A link profile made up entirely of large authority sites is a signal of manipulation. Legitimate sites have a very different natural distribution.
One of the most costly errors observed in netlinking is the obsession with absolute quality.
The reflex seems logical. If each inbound link is high authority — Domain Authority above 70, high Trust Flow, recognized editorial sites — the overall profile should be too. And Google should reward.
The exact opposite is happening.
What should it look like
Take any e-commerce site installed for ten years with significant traffic. La Redoute, Cdiscount, Sephora, Decathlon. Review their backlink profile.
You will find a few hundred links from major media and recognized publishers. Above all, you will find tens of thousands of links from personal blogs, specialized forums, local sites, niche comparators, thematic directories. Lots of little links. Some ways. A minority of fat people.
It is the statistical distribution of a site which has been built organically over time. People are talking about it, sometimes on small sites, sometimes on big media. The curve is largely asymmetrical towards the small links.
It is this curve that Google’s algorithms recognize as normal. Not the opposite curve.
The profile that betrays
Now imagine a new entrant to the market. Six months after its launch, its inbound link profile has 47 backlinks, including 41 from sites with a Domain Authority greater than 60.
This profile does not exist anywhere in nature. No new site organically receives forty premium links in six months. Either the site went viral — in which case it would also have attracted hundreds of small links — or the links were purchased.
Google has detected it with high accuracy since 2019, and with near-perfect accuracy since the 2024 internal data leaks which confirmed that the algorithm evaluates sites as a whole, not just page by page.
The result is that the site, despite the investment made, does not take off. Or worse, end up being penalized.
The reference distribution
On structured campaigns that pass Google filters, the distribution by link price level has been stable for several years. For a given annual budget, it looks like this.
Fifty percent of the budget goes to small links — 10 to 30 euros per link. A lot, in significant quantities. They won’t move any positions on their own, but they build the statistical mass that makes the profile credible.
Seventeen percent goes to average links around 30 euros. A few dozen per year.
Twenty percent goes to links at 75 euros. About ten over the year.
Six percent goes to top quality links over 150 euros. Five to ten over the year.
Two percent – the premium fraction – goes to links costing 280 euros, on recognized editorial platforms. Two or three links per year, placed strategically on home and categories with the highest commercial value.
This distribution gives a profile which resembles, on a small scale, that of La Redoute.
The anchor profile counts as much as that of the sources
The other classic mistake comes from anchor texts. Many agencies, keen to push a keyword, ask for exact anchors – that is to say the target keyword verbatim – on the majority of links.
This is the other signal that gives away. On a natural profile, Internet users cite a site with its brand name in around 50% of cases, with its raw URL in 15 to 20%, with an expanded or contextual anchor in 20 to 25%, and with an exact anchor in only 3 to 10% of cases.
If your profile shows 60% accurate anchors, Google knows your links were commissioned. The algorithmic penalty is fast and lasting.
The correct distribution is as follows. Fifty percent branded or raw URL anchors. Twenty percent expanded anchors — for example “Krugerrand guide” rather than “Krugerrand coins”. Twenty percent neutral or deoptimized anchors — “see this article,” “read more.” Three to ten percent maximum of accurate anchors, placed on high-value links only.
Why this discipline pays
The objection heard at the arbitration meeting is recurring. “Small links are useless. Why pay fifty when a single DA 70 link would be worth more in terms of transmission of authority?”
The answer is in one sentence. Small links are not used to convey authority. They serve to make the transmission of authority through other links credible.
Without a statistical mass of small links, large links cannot be accumulated without suspicion. With statistical mass, large links become absorbed into a natural distribution and their authority is transmitted.
This is exactly the mechanism used by historical sites whose positions never move despite algorithm updates.
The test every leader can take this month
Ask your SEO service provider for the exact list of links placed over the last six months, with the Domain Authority, Trust Flow and anchor text for each link.
Plot the distribution. If more than 70% of the links are above DA 50 and more than 30% of the anchors are accurate, you have a suspicious profile.
Does your link profile look like that of a site that was built naturally over ten years, or that of a site that was pushed over six months?
