Our 1.25 Humans + 20 AI Agents Closed 140% of What Our All-Human Sales Team Did Last Year. But I’m Not Sure That’s the Real Story.

Here’s the headline number: we replaced most of our human sales team with 20 AI agents, kept 1.25 humans, and closed 140% of what the full human team closed the year before.

That’s a real number. I’m not going to hedge it or qualify it away.

But I’ve been doing this long enough to know that a single number rarely tells you what you think it tells you. And if I’m being honest, I’m not 100% sure what drove what here.

Let me walk through what actually happened.

We Didn’t Just Add AI. We Restructured Everything.

By deploying AI agents to replace humans that left, we ended up concentrating all qualified leads into our best closers.

This matters more than people realize. In a typical sales org, leads get distributed across reps of varying skill levels. Your best closer might get the same number of leads as someone who started three months ago. That’s “fair” but it’s not efficient.

When we went to 1.25 humans, those humans were our best people. Every qualified lead went to someone who knew how to close. No more spreading leads across a bench of mixed-quality reps.

That alone would have moved the number. How much? I genuinely don’t know. Maybe 15%. Maybe 30%. There’s no clean way to isolate it. But this alone might account for 50% or more of our results with AI Agents. The fact that the AI Agents let us concentrate leads in the best human closers.

AI Agents Gave Us Something We Never Had: 100% Coverage

Here’s what actually changed with AI agents, and it’s more boring than the hype suggests.

Before AI, we responded to fewer than 40% of inbound leads, max. That’s all the humans wanted to respond to. It might even have been less. And often, humans took days to get back to them.

With AI agents, we respond to 100% of inbounds. Every single one. Instantly. At 2am on a Saturday. On Christmas.

That’s not intelligence. That’s coverage. And coverage turns out to matter a lot more than most sales leaders admit.

The same thing happened on the outbound side. Our AI agents reached out to every single past prospect in our database. Every one. A human sales team cherry-picks. They look at a list of 10,000 past prospects and work the 500 that look most promising. The AI agents just worked all 10,000.

Were most of those 10,000 dead ends? Of course. But the ones that weren’t were deals we never would have found.

And Then There’s the Tailwind We Can’t Ignore

Here’s where it gets tricky.

AI took off in 2025 and into 2026. Not a little. A lot. And our entire business is now AI + B2B content and community. Our events are AI-focused. Our content is AI-focused. SaaStr.ai itself became a product.

So were we closing more because AI agents are better at sales? Or were we closing more because the market for what we sell got dramatically bigger at the exact same time we deployed AI agents?

I think it’s both. But I can’t tell you the ratio.

This is the part that most “we replaced our sales team with AI” posts leave out. If your product suddenly has massive tailwind at the same moment you restructure your sales org around AI, you can’t cleanly attribute the results to the AI.

The Honest Answer: We Don’t Know the Counterfactual

What I really want to know is this: what if we had kept the full human team AND had the AI tailwind AND concentrated leads in our top closers? Would we have closed 180%? 200%?

We’ll never know. There’s no A/B test for this. You can’t run two parallel versions of a company.

What I can tell you is what we actually observed:

AI agents worked fine. They didn’t embarrass us. They didn’t send weird emails. They didn’t hallucinate pricing. They handled first-touch qualification, follow-up sequences, and re-engagement of old prospects at a level that was… fine. Not magical. Fine. Professional and consistent.

Concentration of leads in top closers worked. This is the unsexy part. Taking your best people and giving them all the at-bats is a strategy that predates AI by decades. It just works.

100% coverage is a real unlock. Going from <40% response rate to 100% is not incremental. That’s a step function. Every company with inbound leads is leaving deals on the table simply because nobody responds fast enough or at all.

So What’s the Takeaway for Founders?

Don’t deploy AI agents because you think AI is magic at selling. Deploy them because:

**You can’t afford 100% human coverage and you’re leaking leads.**If you’re responding to less than half your inbounds, you have a coverage problem. AI fixes coverage problems well.**You want to free up your best closers to only close.**If your top rep is spending 40% of their time on qualification and follow-up, that’s 40% of their closing capacity you’re wasting.**You have a database of past prospects nobody is working.**Every B2B company has thousands of past prospects sitting in a CRM collecting dust. AI agents will work that list without complaining about it.

But don’t fool yourself into thinking the AI is what’s closing deals. In our case, the AI agents created more at-bats for humans who are great at closing. That’s the real story. More at-bats, better batters, and a tailwind that made the ball carry further.

Did we close 140% because of AI? Partly. Did we close 140% because we concentrated talent? Partly. Did we close 140% because AI as a market exploded? Partly.

The honest answer is we did something that worked, and we’re not entirely sure which lever mattered most. And anyone who tells you they know exactly how much AI contributed to their sales results is probably not being straight with you.

What I do know: 20 AI agents and 1.25 humans is a lot cheaper than a full sales team. And 140% of last year’s number is 140% of last year’s number, whatever the cause.

That’s good enough for now.