Consumer link building has one reader. B2B link building has a committee, and the committee does not read together.
Gartner's number for a mid-market software purchase is six to ten people. Our audits land in a similar range: five to eleven with real influence, of whom two or three can stop the deal outright. They arrive at different moments, they need different things, and the publications they trust have almost no overlap.
Almost every B2B link programme is built as though there is one buyer. That is the structural reason so many of them produce respectable link counts and no pipeline.
Who is actually on the committee
The composition varies, but four roles show up in nearly every mid-market software purchase and each one reads somewhere different.
The end user or practitioner. Whoever will use the thing daily. They start the search, they form the shortlist, and they are the only one who reads practitioner content. They live in newsletters, subreddits, Slack communities and the trade press for their function.
The economic buyer. The budget holder, usually a director or VP. They arrive late, they read almost nothing in your category, and they read business and functional-leadership publications. They are looking for reassurance that this is a normal choice.
The technical gatekeeper. Security, IT, data. They can veto and rarely champion. They read compliance and security press, vendor documentation, and standards bodies. Nothing you place in a marketing publication reaches them.
The finance or procurement reviewer. Arrives last, cares about contract terms and total cost. Reads almost nothing you can influence — but is influenced heavily by whether the other three have converged.
Map targets per persona, not per company
This is the practical change, and it is a change to the target list rather than to the outreach.
Take the four roles that matter for your product. For each, answer three questions with evidence rather than assumption: what do they read, what question are they asking at the moment they encounter you, and what would have to be true for them to stop objecting?
Then build a column of target publications per persona. The columns will barely overlap, and that is the finding. A single "top publications in our space" list is almost always the practitioner's list with a couple of general-business titles bolted on, which means the economic buyer and the gatekeeper are unserved.
The way to answer the first question honestly is to ask. Five customer conversations produce a better list than any tool, and they surface things no database indexes — the newsletter with four thousand subscribers that every operations lead in the category reads on Monday morning.
Different placements make different arguments
The same link in the same publication is worth different amounts depending on which argument the page is making, and each persona needs a different one.
For the practitioner: how it works, what it replaces, what breaks. Comparison pages, teardowns, workflow write-ups. This is where partial-match anchors sit naturally and where a placement can be genuinely useful rather than merely present.
For the economic buyer: that this is a category, that the category is legitimate, and that you are a normal choice within it. Category definition pieces, analyst-adjacent commentary, funding and market coverage. The link matters less than the association.
For the gatekeeper: that you will not create a problem. Security posture, compliance, integration behaviour, data handling. Almost always a different set of publications, and almost always ignored because it does not feel like marketing.
For everyone at once: the round-up and the alternatives page. These are the placements that produce pipeline independently of rankings, because a buyer at any stage lands on them from a search and forms an opinion in ninety seconds.
Where the committee model changes the arithmetic
Three consequences, all of which affect what you should buy rather than how you should pitch.
The reachable list is larger than a single-persona list suggests, and thinner per persona. A category might have a hundred and twenty credible publications in total but only fourteen that the technical gatekeeper actually reads. Volume planning has to happen per column, not in aggregate, or the gatekeeper's column gets exhausted in month two and quietly abandoned.
Some personas are unreachable by link building, and saying so is useful. Procurement is rarely influenced by anything you can place. Trying anyway is a common way to spend a quarter's budget with nothing to show.
Attribution gets harder in a specific way. A placement read by the practitioner in March shows up as a demo request. A placement read by the economic buyer in September shows up as a deal that did not stall. The second is invisible in any analytics tool and it is often worth more. This is one reason the reliable early indicator of a working programme is qualitative — sales starts sending placements to prospects — rather than a chart.
Doing the mapping
An afternoon, and it is worth doing before you commission anything.
Write the four roles down the left of a page. For each, list what they read — from customer conversations, not from a tool. Then mark which of those publications you already appear in. The empty rows are the finding, and in most B2B SaaS companies the empty rows are the economic buyer and the gatekeeper, because every previous programme optimised for the practitioner.
Then decide, deliberately, how to split the volume. There is no correct ratio. There is only the difference between a split you chose and a split that happened because the practitioner's publications were the easiest ones to reach.
That difference is most of what separates a programme that produces a link count from one that produces deals which stop stalling in the second meeting.