You lower an agency's cost per meeting by raising conversations per SDR, not by cutting SDR pay or by hiring a sixth SDR to do what five could do with the right tools. Cost per meeting is total production cost divided by meetings delivered, and the denominator is where the leverage is. Here is the model, the five levers in order of impact, and when hiring actually becomes the right call.

The cost-per-meeting model

Add up everything that produces meetings: SDR salaries and charges, management time, tools, data, telephony. Divide by qualified meetings delivered in the month. That is your cost per meeting, and your margin is the gap between that number and what clients pay you per meeting.

Two things follow. First, salary is the bulk of the numerator, so cutting tools to save money barely moves the number. Second, the denominator is driven by one chain: dials, connects, conversations, meetings. Change any link and the cost moves.

Our benchmark of 1,000,000 B2B cold calls in France shows an 11.3% average connect rate. A SDR dialing by hand places 50 to 60 dials a day and has 5 or 6 conversations. If 1 conversation in 5 becomes a meeting, that is roughly 1 meeting a day, 20 a month. The whole cost base of that SDR sits on 20 meetings.

Why hiring is the most expensive lever

A new SDR costs more than their salary: recruiting, onboarding, 3 to 4 months to full productivity, and the risk that they leave at month 8. Worse, a new SDR on the same tools has the same ceiling: 5 or 6 conversations a day. You have doubled the numerator to double a denominator that was capped by process, not by people.

The founder test we use: if your SDRs are in conversation less than 20% of their calling time, you have a tooling problem, and hiring into a tooling problem buys you two people waiting on ring tones instead of one.

The five levers, ranked by impact on cost per meeting

1. Conversations per SDR (the dialer)

A parallel dialer calls several numbers at once and connects the SDR to the first human. In our data it gives 5x more conversations for the same hours. Even at a lower meeting rate per conversation (reps cannot prep as deeply at that pace), the arithmetic is decisive: 3 to 4 times more meetings per SDR means cost per meeting falls by two thirds without touching headcount or pay.

2. The list (who gets called)

Conversion per conversation doubles when the call has a reason. Trigger-based openers reach a 7-10% meeting rate in our benchmark against 2-3% for pitch-first openers. For an agency, that means building each client's list on signals (job posts, website visits, funding, job changes) rather than on a static export, and re-enriching numbers before the first dial so the dialer is not wasted on dead data.

3. The time slot

Free, and ignored. In our data 4pm connects at 15.8%, 10am at 14%, 8am at 9.8%. Founders pick up around 10am, marketing around 4pm, HR around 1pm. Move each client's calling blocks to the persona's slot and connect rate moves without a single new tool.

4. Admin time

Logging outcomes, writing notes, updating 8 client CRMs by hand: an hour a day per SDR in our experience. Automated logging from the dialer gives that hour back, and an hour of calling at 4pm is worth more than an hour of CRM hygiene at 6pm.

5. Conversion coaching

The slowest lever and the most durable. Discovery questions starting with "how" lift conversion by 190%; asking "what do you mean?" before answering an objection lifts it by 197%. These are habits, and a coaching layer on the recordings shows which SDR has them. Coach on conversion, not on volume; the dialer already handles volume.

Why cutting SDR pay raises cost per meeting

It looks like the fastest lever and it is the only one that moves the number the wrong way. Lower pay means your best SDRs leave for the agency that bought a dialer, and the replacement costs you 4 months of ramp and a client who noticed. Pay at market, then make each SDR worth more: that is the only version of this that holds for more than a quarter.

When hiring becomes the right call

Hire when all five levers are in place and the constraint is hours, not process: talk time above 40% of calling time, conversations per SDR at or near 20 a day, and signed client volume that the current team cannot serve. At that point a new SDR inherits a working system and ramps in weeks, and the cost per meeting barely moves.

How Sonik approaches cost per meeting for agencies

Sonik is a parallel dialer with AI call prep and coaching, built in France and used by lead-gen agencies with per-client number pools, per-client reporting and automated logging to each client's CRM. Our north-star funnel is 20 conversations per SDR per day, and the mission is 2x more meetings through 5x more conversations with the right contacts.

The signals layer covers lever 2, sorting each client's list by reason to call. Coaching covers lever 5, flagging the moments that decide a call so managers coach on evidence. We also wrote about the in-house version of this problem: how to make more sales calls without hiring more SDRs.

Conclusion

Cost per meeting falls when conversations per SDR rise, and conversations rise with the dialer, the list, the slot, less admin and better coaching, in that order. Hire last, and never cut pay. This week, compute your real cost per meeting for one client, then time one SDR's talk time for an hour; the gap between those two numbers is your plan.