How to work out your real cost per booked meeting on LinkedIn

What cost per booked meeting actually measures

It measures one thing: what you paid for one held conversation with someone who fits your ideal customer profile. The formula is total channel cost divided by meetings held. Both halves are usually recorded wrong, which is why two companies running near-identical campaigns can report wildly different costs and both be sincere.

Start with the denominator, because it is where most of the self-deception lives. A meeting that was booked and then no-showed is not a meeting. A meeting rescheduled twice and finally held counts in the month it happened. Count held conversations with people who could plausibly buy.

Then the numerator. Everything that only exists because this channel exists belongs in it:

• The vendor retainer, or your in-house SDR's fully loaded salary for the share of their week spent on LinkedIn
• Any seats billed to you directly — Sales Navigator, data enrichment, email verification, a scheduling tool
• Fees for any managed or rented profiles doing the sending
• The hours your own people spend on reply handling, call prep and reporting, priced at what those hours cost you
• Anything you bought once for this channel, spread across the months it will serve

Leave out costs you carry anyway — the CRM licence, the website, the AE who would otherwise be taking an inbound call. And leave out meetings that came from referrals in the same month, or you will credit LinkedIn with pipeline it did not create.

The chain from spend to a held meeting

Cost per meeting is an output, never an input. Five stages produce it, and each one fails in its own way, so the chain is worth writing down before the total is.

Multiply your send volume by those four rates and you get meetings. That is the whole equation: meetings = sends × acceptance rate × reply rate × positive rate × booking rate. Cost per meeting is then your monthly spend divided by that result.

Do the arithmetic by hand once, with your own data. If 100 invitations produce 40 accepted connections, your acceptance rate is 40 divided by 100. If those 40 connections produce 6 replies, your reply rate is 6 divided by 40. Carry it through to held meetings and you have a chain you can improve one link at a time. Do not import anyone's benchmark into this — including ours. The chain is only useful because it reads your account, your list and your offer.

If you do not have this data yet, the chain doubles as a reporting spec. Any provider running managed LinkedIn outreach tracks all five stages; one who can only report meetings cannot tell you why the number moved.

Carrying the number through to cost per customer

A cost per meeting is not a decision. A cost per customer is, and it takes one more step.

1. Pull the last twelve months of first meetings your team held and count how many became customers.
2. Use outbound-sourced meetings only. Inbound converts differently, and blending the two will flatter outbound.
3. Divide meetings by wins to get meetings needed per customer.
4. Multiply that by your cost per meeting. That is your cost per customer for this channel.
5. Compare it to the gross profit on a first contract — not the contract value, because contract value includes delivery cost you still have to pay.

If one in five first meetings becomes a customer, you need five meetings per customer, so your cost per customer is five times your cost per meeting. Whether that is affordable depends on one thing you already know and no vendor does: what a customer is worth to you in gross profit, and how long they stay.

A channel is affordable when the first-year gross profit on one customer comfortably covers the cost of the meetings it took to win them — with enough left over that you would run it again.

Decide the payback period before you sign. A cost per customer recovered inside the first invoice is a different decision from one recovered in month nine, even where the arithmetic ends in the same place.

Why the win rate assumption dominates the answer

Change your win rate assumption and the output swings further than any price negotiation could move it.

That right-hand column moves by a multiple; a vendor's price, negotiated hard, moves by a fraction. So the first question is not "is this price good" — it is "what did our last twenty outbound first meetings convert at".

Two honest cautions. A meeting sourced from cold outreach converts lower than one sourced from a referral, so a mostly-referral history will make you over-forecast. And if you have no outbound history at all, you cannot calculate this yet — you can only run a defined test and treat the first quarter as the cost of learning the number. Budget it as a test rather than a forecast, because an invented win rate in a board deck becomes a real problem two quarters later.

Sanity-check the scale as well. If your meetings-per-customer figure implies more first meetings a month than your target list contains people, the problem is not the price of outreach — the segment is too small for this motion. Our free calculators run that check before you commit to a retainer.

Reading the number when it comes out badly

A bad cost per meeting is a symptom. The chain tells you which stage caused it, and each stage has a different fix — which matters, because the usual reaction is to change the message when the message was never the problem.

The last row is the one buyers get wrong most often. If people are accepting, replying and showing up but never buying, outreach is doing its job and something after the handoff is not. Cancelling the channel there fixes nothing.

The first two rows are usually a profile and list problem rather than a copy problem: a sending profile with a thin history and a vague headline suppresses acceptance before a word of your message is read, which is why profile work belongs before volume. And when the gap sits between positive replies and held meetings, the fix is operational speed — the gap appointment setting exists to close.

What to ask a vendor once you have your own chain

With the chain in hand, vendor conversations get shorter and much more useful. Ask these, in this order:

1. Which stages of the chain do you own, and which stay with my team?
2. Do you report meetings booked or meetings held? If booked, who chases the no-shows?
3. Will I see sends, acceptances, replies and positive replies every month, or only the final number?
4. What is your attribution window — if a meeting books six weeks after the first message, which month does it belong to?
5. When a stage stalls, what do you change first, and how long before we see whether it worked?
6. What does the first sixty days look like, given that conversations opened in week three book in week five or six?

Anyone who has run this properly answers all six without hesitation. Our lead generation service sets out the division of labour, and pricing shows what sits inside a retainer versus what is billed separately — which matters, because a seat you pay for directly still belongs in the numerator.

Key takeaways

• Divide total channel cost by meetings held, not meetings booked — no-shows are spend without output.
• Include every cost that exists only because of this channel, including seats billed straight to you.
• Meetings = sends × acceptance × reply rate × positive rate × booking rate. Track all five stages, not just the last.
• Multiply cost per meeting by your own meetings-per-win figure, then compare to first-contract gross profit.
• Your win rate assumption swings the result by a multiple; a negotiated price swings it by a fraction.

Frequently asked questions

Should I count meetings booked or meetings held?

Held. A booked meeting that no-shows consumed the same spend and produced no conversation, so counting it hides a real problem. Track both if you like — the gap between them is a useful measure of how well the booking and reminder process works.

How long before my cost per meeting is meaningful?

Not in the first month. Conversations opened in week three typically book in week five or six, so dividing month one's spend by month one's meetings measures setup, not performance. Use a trailing window once the pipeline is steady, and judge the channel on a full quarter.

Should I include my own team's time in the cost?

Yes, for the hours that exist only because of this channel — reply handling, call prep, reporting. Exclude costs you carry regardless, such as the CRM licence or the account executive who would otherwise be taking inbound calls. Consistency matters more than precision: pick a rule and keep it.

What if I have no outbound history to get a win rate from?

Then you cannot calculate cost per customer yet, and you should not pretend otherwise in a forecast. Run a defined test for a quarter, record every stage of the chain, and treat that spend as the price of learning your own numbers. After that, the calculation works and every future decision is cheaper to make.