Most comparisons of link prospecting tools put Ahrefs, Pitchbox and Hunter in one table, as though they were alternatives to each other. They are not. They do different jobs, in sequence, and the reason the category is confusing is that no single product covers the sequence.
Split it into four jobs and the picture resolves quickly. Three of them are well served. One is not, and it is the one that decides whether the other three were worth doing.
The four jobs, and who does them
| Stage | The question it answers | What is generally used for it |
|---|---|---|
| Discovery | Which pages could conceivably link to me? | Ahrefs, Semrush, Majestic, Moz, search operators, bought lists |
| Qualification | Which of these pages can actually carry the link I want, today? | A person, opening pages one at a time |
| Outreach | Who do I contact, what do I say, and did they reply? | Pitchbox, BuzzStream, Respona, Hunter for contacts |
| Monitoring | Is the link still there, still followed, still pointing where it should? | Backlink monitors, or a spreadsheet and a reminder |
Discovery and outreach are mature markets with several good products in each. Monitoring is smaller but solved. Qualification is the stage that is still done by hand, and the tools on either side of it both quietly assume somebody else did it.
Why the gap is expensive rather than merely annoying
The assumption is that a list arriving from discovery is roughly usable, and that a bad row costs you one wasted email. Neither holds once you open the pages.
We took a purchased list of 1,001 rows and checked every row by hand, fifty at a time, in September 2026. The full method and the limits are in the study. The short version:
- 999 distinct URLs across 467 hosts. 69.9% of rows shared a host with another row, so a thousand-row list was really 467 sites.
- 788 rows, 78.7%, could be read at all. The rest timed out, blocked, or returned nothing usable.
- 183 rows, 18.3%, had an open comment form. 818 had none of any kind.
- 37 rows, 3.7%, would have carried a followed link. So 96.3% of the list could not do the thing it was sold to do.
- 10 rows out of 1,001 were followed, readable and low spam risk at the same time.
At roughly two minutes a page, which is fair once you include the ones that hang and the ones you have to view source on, qualifying that list by hand is about 33 hours of work to arrive at ten rows worth an email. That arithmetic is why the stage gets skipped, and why skipping it is not free.
Sorting by a domain metric does not substitute for it
The usual shortcut is to sort the discovery export by Domain Rating or Domain Authority and work down from the top. On this list that ordered the rows backwards.
Not one of the 285 rows at DR 50 or above could carry a followed comment link. 34 of the 37 that could sat at DR 10 or below. The cause is that hosted subdomains inherit their platform’s score: 38 rows were *.blogspot.com, all at DR 95, all open, none followed. Another 76 shared one forum platform at DR 61 and 48 shared a church media host at DR 68. Those three platforms were 162 rows, 16.2% of the list, wearing three scores between them.
The full argument, including where the metric is still the right one to use, is in why Domain Rating is not enough.
What to ask of anything claiming to do qualification
Whatever you use, including a person with a browser, these are the questions that separate an observation from a guess:
- Does it check the exact URL, or the domain? Comment forms close per post. A theme update removes one. A plugin turns every comment link nofollow without changing a word of the copy. None of that moves a domain metric.
- Does it read the link attribute, or infer it from the platform? A platform default is an expectation. What the page already gives its commenters is an observation. They belong in different columns.
- Will it say it does not know? An open form with no existing comment links is genuinely unreadable either way. A tool that fills that gap with the likely answer has put a guess where you will read an observation.
- Does it tell you how many links a judgement came from? One nofollow link is weaker evidence than forty, and the difference belongs on screen.
- Does it date the answer? All of this is true on the day it was read. A row without a date is a row you will re-check anyway.
Where WebDiagnosis sits, and what it will not do
We built the Backlink Prospect Checker for the qualification stage only, and only for comment opportunities. You give it a list you already have. It opens every exact page and reports whether a comment form is open, what the existing comment links carry and how many of them that judgement came from, the Ahrefs DR, and how spammy the thread looks. Every field’s method is written out, including the conditions under which it returns Unknown.
What it does not do is most of the category:
- It does not discover prospects. Bring the list from Ahrefs, Semrush, a vendor or a search operator.
- It does not find contacts, send email or track replies. That is what the outreach platforms are for, and they are good at it.
- It does not post comments, sign in anywhere or answer a CAPTCHA. It reads public pages.
- It does not cover guest posts, directories, forums, profiles or editorial placements. Comment opportunities are the whole of the current scope.
- It does not monitor links after placement, and it cannot tell you whether your comment will be approved. Moderation is a person.
So it replaces nothing in your stack. It sits in the gap between the export and the outreach, which is the gap the rest of the category leaves open.
Working out whether you need it
You probably do not, if your lists are short, hand-built and you already open every page. The stage is only expensive at volume.
It starts paying when you are handed a list you did not build, which is where the two worked examples come in: qualifying an export you pulled yourself, where the job is mostly cutting down, and checking a list you were sold, where the first question is whether the seller opened any of the pages at all.
If you would rather do the whole thing by hand, the qualification sequence is written out in the order that rejects most cheaply first. It is the same order the tool runs in, and it works fine in a spreadsheet and a browser. There is also a full sample report if you want to see the columns before signing in, and the free allowance is 500 checks on first use, which is five scans of 100 URLs.