What Does a Meeting Actually Cost? The Outbound Maths Nobody Writes Down

Almost nobody running outbound at a small company can tell you what a meeting costs them. Not because they are careless, but because the biggest input never appears on an invoice: their own time.

Here is the arithmetic, laid out so you can put your own numbers in. It takes about fifteen minutes and it usually changes the plan.

Work backwards, not forwards

The instinct is to start with a list size and see what comes out. Start at the other end.

One closed deal. How many proposals does that take? How many meetings produce a proposal? How many replies produce a meeting? How many sends produce a reply?

Filled in with numbers that are typical rather than optimistic:

So one deal costs roughly a thousand harvested leads, and 720 people contacted, at a 5% reply rate.

Put your own numbers in. Every one of these is measurable from what you already have, and the exercise is worth doing with real figures even if you have only thirty data points, because the ratios that surprise you are the ones worth fixing.

Now price each stage

Harvesting and verification. Machine time. At any realistic per-lead software cost this comes to single-digit or low double-digit currency for a thousand leads. It is genuinely the cheapest part of the whole chain, and it is where most people spend their attention shopping.

Writing the sequence. Perhaps four hours for a segment you understand: five messages, a subject that is not a trick, and three variants of the opening line. This is a fixed cost per segment, not per lead, which is exactly why narrow segments look expensive and are not.

Sending and monitoring. Automated, but not zero. Call it two hours a week of reading replies, fixing what is not working and pulling out the leads who should never have been in there.

Handling replies. The big one. Thirty-six replies, at fifteen minutes each including the back and forth to book, is nine hours.

The meetings themselves. Six meetings at an hour each including preparation is six hours.

Total human time for one deal: roughly twenty-five hours, against a software bill in the tens.

The number that follows

At £50 an hour for your time, one deal costs about £1,250 in labour and perhaps £15 in tooling. Cost per meeting is therefore around £210, and 98% of it is you.

Two conclusions fall straight out of that, and they are the entire point of the exercise.

1. Software price is a rounding error. The difference between a £30 platform and a £90 platform is £60 against a £1,250 cost. If the more expensive one saves you two hours a month it has paid for itself several times over. Optimising the tooling bill is optimising the smallest number on the page.

2. Reply rate is the whole game. Look at what happens if the reply rate goes from 5% to 8%, which is a realistic effect of a better segment:

A 3 point improvement in reply rate takes nearly 30% off the cost of a meeting. Nothing else on the page moves it that far.

Where reply rate actually comes from

Since reply rate is the lever, it is worth being precise about what moves it. In rough order of effect:

1. Segment. Whether the person has the problem you solve. This dwarfs everything else and it costs nothing to change.
2. The reason for writing. A specific observed gap versus a generic introduction.
3. Follow-up count. A large share of replies arrive after the third touch, and most people stop at two. See the follow-up sequence.
4. Deliverability. Not a copy problem, but it sets the ceiling on all of the above. An email in spam has a reply rate of zero regardless of how good it is.
5. The ask. Smaller asks get more replies, reliably.
6. Subject line and phrasing. Real, and smaller than the five above. A/B testing cold email at low volume covers why testing this first is a trap.

Notice that the top two are free, the third is free, and the fourth is mostly a setup cost. The cheapest levers are the strongest ones, which is not usually how the world works and is worth taking advantage of.

The trap: cheap meetings with the wrong companies

One caution before you go and optimise cost per meeting.

Cost per meeting is a means, not an end. It is entirely possible to halve it by broadening the segment and lowering the bar for what counts as a meeting, and to end the quarter with more meetings, more hours spent and less revenue.

So track cost per closed deal alongside it, and track average deal value by segment. If one segment produces meetings at half the cost and deals at a third of the value, the expensive segment is the good one, and cost per meeting would have told you the opposite with great confidence.

What to do with the number

Three things, in order.

Calculate it once, honestly, with your own hours priced at what you would actually pay someone. Not zero. Never zero.

Then attack the top of the funnel, not the bottom. The instinct is to work on closing, because that is where the deal is. The arithmetic says the leverage is in reply rate, which is upstream of everything, and mostly in whether you picked the right people to write to.

Then automate the hours, not the judgment. Twenty-five hours per deal, and nine of them are reply handling. That is the part where drafting assistance genuinely pays for itself, which is what the reply drafts in our AI sales agents are for: they remove the blank page and wait for you to press send. The judgment about who to write to, which is where the actual money is, stays where it belongs.

If you want the seven day version of getting the first numbers on the board, your first 100 leads is the practical companion to this one, and cold email benchmarks will tell you whether the conversion rates you just wrote down are normal.