After this lesson you should be able to
- Write a target set definition that names both the citer and the reason
- Identify the categories of site that plausibly link in your market
- Size a target set honestly instead of by wishful arithmetic
- Recognize the keyword-first prospecting error before it costs you a month
The keyword-first error
Here is the workflow I see most often. Someone picks the keyword the client wants to rank for. They put it into a link tool, export the top hundred ranking domains, and call that a prospect list. Then they wonder why the response rate is close to zero.
Think about what that list actually contains. It contains the sites currently beating you. Those sites are your direct competitors for the same search demand. Asking them to link to you is asking them to hand over the thing they are competing for. A small number will, because some publishers are genuinely editorial and some are indifferent to your niche, but as a definition of a target audience it is close to inverted.
The keyword-first error comes from a category confusion. A keyword describes what a searcher wants. It does not describe who publishes. The people who publish about a topic and the people who compete for traffic on a topic overlap far less than the tools imply, because the tools only measure the second group.
Work in the other direction. Start with the question: who, in the ordinary course of doing their job, would have a reason to point a reader at something I have? Answer that in prose before you open any tool. If you cannot answer it, you do not have a link building problem. You have a nothing-worth-citing problem, and the fix for that is upstream of this entire module.
The three questions that define a target set
A usable target set definition answers three questions in writing. I mean written down, in a document, in sentences — not held loosely in your head, because a loosely held definition quietly expands until it includes everything.
- Who already writes about this subject, for an audience that is not me? The qualifier matters. Trade press, practitioner blogs, association newsletters, university department pages, local news, hobbyist communities, adjacent-industry publications. Not your competitors' marketing blogs.
- What do I have that gives them a reason to point at me? A tool, a dataset, a definitive explainer, a free resource, a piece of original research, an unusual physical location, an expert who will comment on record. Something. If the honest answer is our homepage, stop.
- What is the specific page they would be linking to? Not the domain. The URL. If you cannot name it, either the asset does not exist yet or you are planning to ask strangers to link to a commercial page for no reason, which is the request that fails most reliably.
The output is a paragraph, something like: independent gardening writers and regional horticultural societies who cover rose cultivation, who would link to our pruning calendar because it is the only one organized by USDA zone. That paragraph is worth more than a thousand-row spreadsheet.
Categories of plausible citer
Once you have the definition, expand it into categories. Categories matter because each one needs a different pitch, a different qualification standard, and often a different person doing the outreach. Mixing them into a single undifferentiated list is how campaigns become generic.
- Practitioners in adjacent fields. Not your competitors — the people whose work touches yours. An accountant links to a payroll explainer; a wedding photographer links to a venue.
- Trade and vertical publications. Smaller circulation than mainstream press, far higher relevance, and usually a named editor who answers email.
- Organizations you already have a relationship with. Suppliers, customers, partners, trade bodies, the conference you sponsored, the charity you support. This is the least worked and most reliable category in most markets.
- Geographic and community sites. Chambers of commerce, local news, city guides, community organizations. Weak topical relevance, strong local relevance, and genuinely difficult for a national competitor to replicate.
- Reference, education and library pages. Course reading lists, library guides, department resource pages. Slow, and the ones worth having cannot be bought.
- Communities and associations. Member directories, forums with real editorial pages, professional bodies.
Write down which categories apply to you and which do not, and say why the ones you excluded were excluded. That record stops you re-litigating the decision every month.
Sizing the set honestly
People wildly overestimate how large a real target set is, and the overestimate is what produces the ten-thousand-row spreadsheet nobody ever works.
Some grounding from my own data. Across twenty-plus years of publishing at billhartzer.com, the profile contains 22,260 referring domains. Of those, only 446 — 2.0% — are Trust Flow 41 or above. Trust Flow is Majestic's measure of how trustworthy the sites linking to a page are, on a 0 to 100 scale, and 41+ is a reasonable working threshold for a domain that is genuinely established. Four hundred and forty-six domains over two decades works out to roughly twenty-two a year, from a site that publishes constantly, speaks at conferences, and gets cited by the industry press. Meanwhile 69.3% of all 22,260 referring domains are Trust Flow 0, on a profile where no link was ever bought.
Read those two numbers together and you get a realistic shape for a target set. The pool of genuinely good potential citers in any specific niche is usually in the low hundreds, not the tens of thousands. The tens of thousands exist, but they are the Trust Flow 0 tail — and that tail arrives on its own, as a byproduct of being visible, rather than through outreach.
So size accordingly. For most campaigns a well-defined target set of 150 to 400 domains, worked properly, beats a list of 5,000 worked badly. If your definition produces 5,000, the definition is too loose, not the market too large.
Writing it down before you search
Produce a short target definition document before you touch a tool. Mine has five parts and fits on one page.
- The citer. One paragraph, in prose, describing who they are and what their job is.
- The reason. What we have that makes the link editorially defensible from their side.
- The destination. The specific URLs that will be linked to.
- The inclusion criteria. The minimum a site must meet to enter the list at all.
- The exclusions. Categories explicitly out of scope, with the reason. Ours usually excludes sites that publish sponsored posts as their primary business model, any site whose outbound links are predominantly to unrelated verticals, and anything already carrying a link to us.
The exclusions section is the one people skip and the one that pays. Writing exclusions down converts a judgment call you would otherwise make three hundred times, inconsistently, into a rule you apply once.
It also gives you something to hand a client, and it makes disagreements happen at the right time. A client who objects to your definition in week one has saved you a month. A client who objects to your prospect list in week five has cost you one.
Common mistakes
Defining the set by metric alone. Domains above a certain authority score is not a target set. It is a filter you apply to a target set. Relevance comes first, then metrics narrow it. Reversing the order produces lists full of high-scoring sites with no reason on earth to mention you.
Defining it by what is easy to find. If your definition suspiciously matches the output of one tool, you have let the tool define your market. Every tool has coverage gaps, and the gaps are not random.
Confusing audience with citer. Your customers read consumer blogs. Your citers may be trade publications your customers have never heard of. These are different lists and they need different assets.
Skipping the destination page. The single most common structural failure. A campaign with no defined link destination becomes a campaign asking for homepage links, and homepage links from strangers are the hardest ask in this business.
Never revisiting it. A target definition written eighteen months ago describes a market that has moved. Re-read it at the start of every campaign cycle and change what is wrong.
Questions
Should competitors ever be on a prospect list?
Occasionally, and only when they are competitors in the search results rather than in the market. A publication that ranks for your terms but sells advertising rather than your product is a legitimate prospect. A company selling the same thing to the same buyer is not, and pitching them wastes an email and a little credibility.
How many prospects do I need for a first campaign?
Fewer than you think. A tightly defined set of 150 to 400 domains gives most campaigns enough runway for a quarter, once qualification has removed the majority. If you cannot find 150 plausible citers, that is useful information: it usually means the asset is not distinctive enough, and more prospecting will not fix it.
What if my product is genuinely boring?
Then your citers are not consumer blogs, and pretending otherwise wastes months. Boring products are cited by trade press, professional associations, procurement and comparison resources, and the suppliers and customers already in your commercial orbit. That set is smaller, less contested, and considerably more durable.
Can I buy a prospect list instead of building one?
You can buy lists, and I would not. A purchased list encodes somebody else's definition of relevance, which is usually no definition at all, and it has almost certainly been sold to others in your niche. The contacts have been emailed repeatedly by people with the same template. You inherit the fatigue without the relationship.