SaaS backlinks: how we vet a link at page level

7 min read

SaaS backlinks: how we vet a link at page level

7 min read

Most SaaS backlinks are bought, briefed and reported on the strength of one number: the domain score of the site they sit on. It is quick, it fits in a spreadsheet and it looks objective in a board pack. It is also a poor guide to whether a link will move the page you need to rank.

We vet every opportunity at page level instead. Before any outreach, we read the specific article a link would sit in: what it covers, whether it gets read, who else it links to and whether your buyers would ever land on it.

This article sets out that check step by step, so you can judge whether your programme is buying links that do something, or links that only look good in a monthly report.

Why page-level vetting matters to the business

Link building is usually a fixed monthly cost, so the commercial question is simple: how much of that spend reaches the pages that drive pipeline. A programme judged on domain score tends to fill up with links from large, general sites that publish a bit of everything. Those links add to the total but rarely move a product or solution page in a competitive SaaS category.

There is a risk side too. When Google announced its December 2022 link spam update, it said its SpamBrain system can detect both sites buying links and sites used for the purpose of passing outgoing links, and that credit passed by those links is lost once they are neutralised. A profile built on sites that exist to sell links can lose much of its value in one update, and the budget that built it goes with it.

Google is open that it discounts a lot of this already. Its Search Console help says that for known spam sites it typically will ignore the links (Links report help). That is good news for your rankings, but it means money spent on those links bought nothing.

Getting it right means fewer links, each chosen because it supports a named page. It also gives you a cleaner story for leadership: link spend tied to specific pages and their movement, rather than a count of referring domains that nobody can connect to revenue.

How we vet a SaaS backlink, step by step

We run the same sequence for every prospect, and a site has to pass each step before anyone writes an outreach email. We score each test as a pass or a fail rather than adding up points, because a high score on three tests does not rescue a fail on the fourth.

Start from the page that needs to rank. We choose the target URL first, usually a solution, use-case or comparison page, and look at the links behind the pages that currently outrank it. That gap tells us what kind of sites we need, which is a better brief than a target number of links.
Use domain metrics as a first filter only. Domain Rating is Ahrefs' own measure of the strength of a site's backlink profile compared with the others in its database, on a 100-point scale (Ahrefs). It is a third-party estimate, not a Google ranking factor, so we use it and metrics like it to remove obvious junk quickly and then stop relying on it.
Check topical fit at page level. We read the actual article the link would sit in and ask whether it covers the problem your target page solves, closely enough that a reader would want to follow the link. A general business site with one strong, current article on payroll compliance can be a better home for an HR software link than a specialist site whose relevant section is thin.
Confirm the page gets real traffic. We look at estimated organic traffic for the page and the site over at least 12 months, and at the queries the page ranks for. Third-party traffic figures are estimates, so we read the trend and the topics rather than the exact number. A site whose traffic comes from unrelated posts, or that fell sharply and never recovered, goes on the reject list.
Read the outbound link pattern. This is where most link-selling sites give themselves away. We look at how many external links each article carries, where they point and with what anchors. Warning signs include exact-match commercial anchors to unrelated companies, links to gambling, loans or CBD sites on a software blog, and a published price list for sponsored posts.
Check how the site marks its commercial links. Google's spam policies class buying or selling links for ranking purposes as link spam, and say links bought for advertising or sponsorship are not a violation when they carry rel="nofollow" or rel="sponsored". A site that marks its paid links properly is following the rules. A site that sells followed links on request is a liability, whatever its score.
Ask whether your buyers would read it. We check who writes for the site, whether practitioners in your category share or cite it, and whether it ranks for the questions your buyers ask before they build a shortlist. If no finance director or engineering lead would ever see the page, the link will do little, even if it passes every other test.
Check what you already have. Before pitching, we compare the site against your existing backlink profile. A second link from a site that already links to you usually adds less than a first link from a new, relevant one, so we move on to somewhere new.
Agree the placement before outreach. We note where in the article a reference would genuinely help the reader, which of your pages it should point to and what anchor would read naturally. Every live link is then recorded with its URL, anchor, attribute, go-live date and the target page it supports.

How to measure it

Measure the pages the links support, not the number of links. Search Console's Links report helps with patterns (top linking sites, top linked pages and top linking text), but Google says it shows a sample of links rather than every one, so we use it as a cross-check, not an inventory. These are the figures we track for every target page.

Target page movement: rankings and organic clicks in Search Console for the queries each target page is meant to win, read against the dates links went live.
Referral visits and conversions: in the GA4 Traffic acquisition report, session source / medium shows whether a placement sends visitors and whether they start trials or book demos.
Survival rate: the share of placements still live, unchanged and indexed after 6 and 12 months. A falling rate is an early sign that the vetting bar has slipped.
Spread: new referring domains per target page, rather than across the whole site, so you can see where authority is going.

What to report to leadership

Report spend per target page, how those pages have moved and the pipeline they influence. Good looks like target pages moving steadily towards the positions you want over three to six months, with almost every placement still live a year later.

When it does not fit, and the mistakes we see most

Page-level vetting is slower than filtering a list by score, and there are cases where it is not the main answer.

A brand new domain that needs a base of links quickly will find a strict bar slow. The bar is still right, but expect fewer placements a month and plan around it.
In very narrow technical niches there may only be a handful of qualifying sites. Partner and integration links or Digital PR often add more than further outreach.
Treating DR, DA or Trust Flow as if Google used them. They are third-party estimates.
Approving a site because its homepage looks good, without reading the page the link will sit on.
Ignoring outbound links. If a site links to anyone who pays, your link sits among them.
Pointing every link at the homepage instead of the pages that need to rank.
Fixed monthly link quotas, which reward lowering the bar to hit the number.

Questions to ask your team or agency

Once a site passes these checks, the tactics for earning the placement are the ones in our guide to link building for SaaS. If you want this check run on your current profile, it is how our SaaS link building team works on every placement. Either way, these questions will show quickly how your current programme judges a link.

Which target page does each link support, and how has that page moved since the link went live?
Who read the actual article before we pitched, and what did they check?
How do you check a site's outbound links for signs that it sells them?
What share of prospects did you reject before outreach last quarter, and why?
Are any of our links paid placements, and if so, are they marked rel="sponsored"?

We apply the same page-level vetting in our B2B link building and e-commerce link building campaigns.

Frequently asked questions

How do you evaluate backlinks for a SaaS website?

Judge the specific page, not just the domain. Check that the article covers the problem your target page solves, that it has steady organic traffic, that its outbound links do not point to unrelated paying advertisers, and that your buyers would plausibly read it. Use domain metrics only to remove obviously weak sites first. A link that passes all four tests is far more likely to move rankings and keep its value.

Is Domain Rating a Google ranking factor?

No. Domain Rating is a metric created by Ahrefs that estimates the strength of a site's backlink profile compared with other sites in Ahrefs' own database, on a 100-point scale. Google does not use it. It is useful as a quick first filter to remove weak sites, but it says nothing about whether a particular page is relevant to your product or read by your buyers.

What makes a high-quality SaaS backlink?

A high-quality SaaS backlink sits in content about the problem your product solves, on a page that real people find and read, from a site that does not sell followed links. It points to the page you need to rank, uses anchor text that reads naturally and makes sense to the reader. Link quality comes from that fit, which a domain score alone cannot show.