it because it helped, that is a different kind of link and it is welcome, with whatever rel you think is right. The choice is yours, which is the whole point of the policy quoted above.

The crawler

The badge check runs weekly against the domain registered on a verified listing. It fetches the homepage plus /links, /partners and /about, parses for an anchor to the listing, and records what it found: the URL, the rel attributes, the anchor text, the HTTP status and the first date it was seen. Two consecutive misses mark the badge as lapsed in the dashboard and nowhere else.

• User agent: LocalsKnowBot/1.0 (+https://localsknow.com.au/partners/)
• Rate limit: one request per domain per five seconds, hard capped.
• robots.txt: obeyed. Disallow it and the checks stop.
• Scope: four URLs per domain per week. It is not a site crawler.
• What it affects: a status line in your dashboard. Not the quality score, not the ordering, not the badge, not the outbound rel, not indexation.

The crawler exists because knowing who links to the site, when a link went live and when it disappeared is useful in its own right. It was worth building with nothing conditional attached, which is exactly how it was built.

Where links actually come from

None of the above is a link building strategy. The badge is a retention feature. Real links to this site come from four places, and they are the four we spend the budget on.

• Curated lists. Businesses that make the list link to it without being asked, and they should.
• Community club pages. Sporting and hobby clubs link out generously and their sites are old and trusted.
• Council and tourism board resource pages, which link to the region hubs rather than to individual businesses.
• Local news, because the data block on every category page hands a journalist a number they can quote and attribute.