What outbound actually costs per conversation

Ask a sales leader what a meeting costs and you will usually get the software bill divided by the number of meetings. Ask what a conversation costs — fully loaded, including the salaried hours that went into producing it — and you will usually get a pause.

The pause is reasonable. The work that produces a meeting is spread across research, list triage, qualification, writing, sending and chasing, and none of it appears on an invoice. It is the largest line item in outbound and the only one nobody costs.

Here is how to do the sum, and what it usually reveals.

Why nobody does this sum

Three reasons, all of them understandable.

The cost is salaried, so it feels fixed. A rep is paid whether they spend Tuesday researching or selling. That makes the hours feel free, which they are not — they are the most expensive hours in the company being spent on the least skilled part of the job.

The work is invisible. Nobody logs “forty minutes deciding which of these 1,500 rows to look at”. It happens between the things that do get logged.

The output is uncertain. You cannot attribute a closed deal to the hour spent triaging the list that contained it, so the hour never gets attributed at all.

But the sum is worth doing, because it changes which problem you try to solve. Teams that have not done it reach for more volume. Teams that have reach for better selection, because the arithmetic points there.

Where the hours actually go

For a rep working a named list by hand, roughly:

Activity	Share of outbound time
Deciding who to contact from a list	25–35%
Researching a specific prospect before writing	20–30%
Writing and personalising	15–25%
Sending and logging	10–15%
Chasing, following up, remembering	10–20%

These are ranges, not measurements, and your team’s split will differ. The shape is what matters: more than half the time goes on deciding and researching, before a single word is written.

That is the part most tooling does not touch. Sequencers make sending faster. AI writers make writing faster. Neither helps with the sixty per cent spent working out who deserves the message, which is why teams often buy both and find the calendar no fuller.

The arithmetic

Take a four-person team, fully loaded at $70,000 each — salary plus employer costs, not just salary. That is $280,000 a year, or about $135 an hour across the team at 2,080 hours each.

Say each rep spends 60% of their week on outbound. That is roughly 25 hours a week each, 100 across the team, or about $13,500 a week in salaried outbound time.

Now the output. If that team produces 12 qualified conversations a week:

• $1,125 per qualified conversation, in salaried time alone
• Software at, say, $600 a month adds about $12 per conversation
• So the tooling is roughly 1% of the cost of a conversation

That ratio is the finding. Most teams negotiate hard over the 1% and never examine the 99%.

If 60% of those hours go on selection and research, that is about $8,100 a week deciding who to contact. Whether that is money well spent depends entirely on whether the decisions are good — and in most teams nobody can check, because the reasoning is not written down anywhere.

Which lever actually moves it

There are four, and they are not equally useful.

Contact more people. Increases conversations and cost roughly together. The cost per conversation barely moves, and account risk goes up. This is the lever most teams pull first.

Improve reply rate. Real but bounded. Going from 8% to 12% is excellent work and a big lift; going to 40% is not available.

Spend less time deciding. This is where the money is. If selection drops from 60% of outbound hours to 25%, the same team produces roughly the same conversations in substantially fewer hours — or more conversations in the same hours. Either way the cost per conversation falls hard, and it falls without contacting more people.

Pay less for software. Moves 1%.

The third lever is the only one with real headroom, and it is the one that requires a decision to be made systematically rather than by a person reading rows.

What tooling actually changes

Be precise about this, because the category is not.

Sequencers and automation tools compress the sending and chasing steps — the 20–35% at the bottom of the table. Genuinely useful, and the wrong end of the problem if selection is your constraint.

AI writers compress the writing step, 15–25%. Also useful, and only after you know who you are writing to.

Data and enrichment tools compress research, 20–30%. Closer to the money.

Scoring and qualification compresses the 25–35% spent deciding, which is the largest single block and the one nothing else touches.

This is not a claim that one category is better. It is a claim about sequencing: automating the sending of messages to a list nobody has qualified makes the wrong outreach faster, and the arithmetic above is why that feels like progress without being it.

There is also a cost the sum does not capture. Automation carries account risk, and a restricted LinkedIn account takes a rep’s network with it — which is a large, lumpy cost that does not appear in any per-meeting figure until the week it does.

Hire, or fix the process?

The question usually arrives as “we need another SDR”. Run this check first.

If your existing reps spend more than half their outbound time on selection and research, a new hire will also spend more than half their time there. You are buying more of the expensive activity, at the same ratio, and the cost per conversation stays flat.

If they spend most of their time in conversations and the constraint is genuinely hours, hire — that is what hiring is for.

The honest version of the point is not “do not hire”. It is that when you do hire, the person should spend their time on the part of the job you hired them for. A new SDR whose first three months go on list triage is an expensive way to buy list triage.

Do it for your own team

Six numbers, fifteen minutes:

1. Fully loaded cost per person on outbound — salary plus employer costs, not salary.
2. Share of the week each spends on outbound. Ask them; the answer is usually lower than the plan assumes.
3. Multiply out to get weekly salaried outbound cost.
4. Qualified conversations per week. Use your own definition, consistently.
5. Divide: cost per conversation.
6. Add software, and note what share of the total it is.

Then one more, which is the one that tends to change the meeting: ask each rep what share of their outbound time goes on deciding who to contact. Multiply that share by the weekly cost. That figure is what your selection process costs you, and it is usually the largest controllable number in the whole operation.

If it is large and nobody can explain how the decisions are made, that is the thing to fix — and it is what scoring every prospect with a written reason is for. You can run it on your own list and compare the ranking against the order your team works from today, which settles the question faster than any model of it.