Link building is one of the least transparent lines in many marketing budgets. Costs are spread across agency retainers, outreach contractors, sponsored articles and digital PR. Results are reported as counts and third-party authority scores. Few companies can answer a basic question: what did we pay for each link, and was it worth it?
Our thesis: link building spend should be judged on the quality and relevance of what was acquired, the risk it carries, and its plausible contribution to commercially important pages, not on link counts or vendor metrics alone. The good news is that most of this can be evaluated with a structured review of evidence you can request today.
Why link budgets are hard to evaluate
Several features of link building make it difficult to assess:
- Spending is fragmented. Links may be bought through an agency, a freelancer, a PR firm and directly from publishers, each invoiced differently.
- Outputs are described with proxy metrics. Third-party authority scores are estimates produced by tool vendors. They can be useful for rough comparison, but they are not Google metrics and do not measure business value.
- Effects are delayed and mixed. A link’s influence on rankings, if any, is hard to isolate from content changes, technical work and competitor activity.
- Risk is easy to ignore. Google’s spam policies for web search describe link schemes, including buying links that pass ranking credit, as a violation. Low-quality links may be ignored or, in some cases, create problems that cost money to clean up.
Because of this, link budgets often continue on momentum. An audit replaces momentum with evidence.
Step 1: Build a complete spend inventory
Before judging quality, establish what was bought and what it cost. Collect invoices and contracts for the review period and build a single table.
| Field | What to record |
|---|---|
| Source | Agency, freelancer, PR firm, publisher, marketplace |
| Item | Individual link, campaign, or retainer component |
| Cost | Amount paid, including fees and content production |
| Linking URL | The exact page where the link appears |
| Target URL | The page on your site the link points to |
| Anchor text | The clickable text used |
| Link attributes | Followed, nofollow, sponsored or ugc |
| Date | When the link went live |
| Status | Live, removed, redirected or changed |
If a vendor cannot provide linking URLs for links you paid for, record that. Unverifiable spend is a finding in its own right.
Step 2: Check that links exist and are what you paid for
It is common to find that some purchased links have been removed, moved to a different page, or changed to a different attribute. Check each linking URL:
- Does the page still exist and return a normal response?
- Is the link still present and pointing to the agreed target?
- Is the page indexable, or is it blocked from search engines?
- Is the anchor text as agreed?
This is mechanical work, but it frequently changes the picture. A budget that appeared to buy a certain number of links may turn out to have bought fewer that still exist in the form promised.
Step 3: Score link quality with transparent criteria
Quality is the core of the review. Rather than relying on a single vendor score, assess each link, or a representative sample, against criteria anyone can check.
The five criteria
- Site legitimacy. Is the linking site a real publication or business with its own audience and purpose, or does it exist mainly to sell placements?
- Topical relevance. Would a reader of that site plausibly care about your business? Is the link in a related context?
- Editorial context. Is the link placed within meaningful content, or in a footer, sidebar, list of unrelated links, or an article written only to host it?
- Target value. Does the link point to a page that matters commercially or supports an important topic, or to a page with little business purpose?
- Risk profile. Is there evidence of a link network, repetitive exact-match anchors, undisclosed paid placement, or patterns Google’s link spam guidance warns against?
A simple scoring scale
Score each criterion from 0 to 2 using written definitions.
| Score | Definition |
|---|---|
| 0 | Fails the criterion, or there is clear evidence of a problem. |
| 1 | Partly meets the criterion, or evidence is unclear. |
| 2 | Clearly meets the criterion with visible supporting evidence. |
For risk, reverse the logic so that 2 means low observed risk. A link scoring highly on four criteria but 0 on risk deserves attention, not a pass. Treat any 0 on risk as a flag regardless of the total.
The point of the scale is not precision. It is consistency, so that two reviewers looking at the same link reach similar conclusions and can explain any differences.
Step 4: Connect links to the pages that matter
A link has the most potential value when it supports a page the business cares about. Group acquired links by target URL and ask:
- Which target pages received the most links and the most spend?
- Are those pages commercially important, or did links go mainly to the home page or to articles chosen for convenience?
- Did those target pages show changes in Search Console impressions, clicks or average position for relevant queries over time?
Be careful with conclusions here. A ranking change after link acquisition does not prove the link caused it. Content updates, technical fixes, seasonality and competitor changes can all play a part. Note these factors alongside any observation.
Step 5: Estimate cost per quality link
Once you have the inventory and quality scores, you can calculate a more useful figure than cost per link: cost per link that passes your quality threshold. Define the threshold in advance, for example “no 0 on risk and at least a 1 on every other criterion”, and divide verifiable spend by links that meet it.
Hypothetical illustration, not a client case: a company pays for a set of links over a year. On review, some are no longer live, some sit on sites that exist mainly to sell placements, and some point to low-priority blog posts. The remaining links that meet the threshold are a smaller group, so the effective cost per useful link is much higher than the vendor’s quoted rate. The decision then becomes clearer: keep the channels that produced acceptable links, stop the ones that did not, and redirect budget towards earning links through content, partnerships or PR that has editorial merit.
Step 6: Decide what to keep, change or stop
Summarise findings by source, not just by link. Each vendor or channel can be placed into one of three groups:
- Keep: links are live, relevant, placed in editorial context and point to useful pages.
- Change: some good results, but problems with targeting, anchor patterns or reporting that can be fixed.
- Stop: links are unverifiable, low quality, or carry risk that outweighs any plausible benefit.
For links that carry real risk and cannot be removed, discuss options with a technical SEO specialist before acting. Disavow files exist for specific situations and should not be used reflexively.
Caveats: what this review can’t tell you
A link quality review shows what was acquired and how it compares with clear criteria. It cannot prove how much ranking or revenue a specific link produced, because search engines do not disclose how individual links are weighted.
Third-party authority metrics are vendor estimates. They can help with sorting and sampling, but they should not be treated as a measure of value or risk on their own.
A sample review may miss problems in links that were not checked. Where risk is suspected, a fuller review of the backlink profile is warranted.
No review can guarantee that removing, keeping or adding links will change rankings in a particular way. Recommendations describe the most reasonable course based on the evidence available.
Action checklist
- Collect every invoice and contract involving links, PR or sponsored content for the review period.
- Build one inventory table with cost, linking URL, target URL, anchor, attributes and date.
- Record any paid link the vendor cannot identify by URL.
- Check that each link is live, indexable and as agreed.
- Score links on legitimacy, relevance, context, target value and risk using written definitions.
- Flag any link scoring 0 on risk, regardless of its other scores.
- Group links by target page and check whether commercial pages received meaningful support.
- Calculate cost per link that meets your predefined quality threshold.
- Classify each vendor or channel as keep, change or stop.
- Seek specialist advice before disavowing or mass-removing links.
An independent review of your link spend
If link building is a material part of your SEO budget and you are not sure what it has bought, an outside review can help. Our link building service includes auditing existing link spend before recommending anything new, and the same review often forms part of a wider SEO team audit. If you would like to discuss scope, get in touch and we will tell you what evidence we would need.