Cost of sales goes down when cost per held meeting goes down, and that number is moved by five levers: conversations per rep hour, the opener, the time slot, the signals already in your CRM, and the share of the day reps actually spend selling. None of them requires cutting headcount or pipeline. Here are the metrics that make cost of sales visible, the five levers with their effect in our data from 1,000,000 B2B cold calls in France, and the two cuts that look like savings and are not.

The metrics that make cost of sales visible

Cost of sales as a percentage of revenue is the number the board sees, and it is six months late. By the time it moves, the cause is three quarters old. For an outbound team, four unit costs tell you what is happening now.

• Cost per conversation. Monthly fully loaded cost of the rep and tools, divided by conversations held (connects that got past the opener with the right person).
• Cost per held meeting. Same numerator, divided by meetings that happened. This is the one to put on the wall.
• Cost per qualified opportunity. The number your AEs and your CFO can both read.
• Selling-time share. The share of a rep's week spent in conversations or preparing them. Salesforce's State of Sales research has put the time reps actually spend selling at under 30%; everything else is cost without output.

Compute them per rep and per month. Three months of history is enough to see which lever to pull.

Five levers that cut cost per meeting

1. Conversations per rep hour

At the French average connect rate of 11.3%, a rep on a single line spends most of the hour listening to ringing. Parallel dialing gives reps about 5x more conversations in the same hours. Cost per conversation falls by roughly the same factor, with no change in salary and a tool cost that is a fraction of it. This is the largest lever and the one most teams pull last.

2. The opener

Trigger-based openers reach a 7-10% meeting rate per conversation in our data; pitch-first openers reach 2-3%. 65% of reps open by explaining what they sell. Moving a rep from one to the other divides cost per meeting by three and costs a coaching session.

3. The time slot

Connect rate is 15.8% at 4pm, 14% at 10am and 9.8% at 8am. A rep who moves two hours of dialing from 8am to 4pm gets 60% more connects per dial hour. The cost of this lever is a calendar change.

4. The signals you already paid for

Reps use about 20% of the signals already sitting in their CRM: website visits, job changes, closed-lost deals, dormant leads. 50% of our own pipeline comes from those signals. A dormant lead called with "How have you been?" converts around 10% versus about 1.5% for a standard cold open. These are meetings at near-zero data cost, because the data is already in a system you pay for.

5. Selling-time share

List building, CRM logging, pre-call research and voicemail drops are hours a week that produce no conversations. Automatic logging, AI call prep and signal-driven lists give those hours back. In our experience this is the lever that also reduces turnover, which is the hidden cost of sales nobody puts in the model.

Two cuts that look like savings and are not

Cutting the tool budget. A dialer that produces 5x conversations per rep is the cheapest headcount you will ever buy. Cutting it to save a few hundred euros a month per seat raises cost per meeting by a multiple.

Cutting the data budget. A rep calling a list with 40% wrong numbers has a connect rate you cannot coach. In Europe, mobile coverage decides connect rate more than anything the rep does; the data line is where cost per meeting is often lost, not saved.

The honest cut is the one the unit costs point to: a persona nobody connects with, a slot nobody answers in, a channel with a cost per opportunity three times the others.

A European note on cost of sales

Three things make the numbers different for a French or European team. Connect rates are modest (11.3% on average in our benchmark of 1,000,000 B2B cold calls in France), so the same meeting target needs more dials and the volume lever matters more. Fully loaded headcount is more expensive to adjust, so levers that work with the existing team are worth more than hiring plans. And August removes a month of selling from the year; a cost model that spreads activity evenly will overstate the output of every rep.

How Sonik approaches cost of sales

Sonik's product is built around the first four levers. Parallel dialing multiplies conversations per rep hour. CRM signals (website visitors, job changes, closed-lost, dormant leads, social intent) decide who is on today's list, so the meeting rate per conversation is higher and the data is already paid for. AI call prep and summaries take the research and logging out of the rep's day.

The dashboard reports conversations, meetings and connect rate by slot per rep, which is enough to compute cost per held meeting without a spreadsheet. Our mission is 2x more meetings from 5x more conversations with the right contacts, which in cost terms means the same team producing twice the pipeline.

Conclusion

Cost of sales is a unit cost problem with five levers, and four of them cost nothing but attention. This week, compute cost per held meeting for each rep over the last three months. The spread between the best and worst rep is usually the opener or the slot, and both are fixable before the next planning cycle.