Commercial structure gets treated as an administrative detail. It is not — it determines what behaviour is rewarded, and suppliers respond to incentives like everyone else. Worth choosing deliberately.
Per-link pricing
You pay: a fixed amount per placement delivered.
The supplier optimises for: placement count, and therefore for the cheapest placements that pass whatever check exists.
This is the most transparent structure and the one with the sharpest failure mode. Because revenue is per unit, there is a direct incentive to source cheaply and to define "acceptable" generously — particularly in a month where volume is behind.
It works when the specification is tight and enforced. It fails when quality is assessed subjectively, because the pressure always runs one direction.
| Good for | Bad for |
|---|---|
| One-off needs | Programmes needing strategy |
| Buyers with in-house vetting | Buyers relying on supplier judgement |
| Filling a specific gap list you already own | Anything requiring the supplier to say no |
Monthly retainer
You pay: a fixed monthly fee for an agreed scope.
The supplier optimises for: retention — which is a genuinely better incentive, because keeping you for eighteen months requires results rather than throughput.
The characteristic failure is opacity rather than corner-cutting. Scope flexes quietly, deliverables shift, and by month four nobody can reconstruct what was actually delivered for the money.
The fix is not to avoid retainers. It is to specify the scope numerically inside one: a monthly range of verified placements, a minimum standard per placement, and reporting that can be spot-checked.
Project pricing
You pay: a fixed fee for a defined piece of work with a defined end.
The supplier optimises for: completing the deliverable efficiently.
Excellent for bounded work: a profile audit, a disavow programme, a linkable asset build, a digital PR study. Poor for acquisition, which is inherently continuous — a project-priced "50 links" is per-link pricing with a bulk discount and the same incentives.
The hybrid we run, and why
Monthly fee, with a stated placement range, billable only on live-and-indexed.
The monthly fee removes the incentive to maximise count. The stated range removes the retainer's opacity. The billing gate means placements that fail the checks cost us rather than you, which puts the specification on our side of the ledger.
Shortfalls carry forward rather than being refunded — which is the concession the client makes in return, and a fair one, because editorial timelines genuinely are irregular.
The clauses that matter more than the model
Four terms that determine outcomes regardless of which structure you choose.
When a placement becomes billable
Live, indexed and verified. Not outreach sent, not "secured", not published-but-unindexed. This single clause reallocates most of the execution risk and it is the one most buyers never ask about.
What happens to a shortfall
Carry-forward is reasonable. A refund is cleaner but creates pressure to hit numbers. What is not reasonable is a shortfall that simply disappears.
The replacement warranty
Twelve months, defined triggers, replacement rather than credit. Links decay predictably; a structure that ignores this has moved a known cost onto you.
Minimum term and notice
Three months is reasonable — outreach sent in month one publishes in month two and is measurable in month four. Twelve months with no exit is a term to decline regardless of price.
Choosing
| Your situation | Structure |
|---|---|
| You own a vetted gap list and want it filled | Per-link, with a written spec |
| You want predictable monthly output you can audit | Monthly fee with a stated range |
| You need a bounded piece of work | Project |
| You need strategy owned as well as execution | Retainer, with numeric scope inside it |
| You are testing a supplier | Three months, monthly fee, small tier |
The question to ask about any structure: what does this make them do in the third week of a month when they are behind? Every commercial model has an answer, and it is worth knowing yours before you sign it.
One structure to avoid
Performance pricing tied to rankings. It sounds aligned and it is not. Rankings depend on your page, your technical health, your competitors and factors nobody controls — so a supplier paid on rank has an incentive to target keywords that are easy rather than valuable.
We have seen programmes where the agreed keyword list quietly drifted toward low-volume terms nobody searched. Everyone hit their numbers. Nothing happened commercially.