How to Run a Pay-Per-Meeting Pilot: The 30-Day Test That Proves or Kills the Model

Pay-per-meeting sounds like the easiest model in B2B sales to evaluate. You pay for meetings that happen, so either the meetings are good or they are not. In practice it is one of the easiest to evaluate badly. The first qualified meeting lands in week two, the AE says the prospect was "fine but early", the second one no-shows, the third is a great fit, and by week five everyone has an opinion and nobody has a number. The deal gets renewed or cancelled on mood.

This guide is the fix. It is the 30-day pilot structure we ask every new company on Bountii to run, written so you can use it with any provider, agency or marketplace. It assumes you already know what the model is; if not, start with the pay-per-meeting guide. It also assumes the contract side is handled, which the pay-per-appointment contract guide covers clause by clause. What follows is purely about running the test so that on day 30 the decision makes itself.

Thirty days is long enough for a provider to source, book and hold a real sample of meetings, and short enough that nobody can hide behind "ramp". A retainer agency will ask for 90 days because its model needs that long to show anything. A pay-per-meeting provider does not get that excuse: there is no ramp cost on your side, so the only question is whether the meetings arrive and whether they are the right ones.

The pilot matters because pay-per-meeting shifts risk in a way most buyers have never measured. Under a retainer you pay for activity and hope for outcomes, so you judge the vendor on effort. Under pay-per-meeting you pay for outcomes, so you have to judge the outcomes, which means defining them in advance, counting them honestly, and comparing them to what you pay today. Teams that skip that step end up applying retainer instincts to an outcome contract and conclude, wrongly, that it "did not do much" because the inbox was quiet.

Every pilot that produced a clean decision had these four numbers on a single page before any outreach started. Every pilot that ended in an argument was missing at least one.

One: your current cost per qualified held meeting. Not cost per lead, not cost per SDR, not agency fee per month. Take everything you spend on outbound in a quarter, including salaries, tools, data and management time, and divide it by the number of meetings that were held and that your AEs accepted as qualified. For a team with one fully loaded SDR at $100,000 to $130,000 producing eight to twelve qualified held meetings a month after ramp, the honest number is usually $900 to $1,400. If you do not have the number, estimate it and label it an estimate. The cost to hire an SDR has the line items if you are building it from scratch.

Two: the target cost per qualified held meeting for the pilot. This is the price you will pay per meeting, adjusted for the show rate and qualification rate you expect. A $400 bounty with an 85 percent show rate and an 80 percent qualification rate is a $590 qualified held meeting. Set the target below number one by a margin you would act on, usually 30 percent or more.

Three: the show-rate floor. Below what percentage of booked meetings actually being held do you stop the pilot early? For warm-sourced meetings we use 75 percent; for cold-sourced agency meetings, where no-shows commonly run 20 to 40 percent, 60 percent is already generous.

Four: the qualification pass-rate floor. What share of held meetings must pass your written definition for the channel to be worth scaling? We use 70 percent. Below 50 percent the definition was either unclear or ignored, and either way the pilot has already told you what you need to know.

The most common way to waste a pilot is to make it look like the whole go-to-market. Pick one ICP segment you already close well, one product or offer, and one geography. Give the provider a list of 20 to 60 target accounts or a tight description that produces one. A pilot across three segments with 300 accounts generates meetings nobody can compare to each other, and "the enterprise ones were great but the SMB ones were weak" is not a decision, it is a shrug.

Set a meeting target that produces a sample, not a trophy. Five to twelve qualified held meetings in 30 days is enough to see a pattern. If your provider says it can deliver 40 in the first month from cold, be suspicious of the definition, not impressed by the volume.

Pricing the meeting is its own subject and B2B appointment setting cost in 2026 walks through it by title and segment. The short version for a pilot: pay slightly above the market rate for the seniority you want. A pilot priced to save money attracts the providers and hunters who cut corners, and then you learn nothing about what the model does at a fair price.

The pilot lives or dies on a document that fits on one page. It states who counts (titles or seniority, company size, industry, region), what must be true (a live need, a trigger, a budget owner present, whatever your qualification criteria are), what the meeting is (length, format, who from your side attends), and what disqualifies it (a reseller, a job seeker, a student project, an existing customer, a competitor). It also names the reviewer, the review window and what happens to disputed meetings.

Write it with your AEs, not for them. If the people taking the meetings did not agree to the definition, they will reject meetings that match it and accept meetings that do not, and your pass rate becomes a measure of mood. What makes a meeting qualified has a framework and examples you can adapt.

Share the full page with the provider or the hunters before the first outreach. Hiding it, on the theory that they will game it, guarantees the opposite problem: they will guess, and guess wrong.

Week one is setup and nothing else. Definition agreed, accounts listed, bounty or price fixed, calendar links and intake form live, AEs briefed on what is coming and how to score it, review owner named. On a marketplace the bounty goes live in this week and hunters start claiming accounts; with an agency the first sequences start. Expect zero meetings and do not treat that as a signal.

Week two is first meetings and review discipline. The first two or three meetings matter less for what they are than for how you handle them. Review each within 48 hours against the one-pager, record the result on the scorecard with a one-line reason, and send feedback to the provider the same day. A pilot where the first meetings are reviewed a week late has already lost its calibration loop.

Week three is volume. This is when the sample forms. Keep reviewing within 48 hours, keep the AE feedback flowing, and watch two things: the show rate on booked meetings and whether the qualified ones cluster around a particular kind of account or trigger. That clustering is the most valuable output of the whole pilot, because it tells you what to scale.

Week four is scoring and the decision. Stop booking new meetings two or three days before day 30 so that everything booked can be held and reviewed inside the window. Fill the scorecard, compare to the four numbers, and decide. Do not extend "to get a few more data points" unless a floor was missed for a reason you can name; extensions are how pilots become quiet renewals.

Booked-to-held rate: meetings held divided by meetings booked. Pass at 75 percent or above for warm-sourced, 60 percent for cold-sourced. This is the first thing to go wrong with cold channels and the first thing to check.

Held-to-qualified rate: meetings that passed the definition divided by meetings held. Pass at 70 percent. Below that, either the definition was unclear or the provider is not targeting it.

Cost per qualified held meeting: total paid divided by qualified held meetings. Pass if it is at or below the target you set before day one, and a clear win if it is 30 percent or more below your current number.

Time to first qualified held meeting: days from go-live to the first meeting that passed. Pass at 14 days or less. Warm-sourced channels usually land it in seven to ten; longer than three weeks on a 30-day pilot means the second half is the whole sample.

AE acceptance: the share of qualified meetings your AEs voluntarily took a second call on, or moved to a next step. Pass at 40 percent. This is the number that separates "met the definition" from "worth having", and it is the one most buyers never record.

ICP concentration: the share of qualified meetings inside the single segment you scoped. Pass at 80 percent. Drift here means the provider is filling the quota from wherever it can.

Dispute rate: meetings you rejected that the provider contested. Pass at 10 percent or below. A high dispute rate with a high pass rate means the definition has a gap; a high dispute rate with a low pass rate means the relationship will not survive scaling.

AE time per meeting: minutes your team spent per qualified meeting on prep, the call and follow-up, including the rejected ones. There is no pass mark; record it so that the cost per meeting you compare on day 30 includes your own side of the ledger.

Judging on volume instead of yield. Eight qualified held meetings at $500 each is a better result than 25 booked meetings at $250 of which nine were held and five qualified, yet the second looks busier. The scorecard exists to stop this comparison being made by feel.

Not briefing the AEs. The person taking the meeting decides whether it was qualified. If they do not know a pilot is running, what the definition says, or that they are expected to score it within 48 hours, the pass rate measures their patience on the day.

Scoping an ICP you do not already close. A pilot is a test of the channel, not a test of a new market. If you have never sold to the segment, a low AE-acceptance rate tells you about the segment, and you will blame the provider.

Running two providers on the same accounts. It feels like a fair comparison; it produces two sets of outreach to the same people, lower response for both, and a dispute about who sourced which meeting. Split accounts cleanly if you want to compare.

Skipping the review window. If every booked meeting is billed automatically, the pilot cannot distinguish the provider's judgment from yours. Insist on a window, even a short one, and use it. On a marketplace with a built-in review step this is already handled, which is one reason pilots there tend to produce cleaner numbers.

Scale when all four floors were cleared and the cost per qualified held meeting beat your current number by the margin you set. Scaling means widening the account list inside the same segment first, then adding a second segment with its own one-pager, not doubling the budget across everything at once.

Adjust when the yield numbers passed but the cost did not, or the cost passed but AE acceptance was weak. The first usually means the price was set too high for the seniority delivered and can be corrected; the second means the definition is letting through meetings that match on paper but not in the room, and needs a line added. Run a second 30 days with the change, not a 90-day extension.

Stop when a floor was missed and you cannot name a fixable cause, or when the dispute rate says the provider does not share your view of what qualified means. Stopping a pay-per-meeting pilot costs you nothing beyond the meetings you already paid for, which is the point of the model. Write up the scorecard anyway; it becomes the baseline for the next provider.

On Bountii a pilot is a bounty. You post the one-page definition, the target accounts and the price per qualified held meeting, from $200, under a Growth subscription at $199 a month that covers ten open accounts. Hunters with existing relationships in those accounts claim them, with a 14-day exclusive window per claim, and book meetings through a warm introduction rather than a cold sequence, which is why show rates on the platform sit well above the cold-channel range.

Every meeting arrives with the hunter's context and goes into a 48-hour review where your team scores it against the definition. Qualified meetings release the bounty from the funded pool; rejected ones cost nothing. The scorecard above maps directly onto what the dashboard records, so day 30 is a read-out rather than a reconstruction. Companies can also split a pilot between an agency and a bounty on separate account lists, which is the one form of head-to-head comparison that produces a clean answer.

If you want the one-page definition template and the scorecard as a sheet, book a demo and we will send both and help you set the four numbers.