I've spent a lot of time in B2B sales and talked to dozens of companies about their outbound growth. The objection I keep hitting is 'we already have an in-house SDR.' That never made sense to me, so I ran the numbers. Here's why cold email infrastructure usually beats hiring a human for prospecting - and the one case where it doesn't.
1. Real cost of an SDR
Checked job posts, recruiter data, and salary surveys. Base pay is just the start. Stack commission, benefits, payroll taxes, tools, management time, recruiting, and the 3 months of ramp where they produce almost nothing. Fully loaded, you're looking at low-to-mid six figures per rep per year. Remote rep with basic tools sits at the bottom, an enterprise rep with full stack hits the top. Either way, it's a lot more than the advertised salary.
2. Infrastructure cost
Typical stack: a sequencer, email infrastructure with multiple providers to hit 200k sends/month, and lead data. All-in about £800-1000/month. Annualised, that's a small fraction of one SDR.
3. Output comparison
That same infrastructure at 215k sends/month with a 2% reply rate yields roughly: 4,300 replies → 430 positive → 140 booked meetings → 100 live calls → 10-20 closed deals. Per month. A decent operator hitting 3.5% nearly doubles that - call it 300 live calls a month off one stack. To match 300 calls/month with humans, you'd need 15-20 SDRs at 15-20 booked calls each. That's a seven-figure payroll line. Cost per booked meeting gap from recent benchmarks works out to roughly 18x in favour of cold email.
4. Turnover tax
Average SDR tenure is 14-18 months. Annual turnover sits around 35-40%. One departure costs recruiting, onboarding, lost pipeline, and the slow months before they quit. If you hire three reps, expect to lose one within a year. Cold email has zero turnover. A domain might burn, but replacing it takes 24 hours and costs almost nothing. Replacing a burned-out SDR takes three months. Over three years, a single SDR seat runs into the high six figures, the same window of email outreach is a tiny fraction. Roughly half a year's salary saved per replaced seat, every year.
So when do you hire a human? Only for large enterprise deals that need multi-threading across 5-10 stakeholders, sales cycles of 6+ months, or products that require a live demo. Or if you're in an industry like trades or local services where email just doesn't land.
Framework I share with founders:
- Pre-revenue / early stage → infrastructure only, founder runs it
- Finding traction → infrastructure + founder selling, if you hire, hire a closer, not a prospector
- Scaling → infrastructure + one SDR for high-value accounts only
- Established → infrastructure for volume, SDR team for named enterprise accounts
If you're a CEO thinking about replacing an SDR: put a little into infrastructure and use the rest to hire a closer. Cold email generates meetings, closers turn meetings into revenue. Point infrastructure at volume, point humans at deals that need humans.
Curious if anyone else has run the numbers and come to a different conclusion.
