If AI Takes Over the SDR Seat, Where Will Your Future AEs Come From?

著者: Brian Tr · 更新日: 2026年10月11日

Most conversations about AI and the SDR role stop at one question: can software book the meeting as well as a person? That's a fair question for this quarter's budget. It's the wrong question for your 2029 revenue plan.

Our view: the SDR seat was never only a prospecting job. It was the cheapest place in the company to teach someone to talk to buyers, take rejection, read an account and earn the right to carry a quota. If you automate the seat and don't build something to replace that training, you'll save money now and discover in about three years that you have no bench of AEs to promote and a hiring market full of people who were never trained.

That doesn't mean you shouldn't automate outbound. It means automation is half of the decision, and the other half is where your next closers learn. The rest of this piece lays out the evidence, the failure pattern, and a table for choosing a replacement path.

Key Facts

• The Bridge Group's 10th SDR Models and Metrics report, based on 351 B2B companies, puts average SDR tenure at 1.9 years, with 58% of companies reporting an average tenure of 12 to 23 months. The same report lists average ramp time at 3.0 months.
• In that report, promotions accounted for 16% of annual SDR attrition in 2024, down from 34% in 2020. The path to AE was already narrowing before AI-first outbound arrived.
• In a 2017 analysis of about 205 North American AEs who had been SDRs, Matt Bertuzzi of The Bridge Group found that 55% of SDRs promoted after 11 or fewer months failed in the AE role, against 6% of those promoted after 16 or more months. Failure meant leaving the AE seat within six months. It's a small, LinkedIn-sourced sample, so treat it as directional.
• Unify, a go-to-market software vendor, analyzed roughly 180,000 people who started an SDR or BDR role between January 2021 and January 2026. It reports that 22.3% reached AE by July 1, 2026, at a median of 12 active SDR months. The share reaching AE within 12 months fell from 11.1% for the 2021 starting class to 7.2% for 2023. This is the vendor's own dataset.
• Stanford's Digital Economy Lab, using ADP payroll data through June 2026, finds employment for workers aged 22 to 25 in AI-exposed occupations is 19% below where it would have been had it kept pace with less-exposed peers. The authors report declines concentrated where AI substitutes for human tasks.

What the SDR seat actually produced

Ask a sales leader what an SDR does and you'll hear "books meetings." Ask where their best AEs started and half will say the SDR team. Both answers are true, and they describe two different products.

The visible product is pipeline: booked meetings, qualified opportunities, a cost per meeting you can compare to a software license. That's what AI SDR tools are sold against, and on that measure a tool can look very good.

The invisible product is people. Over twelve to eighteen months an SDR sees a few thousand real buyer reactions. They learn which opening lines get hung up on, what a real objection sounds like next to a polite brush-off, how a buying committee forms, and how to handle being told no forty times in a day. That's an apprenticeship in the one skill every AE needs and few can be taught in a classroom. Nobody put it on the budget line, so nobody notices it leaving when the headcount does.

The Bridge Group's data shows how much of the supply chain depends on this. Average tenure is 1.9 years, which gives a team a window of about a year of useful, ramped output and a similar window of learning. Promotion is where that learning pays off. And promotion was already thinning: 16% of attrition in 2024, against 34% in 2020.

The bench problem arrives late

Here's why this mistake is easy to make. When you cut an SDR seat, nothing breaks. Pipeline can hold up, cost per meeting drops, and the board likes the chart. The damage shows up on a delay equal to roughly the time an SDR would have needed to become an AE.

Illustrative example: a company with eight SDRs promotes two or three a year into AE roles. It replaces most of the SDR team with automation in year one. In year one, nothing visibly changes. In year two, the last promoted SDR is a year into the AE seat and the next cohort doesn't exist. By year three, an AE resigns, a territory opens, and the only candidates are external hires at market rates who don't know the product, the buyers or the way your company sells. That's an invented scenario, not a case study, but the arithmetic is the point: a pipeline that feeds on a role will starve a few years after the role disappears.

Two other forces make it worse.

First, the outside market won't rescue you. Stanford's payroll analysis shows young workers in AI-exposed occupations are the group whose employment has dropped relative to peers, which is consistent with many companies trimming the same entry rung at once. If everyone empties the same rung, the supply of experienced AEs a few years out gets thinner for all of them. Our piece on the broken bottom rung covers that dynamic across functions.

Second, promoting faster isn't the answer. Bertuzzi's 2017 sample suggested that SDRs moved to AE after eleven or fewer months failed far more often than those with sixteen or more. It's one small study, but the direction fits what most sales managers will tell you: seat time with real buyers matters. If you automate half the seat and then push the remaining SDRs up early to fill AE gaps, you trade a pipeline problem for a failure-rate problem.

What the data does and doesn't tell us

We should be plain about the limits.

What's well supported: SDR tenure is short, ramp is fast, and promotions have fallen as a share of attrition. Those come from a recurring survey of 351 companies. Entry-level employment in AI-exposed occupations has also dropped relative to peers, per payroll records.

What's suggestive: the Unify dataset shows the share of SDRs reaching AE within 12 months falling by cohort, and that 22.3% overall reached AE. But Unify is a vendor with a stake in the go-to-market stack, and a cohort that started in 2023 hasn't had as long to be promoted as one from 2021, so the cohort comparison is partly a function of time. Bertuzzi's failure-rate figure is a single LinkedIn-based sample.

What nobody has shown: that companies who automated SDR work then ran short of AEs. It's too soon, and no one has published a matched comparison. Our claim is a mechanism and an arithmetic one, not a measured result. If you read it as a forecast, you'd be right to test it against your own promotion history before acting.

Where the apprenticeship can move

If the SDR seat shrinks, the learning has to live somewhere. There are five realistic options, and each has a cost.

None of these is free, which is the point. The SDR seat was a cheap way to train people, and you'll pay more per head to do the same job through any other route. The honest comparison isn't "AI tool versus SDR salary." It's "AI tool plus the cost of the replacement path versus SDR salary."

How to tell whether your pipeline is at risk

Pull your own numbers before copying anyone's benchmark. Three questions do most of the work.

1. Where did your last ten AEs come from? If more than a third started as SDRs, the seat is load-bearing for you. If nearly all were hired externally, the seat matters less and your cost of cutting it is lower.
2. How long do your promoted SDRs stay in the AE seat? If your newly promoted AEs mostly succeed after twelve or more months of SDR time, you've got a stable path to protect. If many fail, fix the path before you shrink it.
3. How many AE seats will open in the next three years? Count growth hires plus expected attrition. If the answer is more than the number of people you could promote or hire from outside, you're already short.

The answers feed straight into sales org design, because the SDR layer, the AE layer and the promotion path between them are one design decision. They also connect to retention and new hiring: a seat that offers a visible path up is also a seat that keeps good people.

Objections you'll hear

"AI SDRs book meetings at a fraction of the cost." They might, and that's a fine reason to automate the repetitive part. It doesn't address where closers come from. If you want to see how work gets rebalanced rather than removed, our note on SDR workload balancing with AI is a starting point.

"We'll hire experienced AEs from the market." You can, at higher cost and with more attrition risk. And if competitors do the same, you're all bidding for a smaller pool. Hunting from competitors has its own failure modes.

"Our SDRs never became AEs anyway." Then the argument doesn't apply to you, and you can automate with a clear conscience. Check the data first. Many leaders underestimate how many closers started in the seat.

"The best SDRs will find a way up regardless." Maybe, but only if there's a ladder to climb. If the entry seat is gone, there's nothing for them to find a way up from.

What to do next

Treat this as a planning exercise before you sign an AI SDR contract, not a reaction after.

1. Map your AE sources. For each AE hired in the past three years, record whether they were promoted from SDR, hired as a junior AE, or hired experienced. That's your dependency on the seat.
2. Forecast AE seats. Count growth plus expected attrition for the next three years, and compare it with your realistic supply from each source.
3. Decide the replacement path before you cut. Pick from the table, price it, and put that cost next to the AI tool's cost in the business case.
4. Keep a human core. If you automate volume, keep a smaller team of SDRs on your best accounts, with a written promotion standard that doesn't push people up early.
5. Make promotion criteria explicit. Tenure, call quality, pipeline quality and coaching feedback, written down and shared. Our guide to coaching reps in one-on-ones covers the manager side, and what a sales manager is for when AI reviews every call explains why that coaching time matters more now.
6. Revisit quotas and capacity. If automation raises each rep's workload, quotas have to move with it, or your juniors will be set up to fail.
7. Review yearly. Check promotion counts, AE failure rates and open seats every year. The warning sign is a drop in internal promotions with no matching drop in AE openings.

If you'd rather see the decision as a sentence: automate the work, not the apprenticeship.