After this lesson you should be able to
- Apply hard disqualifiers that remove most of a raw list
- Score the survivors on signals that predict a real link
- Sort prospects into tiers with different handling
- Keep the rubric honest as the market and your site change
The pass that saves the campaign
A raw prospect list — intersect output plus operator results plus whatever the team gathered — is mostly not prospects. In my experience a disciplined qualification pass removes between 60% and 80% of it before anyone writes a word of outreach, and that figure has held steady across markets for a long time.
People resist this because a shorter list feels like less work done. It is the opposite. Every unqualified prospect costs you an email you will not get a reply to, a follow-up, a row in a tracking sheet, a small increment of sender reputation, and a slice of the attention you owed to a prospect who might have said yes. The cost is not zero and it is not linear: send enough unqualified mail and you damage deliverability for the whole list, which is how one bad campaign poisons the next three.
Do this in two distinct passes, and do not merge them. First, hard disqualifiers — binary, fast, unarguable, applied as a delete. Second, scoring — judgment applied only to what survived. Merging them turns a five-second decision into a five-minute one, three hundred times over.
One instruction that sounds trivial and is not: delete what fails, do not flag it. A column marked no is a column somebody will filter differently next month, and disqualified prospects have a way of walking back into the campaign.
Hard disqualifiers
These are binary. If a prospect meets any of them, it leaves the list, regardless of how good its metrics look.
| Disqualifier | How to check it | Why it disqualifies |
|---|---|---|
| Already links to you | Cross-reference your own referring domains | Pitching an existing linker damages a relationship you already have |
| No outbound editorial links anywhere | Open three pages and look | A site that never links out will not start with you |
| Sells links, or advertises sponsored posts | Check the site's own advertising or write-for-us page | The ask becomes a purchase; the link inherits that character |
| Publishes across unrelated verticals | Scan the last twenty posts for topical coherence | A site covering casinos, CBD and insurance is a placement network, not a publication |
| Not indexed | Search for a distinctive phrase from the homepage | An unindexed page passes nothing and signals a bigger problem |
| Dead or abandoned | Date of the most recent post or update | Nobody is home to make the edit you are requesting |
| Language or region mismatch | Read the site | You cannot pitch, and their readers cannot use you |
| Domain repurposed after expiry | Compare current content to archived versions | The history that made it look attractive belongs to a different site |
| Direct commercial competitor | Read what they sell | You are asking them to promote a rival |
| No contact route at all | The work from Lesson 4.4 | An unreachable prospect is not a prospect |
Run these in that order, roughly, because the cheap checks come first. Already-links-to-you is a spreadsheet join and takes seconds across the whole list. Domain repurposing needs a human looking at an archive and should only ever be applied to the survivors.
Signals worth scoring
What is left gets scored. Scoring is judgment, so keep the number of factors small — four or five — or the rubric becomes theater and people start reverse-engineering the score they wanted.
- Topical relevance to the destination page. Not to your site in general. A link about the wrong subject from the right industry is worth less than people assume. Score this highest.
- Evidence they link to sites like yours. Do their outbound links go to independent sources, or only to their own properties and advertisers? This predicts outcome better than any authority metric.
- Plausible mechanism. Is there a page you could realistically be added to, or a beat they cover where you could be a source? A prospect with no mechanism is a wish.
- Authority, as a tiebreaker. Trust Flow, Domain Authority or your metric of choice — used to rank within relevance, never to override it. Module 2 covers the limits of these metrics and they all apply here.
- Audience overlap. Would a reader clicking through be someone you want? A link that sends no plausible reader is worth having and worth less.
Weight relevance and mechanism above authority. I will take a Trust Flow 15 site whose readers are exactly my audience over a Trust Flow 55 general-interest site every time, and the data on link survival supports the instinct: links that make editorial sense to the publisher last, and links that do not get edited out.
A scoring scheme that survives contact with a team
Elaborate weighted rubrics collapse under real use. What works is three tiers with different handling and an explicit rule for what happens to each.
- Tier A — pitch individually. Strong relevance, clear mechanism, a named contact, evidence of linking out. Usually 10 to 20% of the qualified list. These get a genuinely bespoke email that references the specific page. If someone senior only has time to write ten emails a week, these are the ten.
- Tier B — pitch from a structured template. Relevant, plausible mechanism, contact found. The bulk of the list. Template means fixed structure with two or three genuinely specific sentences, not merge fields — Lesson 5.2 makes the distinction concrete.
- Tier C — hold or discard. Weak relevance, unclear mechanism, or a contact you are unsure of. Decide in advance what happens here and write it down.
My rule for Tier C is that it does not get emailed. It gets a different treatment entirely — a follow on social, a comment on something they published, an invitation to a webinar, or nothing. The instinct to email Tier C because the list looks short is the instinct that produces the campaigns this course exists to prevent.
Add one field that most systems omit: a one-line note on why this prospect is here. Not a score. A sentence. They maintain the county supplier list and we are not on it. That sentence is what the person writing the email actually needs, and a score does not contain it.
Keeping the rubric honest
A rubric written once and never revisited becomes superstition. Two habits keep it connected to reality.
Score the outcome, not just the prospect. When a campaign closes, look at which tier the links actually came from. If Tier A converted at the same rate as Tier B, your relevance judgment is not discriminating and the criteria need work. If Tier C converted at all, you were disqualifying something you should not have been.
Check your survivors, not just your wins. This is the part almost nobody does, and the data makes the case for it. In my own profile, the median referring domain stops linking after 1,080 days, and 49.6% of all links ever recorded are already lost. Survival scales sharply with quality: domains at Trust Flow 0 had a median survival of 859 days, while those at Trust Flow 61 and above lasted a median of 3,353 days — roughly four times as long. And 717 referring domains at Trust Flow 21 or above have stopped linking to me entirely.
The implication for qualification is direct. A campaign judged on links acquired at week twelve is judged on a number that will be roughly halved by year three, and the halving is not uniform — the weak prospects you let through are the ones that disappear first. Qualification is not just about response rate. It is about which of the links you win are still there in three years, and on that measure the case for cutting hard is much stronger than the week-twelve numbers suggest.
Questions
How much of a raw prospect list should qualification remove?
Typically 60 to 80%, and a pass that removes much less has usually not been applied honestly. The raw list contains sites that never link out, sites that link to everyone, dead sites, direct competitors and duplicates. Removing them is not lost work; it concentrates the outreach effort on the prospects that can actually say yes.
Should I disqualify a site purely on low authority metrics?
No. Metrics are a tiebreaker within relevance, not a gate on it. On my own profile 69.3% of referring domains are Trust Flow 0, on a site where no link was ever bought — low scores are simply what the ordinary web looks like. Disqualify on relevance, outbound linking behavior and whether anyone is home.
What should happen to the C tier?
Decide before the campaign starts and write it down, because the default is that somebody emails them when the numbers look thin. My rule is that Tier C never receives outreach. It gets non-email treatment — a follow, a comment, an invitation — or it gets discarded. Emailing weak prospects to hit a volume target costs deliverability you will need later.
How often should the qualification criteria change?
Review them at the end of every campaign against what actually converted, and again a year later against what actually survived. If your top tier converted no better than your middle tier, the criteria are not discriminating. If links from a category disappeared within a year, that category needs a harder standard or an exclusion.