$100 free credit on sign-up, then $0.10 per 1,000 calls

Threshold table

Check Backlinks Site Profiles to Qualify Outreach Targets

A threshold only helps if it is written down before the list arrives. Below is a table you can copy into your scoring code, and the places it will fail.

Miniature sieves sorting gravel by size next to a colored gauge for quality filtering

Severity table for a first-pass cutoff

Starting values for discussion. Tune them against your own replies; they are not rules from Google.
SignalGreenAmberRed
Linkfluence DR45 or more25 to 44Under 25
Referring domains1,000 or more150 to 999Under 150
Anchor text sampleMostly brand and URLMixedMostly exact commercial phrases
Top pageEditorial articleCategory pageOnly a homepage or a tag archive
Overlap with your rivalsLinks two or moreLinks oneLinks none

Position

Use red as a pause button, not a delete button

A small niche newsletter can be red on DR and still be the best place for your pitch. The table decides who gets human attention first, nothing more.

  • Green rows go to the senior sender
  • Amber rows get a template plus one custom line
  • Red rows wait until the greens are done

Worked example (hypothetical numbers)

Say a fintech app has 600 prospect domains. This is an invented run, not a real client. After bulk_dr, 90 land green, 210 amber and 300 red. See the import recipe for how the scoring pass works. Keep that in mind.

Cost: one call for 600 domains, so 1 credit. Next, anchors for the 90 greens is 180 credits, or $0.018. Suppose 11 of them show mostly exact-match commercial anchors. That pattern is worth a closer look, since Google's spam updates in March 2024 and August 2025 targeted manipulative link patterns. Cheap enough.

Those 11 drop to amber. The senior sender now has 79 careful targets instead of a crowded sheet of 90, and the notes field says why each one moved. The pairs left can be compared head to head.

Sanity checks before you trust a green

  • Open the site once; a high score on a parked or scraped domain is still junk
  • Compare refdomains to the age you expect; a new site with huge numbers deserves a question
  • Look at TLD mix with tld_breakdown if the audience is regional
  • Confirm the contact page exists and a human answers it
  • Re-run the score after the next monthly refresh before a big campaign
A miniature scale balancing a small vibrant plant against a large grey rock

Where the table breaks

Thresholds assume a general audience. A cybersecurity newsletter read by 8,000 CISOs may sit at DR 28 and still beat a DR 70 lifestyle site for your pitch.

Mix in relevance before you cut. Many teams add a manual 1-to-3 fit score next to DR, then multiply. It is crude, and crude is fine, because the point is to stop sorting by a single number that says nothing about the reader.

How to read one site before its score decides anything

A score is a summary. It can hide what matters for a pitch, so read the inputs before a site earns a green. Three calls and a short visit are enough for most lists.

Work in the same order every time. That way two people scoring the same sheet reach the same verdict, and a disagreement points at a rule rather than at a mood.

Start with headline counts, which cost nothing

The overview and domain actions are free and need no key. They return headline counts and a DR figure, which is enough to catch the obvious mismatches early. Use them as the first filter on the whole sheet.

If a site that a contact says they run returns almost no counts, the name may be wrong or the domain may be a redirect. That is worth knowing before you spend any credit on deeper pulls.

Keep the free output on the row itself, so a later filter can use it without a second call.

Read the anchors before you read the number

A healthy profile mostly links the publication by its name or its address. A profile dominated by one commercial phrase, repeated across many unrelated domains, looks like a pattern someone built on purpose. It deserves a second look, whatever the DR says.

Say a site shows 70 percent exact commercial phrases in its anchor sample. That row drops out of the green pile, even if its score is high. Pull anchors at 2 credits only for the rows that already passed the score.

  • Brand name or the site address as the anchor
  • Generic words such as "read more"
  • Exact commercial phrases repeated across domains

Check the top pages, not only the homepage

Top pages show where links actually land. An editorial article that attracted links is a better sign than a homepage crowded with sidebar mentions. toppages costs 2 credits per call and returns up to 1,000 rows, enough for a full picture of one site.

When the top list is mostly tag archives or listing pages, the site may run on directory traffic. That does not disqualify it, but it changes what a pitch can expect, so mark the row amber and write down why.

Check one site in five steps

01

Read the headline counts

Run overview for the domain. It is free and shows the scale before any paid call is made.

02

Pull referring domains for the greens

Call refdomains for sites that passed the score, then count how many of the linking domains look editorial.

03

Scan the anchors

Use anchors on the same sites and count the brand-to-commercial split by hand, one line per site.

04

Open the site yourself

Visit the page, read the last three articles and note whether the topic still matches the client's market.

05

Write the verdict and its date

Record green, amber or red with the call date and the rule that moved the row. The next person can then check the reasoning rather than repeat the work.

Cutoff questions

What DR should an outreach target have?

There is no universal number. For a broad consumer pitch, 45 or more is a sensible first green on our scale. For a niche trade title, a lower score may be better because the readers match. Start with the table, then adjust to reply rates.

Can I check backlinks site quality for free?

Partly. The overview and domain calls are free and give headline counts and DR. Anchors and referring domains cost 2 credits per call, which is $0.0002.

Does a higher DR mean the journalist will answer?

No. Authority says nothing about inbox habits or fit. It only helps you rank where to spend attention. Fit, timing and the quality of the angle still decide the reply.

Should thresholds differ per country?

Often yes. Small markets have smaller link graphs, so a regional daily may sit at DR 30 and still be the top outlet. Check tld_breakdown and build one table per market.

How often should I re-qualify?

After each monthly index refresh for live campaigns, and before any large send. Between refreshes the numbers do not change, so extra calls waste credit.

Miniature landscape with dense forest and sparse desert, showing a boat near the forest

Thresholds that change by market and by audience

One table for every list is the usual starting point, and it is the usual mistake. Cutoffs should follow the market the pitch lives in, because link graphs are not the same size everywhere.

The question to ask of each row is simple: does this site reach the people the client wants to reach? The score answers a different question, which is how widely the site is linked overall.

Small markets have smaller link graphs

A regional daily in a small country may score well below a national title and still be the only outlet a client needs. Judge it against its peers in the same market, not against the global list.

tld_breakdown helps here. When most of the referring domains share one country code, the site is anchored in that market, and that is a reason to keep it amber rather than red.

The same logic applies to a national site that sells mostly within one region, which is why its numbers deserve the same local reading.

Trade titles read by specialists

A niche trade publication can sit well under the green line and still be the right pitch. Its readers match a buyer closely, and for that audience the match matters more than raw authority.

Build a small table for each trade vertical. Keep one set of cutoffs per vertical, and write the vertical name into the row, so a payroll title is never judged against a crypto one.

Broad consumer sites with high scores

A high score on a broad lifestyle site is easy to find and often hard to win a reply from. The audience can be huge and only loosely about the product, which dilutes the pitch even when the link data looks strong.

If a site scores green but its readers are not your buyers, move it to amber and send the pitch with a narrower angle. The score still counts, it simply stops deciding the outcome on its own.

Mistakes that flip a verdict after the refresh

Three slips cause most wrong verdicts. The first is trusting a green that was pulled before the latest refresh. The second is scoring a host that differs from the publication the contact actually writes for. The third is reading an anchor split from a sample too small to show a pattern.

Each slip is cheap to prevent. Date every pull, record the host that was scored, and require a minimum sample before reading the anchors. When a verdict flips after a refresh, log the change instead of overwriting it, because the flip is information about the list.