AI SDR vs a Human SDR: The Real Cost Math
US BLS OEWS May 2023 puts the median Sales Representative of Services (SOC 41-3091) at about $31.06/hr — the public human baseline an AI SDR has to beat per qualified outcome, since no honest single AI-per-lead price exists yet to cite.
The "AI SDR" pitch is almost always framed as cheaper than a human — but cheaper per what? Vendors quote per seat, per month, per meeting booked, and per lead, and most do not publish a fixed per-outcome price at all. So the defensible way to do the cost math is to anchor on the real, public cost of the human role the AI replaces, then compare any AI quote against it per qualified outcome. This post builds the human SDR baseline from US Bureau of Labor Statistics data and shows the per-meeting math — without inventing an AI lead price.
The human SDR baseline (real, cited)
Sales development sits closest to the BLS category Sales Representatives of Services, Except Advertising, Insurance, Financial Services, and Travel (SOC 41-3091). US BLS OEWS (May 2023) reports a median hourly wage for that group of about $31.06 — roughly $64,600 a year. Inside sales reps span a range, so treat that median as a midpoint, not a fixed salary.
For a more transactional comparison some teams also reference Securities, Commodities, and Financial Services Sales Agents (SOC 41-3031); pick the SOC that matches the role you are actually staffing and cite it explicitly. Either way, the base wage is the floor — fully-loaded cost (benefits, tooling, ramp, management) runs higher and varies by employer.
From wage to cost per qualified outcome
An SDR’s cost only becomes comparable to an AI quote when you express it per qualified outcome — a booked meeting or an SQL — rather than per hour. If an SDR books, say, two qualified meetings a day across an 8-hour day, the base-wage cost is about ($31.06 × 8) ÷ 2 ≈ $124 per meeting; at four meetings a day it is about $62. These are arithmetic illustrations from the cited BLS median and an assumed throughput — not measured WorkForce data — so substitute your own pipeline numbers.
As with support, the value is the unit: dollars per qualified meeting. That is the axis on which a human SDR and an AI SDR become directly comparable, no matter how each one is billed.
Why we do not quote an AI per-lead price
There is no honest single "AI SDR cost per lead" to publish until verified transactions accrue for the category — so WorkForce does not invent one. Where a vendor publishes its own list price (a per-meeting or per-seat figure on its pricing page), record it explicitly as that vendor’s own published number and convert it to dollars-per-qualified-outcome before comparing.
The lesson of the first AI-SDR wave is that price was never the real problem — verification was. When buyers cannot independently check the quality of the meetings an agent books, a low price just hides churn. That is what an independent quality score is for.
How to compare an actual AI SDR quote
Convert the AI quote to dollars-per-qualified-meeting and read it against the human baseline above. Then demand a quality number: what fraction of "qualified" meetings are genuinely sales-accepted? A cheap agent that books unqualified meetings is more expensive than it looks once your AEs’ time is counted.
The right choice is the one whose quality-per-dollar wins — not simply the cheapest seat or the lowest per-meeting headline.
Methodology note
This article cites real BLS OEWS May 2023 figures (SOC 41-3091, median ~$31.06/hr / ~$64,600/yr; public domain) for the human baseline, and names SOC 41-3031 as an alternate role match. The per-meeting figures are arithmetic from that median and a stated, assumed throughput — not measured WorkForce transaction data — and no AI market rate is stated as a live figure. WLI category rates publish with their confidence intervals and sample sizes once transaction coverage exists; see /methodology.
Next steps
Get a free, independent quality score for any sales agent at /eval, and benchmark against the live SDR rate at /wli as transaction coverage accrues. Until then, the BLS baseline above is the honest floor.