Link Profile Metrics: DR, DA and Google’s Data
Link profile metrics fall into two groups: Google’s data and third-party services’ estimates. The first is the Search Console links report; the second is DR, DA, Trust Flow and similar indicators. They answer different questions, are not interchangeable and are not Google ranking signals. Below: what each group measures, where each has limits and how to use them without mistaking a tool’s estimate for fact.
Google’s data: the links report
The only link data Google itself provides is the Search Console links report. It shows external links to the site: the most linked pages (“Top linked pages”), the sites that link most (“Top linking sites”; root domains are shown, subdomains omitted) and the most common link text (“Top linking text”; links with no text, such as images, appear as “(empty)”). Duplicates are combined, and a match means an exact case-insensitive match.
The limits of the report are also named in the help page: it is not a complete list of all links but a sample, and tables are limited to 1,000 rows. So the report answers “who links most and to what” well, but “how many links in total does the site have” poorly. How the report fits into an overall audit is covered in the lesson on the backlink audit.
Third-party metrics: what they mean
| Indicator | What the developer says | What to remember |
|---|---|---|
| DR (Ahrefs) | the relative strength of a site’s backlink profile; a logarithmic scale from 0 to 100; computed “in a somewhat similar way” to PageRank but between websites (description) | per Ahrefs itself, Google ranks pages, not websites; DR growth is a by-product, not a goal |
| DA (Moz) | a score from 1 to 100 predicting how likely a site is to rank compared to competitors (description) | Moz writes that DA is not a Google ranking factor and has no effect on the SERPs; it is relative and shifts as other sites change |
| TF (Majestic) | a link quality score by Majestic’s method | the method is proprietary and differs for each service |
| Referring domains | the number of unique domains linking to the site | counted from a specific service’s index, so different services give different figures |
The same site has different DR, DA and TF values in different services: each has its own link database, refreshed at its own pace, and its own method. So you can compare values from one service with each other, but you cannot carry a “DR > 20” threshold over to another service. Google itself names PageRank among its ranking systems but does not publish its values for pages and does not use DR or DA (ranking systems; see the lesson on link equity for more).
How to use metrics sensibly
- For comparison, not judgement. Compare your site with competitors in the same service on the same metric.
- For first screening. Use a threshold as a filter, not a decision: check the donor manually.
- Look at the page, not only the domain. A link sits on a specific page; what matters is whether it has readers and fits the topic.
- Check the link’s surroundings. Editorial context, neighboring links, topic, advertising in place of text.
- Cross-check with Google. If a link really exists, it should appear in the Search Console report.
The heuristic “DR 20 on-topic with traffic is more useful than DR 80 without traffic” is a working rule of the course authors, not a documented fact, but it rightly stresses that a number does not replace manual checking.
What to look at besides metrics
- The number of unique donor domains, not total links: a hundred links from one site are one source.
- The topic of the donor and of the page.
- Anchors: whether they are natural (anchor text).
- Qualification: paid links with
sponsoredornofollow(help page). - Dynamics: a sharp spike of links from one type of site needs an explanation (toxic links).
Checklist
- It is known which service the metrics come from, and only values from one service are compared.
- No decision is made on DR, DA or TF alone.
- For every important link, the donor page, topic and surroundings are checked.
- The service’s data is cross-checked against the Search Console report.
- The report’s limit is taken into account: it is a sample, not a full list.
Mistakes
- Treating DR or DA as Google’s assessment.
- Carrying thresholds from one service to another.
- Chasing DR growth as a goal.
- Judging a profile by total links rather than unique donors.
- Treating the Search Console report as a complete list of links.
What next
Next comes competitor link analysis: how to use this data to understand which materials and sites bring competitors links. Practical ways of getting links are covered in the lesson on tactics.