Pay per appointment vs retainer vs revenue share

Every pricing model is a risk allocation

Strip away the sales language and a lead generation contract answers one question: if the campaign underperforms, who eats the loss? A retainer says you do. A performance model says the agency does. Pay per lead and pay per appointment say it depends on how tightly the deliverable is defined, which means the definition, not the headline price, is the real negotiation. Keep that lens on the four models below and most vendor claims evaluate themselves. If you came for dollar figures, those live on the cold email agency cost page.

Monthly retainer: you pay for activity

How it works: a fixed monthly fee, usually on a multi-month minimum term, in exchange for the agency running your outbound end to end: list building, copy, sending infrastructure, reply handling. Who carries the risk: you. The invoice is identical in a month that produces twelve meetings and a month that produces zero.

The failure mode is structural rather than moral. A retainer pays for activity, so activity is what gets optimised: sends made, sequences launched, reports delivered. All of it can be done diligently and in good faith while the pipeline stays empty. There is also a floor under the model worth respecting: real volume costs real money. At 50,000 emails per month, the sending infrastructure alone, mailboxes, sequencer, data and verification tools, runs about $2,841 per month at list prices, before anyone is paid to think. A retainer priced below the tooling it claims to include is telling you something.

Pay per lead: volume beats quality unless the contract says otherwise

How it works: a fixed price per lead delivered. Who carries the risk: nominally shared, in practice tilted toward you, because the vendor controls the definition of the unit you are buying.

The failure mode is volume over quality: whatever the contract accepts as a lead is what gets maximised. The funnel maths shows how much room that leaves. At the realistic benchmark we publish in our ROI calculator, 100,000 emails reach 50,000 unique leads at two touches and produce about 1,000 replies, of which 15 percent, roughly 150, are positive. A vendor counting any reply as a lead invoices you almost seven times as often as one counting positive replies only, on the identical campaign. Before comparing per-lead prices between vendors, compare definitions; the price difference is usually just the definition difference wearing a discount.

Pay per appointment: the meeting definition becomes the negotiation

How it works: a fixed price for each sales meeting the vendor books. Who carries the risk: the vendor, on paper. In practice the meeting definition becomes the negotiation, and that is where risk flows back to you.

A booked meeting is a calendar entry. It is not a held meeting, and it is not a qualified one. Across the three scenarios in our ROI calculator, show-up rates run from 65 percent (conservative) through 75 percent (realistic) to 85 percent (aggressive). Take the realistic case: 100,000 emails, reaching 50,000 unique leads at two touches, yield about 38 booked meetings, of which about 28 are actually held. If the contract bills on booked, roughly one meeting in four that you pay for never happens. The questions that decide whether this model works for you are all contractual: does a no-show get replaced or credited, who verifies the prospect matches your ICP, and can the vendor loosen qualification to hit a monthly quota. A vendor that resists writing the definition down has already answered.

Revenue share: the agency bets on closing

How it works: the agency takes an agreed percentage of revenue from deals it sources, usually alongside a lean fixed amount that covers hard costs. That is the model we run: the fixed portion covers the dedicated sending infrastructure, the bulk of compensation comes from the revenue share, and the percentage and attribution window are agreed up front. The full mechanics, attribution and edge cases have their own page: how revenue share works.

Who carries the risk: mostly the agency, which funds months of work that pays little unless deals close. Be honest about what that implies, in both directions. First, revenue share is not automatically the cheapest option: if the campaign works and you close well, the absolute dollars can exceed what a retainer would have cost. What you buy is alignment and a capped downside, not a discount. Second, the model only functions when the agency genuinely expects to close, which cuts two ways: agencies on this model screen clients hard and decline weak offers, and the model still fails if your own sales motion cannot convert, because at realistic benchmarks about 20 percent of held meetings close, and no pricing model rescues a close rate of zero.

The four models side by side

How to choose

Three questions sort most situations. First, what is the worst case you can absorb? If a wasted quarter of retainer payments would hurt, prefer a structure where the vendor carries more risk; our no-retainer lead generation page covers what that looks like in practice. Second, is the billable unit written down? The contract should define the unit (lead, booked, held, or closed), the no-show policy and the attribution window before the first invoice. Third, does the vendor act like it believes its own model? A performance agency that accepts every client is not pricing risk, and a retainer agency that will not report meetings held is telling you what it optimises. Our own model, and what gets agreed on the intro call, is on the pricing page.

Common questions

Is pay per appointment cheaper than a retainer?

Not automatically. You typically pay on booked meetings, and show-up rates in our published scenarios run 65 to 85 percent, so part of what you pay for is never held unless the contract bills on held meetings. The comparison turns on the per-appointment price, the meeting definition and the no-show policy.

What should count as a billable appointment?

A held meeting with a contact who matches written qualification criteria, not a calendar entry. Get the qualification criteria, the no-show and reschedule handling, and the replacement or credit policy into the contract before the first invoice.

Is revenue share always the cheapest model?

No, and any pitch that says so should worry you. When a campaign performs, a percentage of revenue can exceed what a flat retainer would have cost. Its value is a capped downside and an agency paid for outcomes rather than for the invoice.

Which model should a small company pick?

The one whose worst case it can absorb. A vendor carrying real risk screens clients before accepting them; one that signs everyone is charging for optimism. Actual cost ranges are on the agency cost page.