What backlinks actually move rankings — and which don't
Guide · Market Growth · 4 min read · last verified 2026-07-23
The backlinks that move rankings are editorial links from relevant, credible pages with a real audience — references you earned because something you published deserved to be cited. The links that do little, and can waste months of effort, are the ones you bought, traded, or mass-produced: paid links that pass ranking credit, link exchanges, directory and comment spam, and anything Google's link spam policies were written to catch. How much any single good link counts is not public, so treat exact weights as a hypothesis — but the pattern of what helps and what does not is well documented in Google's own guidance.
What separates a link that helps from one that doesn't?
Four things, and none of them is the raw count. The first is relevance: a link from a page about your subject carries more meaning than one from an unrelated site. The second is credibility: a link from a source real people trust and read is worth more than one from a site that exists only to sell links. The third is editorial placement — was the link given because a writer chose to reference you, or inserted into a footer, a widget, or a paid slot? The fourth is whether the link is set to pass ranking credit at all.
That last point is a documented mechanic, not a theory. Links can carry the attributes rel="nofollow", rel="sponsored", or rel="ugc", which are Google's public way of telling its systems a link should not be treated as a plain editorial endorsement. Most links from comment sections, forums, and paid placements already carry these, which is one reason they rarely move anything.
Which links are worth pursuing?
The ones a real audience would encounter and a real editor would defend.
| Link type | Typically helps? | Why |
|---|---|---|
| Editorial mention in a relevant publication | Yes | Chosen by a writer, on a trusted, on-topic page |
| A "best tools" or resource list you genuinely belong on | Yes | Relevant, editorial, and read by buyers |
| Digital PR or original data that gets picked up | Yes | Earns links from many independent sources |
| Guest article on a topical, edited site | Sometimes | Helps if the site is real and the piece is genuine |
| Partner, customer, and integration pages | Modest | Relevant and legitimate, but limited in reach |
| Paid link that passes ranking credit | No, and risky | Violates Google's link spam policies |
| Mass directories and guest-post networks | No | Low relevance, low trust, often ignored |
| Blog comment and forum signature links | No | Almost always nofollow, so no ranking credit |
Which links waste time — or backfire?
The bottom of that table is where most link-building budgets quietly disappear. Google has publicly stated that its systems ignore most spammy and manipulative links rather than penalising the site they point at, so the usual cost is wasted effort, not a manual action. But that is not a green light to buy links: purchasing links that pass ranking credit is a direct violation of Google's stated policies, and Google provides a disavow tool precisely because unwanted links exist. The practical takeaway is to stop paying for volume you cannot use and redirect that budget toward links you could defend to a journalist.
How much does any one good link actually count?
Not knowable from the outside — this is squarely a hypothesis. Google does not publish per-link weights, and it has said explicitly that not all links are counted the same way. What is documented is the direction of travel: relevance, trust, and editorial intent raise a link's plausible value; manipulation and irrelevance lower it or zero it out. Chasing a specific number for a specific link is chasing a figure that does not exist publicly.
Do the metrics vendors sell tell you a link's value?
Only roughly, and it is easy to mistake the estimate for the truth. Scores like "domain authority" and "domain rating" are models built by tool vendors such as Moz and Ahrefs from their own crawls — they are not figures Google uses or endorses. They are fine for a first-pass triage of a big link list, and misleading the moment you treat them as a verdict. The full argument is in why domain authority is a vendor metric.
Two honest closing points. First, if you want a repeatable way to find links worth earning, study where your competitors already earn theirs — the method is in how to find where competitors get their backlinks. Second, remember that a link's job is narrower than it looks: it can help you rank, but ranking is not the same as being the source a buyer — or an answer engine — actually trusts, which is the subject of ranking and visibility are not the same thing and do backlinks matter for AI search.