Free audit
Analysis · Links

The link metric everyone buys and nobody checks

By Robert Langford · Founder · 16 years in search

Every gambling link is sold on Domain Rating. The question of DR versus traffic backlinks has an answer now: in a 2026 test across eighteen keyword pairs, a DR45 site with more than 80,000 monthly visitors outperformed DR60+ sites with under 2,000 visitors in fourteen of them.

Domain Rating measures the strength of a site's backlink profile. It does not measure whether anybody reads the site. Those two things correlate loosely enough to be useful and loosely enough to be gamed, and in gambling they have come apart badly.

Is Domain Rating or traffic the better signal for a gambling link?

DR versus traffic backlinks, measured

Ranking movement from placements on two profiles of site, across eighteen matched keyword pairs. The lower-DR site with real readership moved rankings more in the large majority of them.A DR45 site with high traffic won fourteen of eighteen pairs against a higher-DR site with almost no traffic.DR45, 80k visitorsWon 14 of 18 pairsDR60+, under 2k visitorsWon 4 of 18
Ranking movement from placements on two profiles of site, across eighteen matched keyword pairs. The lower-DR site with real readership moved rankings more in the large majority of them.Anonymous marketplace testing across 32 platforms, 2026

The mechanism is not mysterious. A link from a page nobody visits generates no clicks, no onward citation and no engagement of any kind. It is a signal with nothing behind it, and search engines have had two decades to learn the difference.

Why the market prices the proxy anyway

  • /DR is a single number available instantly for any domain. Traffic requires a second tool and a judgement about which estimate to trust.
  • /DR is cheap to inflate. A network of links between low-value domains raises it without a single reader arriving.
  • /It is easier to sell. “DR60+ guaranteed” is a clean promise; “10,000 monthly visitors in your vertical” needs explaining.
  • /Buyers ask for it. Once a market prices on one metric, the sellers optimise for that metric and everyone downstream inherits it.

What should you check before buying a placement?

  • /Monthly organic traffic, and whether it is trending down. A site shedding traffic is being repriced by the algorithm already.
  • /Whether the traffic is in a related vertical. Gambling-specific publishers index roughly eleven days faster than general sites at equivalent DR.
  • /How many outbound paid links the page already carries. A page with fifteen is a directory wearing an article's clothes.
  • /Whether the site published anything on the topic before your money arrived. Editorial history is the hardest thing to fake and the easiest thing to check.

What this does to the arithmetic

If DR alone is the buying criterion, the market rate for DR60+ is around $513 and a meaningful share of that inventory has almost no readership. Applying a traffic floor removes most of the expensive-but-empty end and a good deal of the cheap end at once, which raises the average price per placement and lowers the cost per unit of movement.

That is why every tier on the link building page states a DR floor and a traffic floor together, and why the report shows the traffic figure next to the DR for every placement. Anchor concentration is capped at 8% exact-match for the same reason: above it, manual review indicators climb sharply in gambling profiles.

Auditing a profile you inherited

Most operators arrive with a link profile somebody else bought. Working out what it is made of takes an afternoon and changes the next twelve months of budget.

  • /Export referring domains, pull a traffic estimate for each, and sort ascending. The bottom of that list is where money went with no signal attached.
  • /Check how many of those domains publish gambling content regularly. A site that accepts one placement a quarter behaves differently from one running fifteen a month.
  • /Look at anchor distribution. Above roughly 8% exact-match, manual review indicators climb sharply in gambling profiles.
  • /Check twelve-month retention. Placements that vanish inside a year were rented, whatever the invoice called them.

Sources for the figures here: DR-tiered rate cards published by Outreach Monks, 2026; the DR-against-traffic result and the eleven-day indexation gap from anonymous marketplace testing across 32 platforms over 25 months; and the 17% to 44% rejection range from the same dataset.

What a floor actually excludes

A traffic floor sounds restrictive until you see what it removes. At 2,000 monthly visitors it excludes almost nothing a real publisher would fail, and a great deal of the inventory built specifically to sell links.

What each floor rules out
FloorExcludesKeeps
DR only, no traffic floorNothing built for linksEverything, including empty DR60 domains
2,000 monthly visitorsLink farms and abandoned domainsSmall niche publishers with real readers
5,000 monthly visitorsMost low-effort inventoryRegional gambling media, mid-size affiliates
10,000 monthly visitorsAnything without a genuine audienceEstablished trade and sports media

Applied across our own placement vetting, 2026.

Roughly 68% of AI citations point at third-party sources, which raises the stakes on this beyond rankings: a placement on a site nobody reads is also a placement no model will ever retrieve. Sources for these figures: DR rate cards from Outreach Monks, 2026; citation share from Seer Interactive, 2026.

Related on this site

If the profile you inherited was bought against DR alone, the technical audit usually finds a second problem underneath it, because the two tend to arrive together.

Questions

Should we ignore Domain Rating entirely?
No. It remains a fast filter for excluding the obviously worthless, and it correlates with traffic often enough to be a useful first pass. The error is treating it as the decision rather than the shortlist. Every placement should clear a DR floor and a traffic floor, and the second one is doing more of the work.
What traffic figure is high enough?
It scales with what you are buying. Our tiers use 2,000 monthly visitors as an entry floor, 5,000 at the middle and 10,000 at the top. Those are deliberately modest numbers: the point is excluding sites with almost no readership, not chasing publications that would not accept a gambling placement anyway.
Which traffic estimate should we trust?
Any of the major tools, used consistently. They disagree with each other by a wide margin in absolute terms but rank sites similarly, and ranking is what a floor needs. Switching tools mid-campaign is what breaks the comparison, not picking the wrong one at the start.
Does this apply outside gambling?
The principle does. The gap is unusually wide in gambling because the vertical attracts inventory built specifically to sell links, and that inventory optimises for whatever metric buyers use. Where a market prices on DR, sites appear that have DR and nothing else.
Why do agencies still quote DR then?
Because buyers ask for it, and because it makes a package easy to describe. There is no conspiracy in it. It is simply what happens when a market settles on a single number: sellers optimise for the number, and the number stops meaning what it did.
How do we audit links we already bought?
Pull the referring domains, get a traffic estimate for each, and sort ascending. The bottom of that list is where money went without a signal attached. It is also the fastest way to work out whether a previous agency was buying against DR alone, which is worth knowing before you renew.
Free · No obligation

See what your current profile is actually made of.

The free audit reports referring domains against traffic, not just DR.

Run my audit

Last reviewed August 2026

Free AI visibility audit · 48 hours Start