I had ten founder calls last week.
Eight of them have never run structured cold outbound. The other two were trying to do it themselves between everything else they're running.
That's the pattern. It keeps coming up.
The story I expected to hear when I started taking these calls was "our outbound isn't working anymore." That isn't what people are showing up with. They're showing up unbuilt. Not broken. Different problem.
A few from the week, anonymized
An IT services founder, US and Canada market. Asked him how they generate pipeline. His answer: "All our business is referrals. We never did any marketing. Now I'm looking to explore that area." Referral-led for years. No outbound infrastructure of any kind.
A VSL agency that's closed roughly $2M on warm intros and never made a cold call. Now wants a cold engine because the warm well is finite and he can see the bottom of it.
A self-storage operator with 20 years of operating experience, drowning in inbound leads. His words: "Somebody could take all these contacts and really nurture them." No system. No process. Just a pile of leads not being worked.
A pickleball publication selling $500 to $1,500 ad placements. "We aren't able to spend even ten minutes on it. We just have other stuff."
A founder building websites for blue-collar contractors. Five months in. Zero clients. He'd scraped 800 plumbers and roofers off Google Maps by hand and gotten maybe 30 to reply. "Man, I was like, damn dude. This is harder than I thought for sure."
The read
Most founders calling sales companies right now are not broken. They've never built the sales function. They grew on referrals to whatever ceiling referrals will take them, and they've hit it. Now they need a process that doesn't currently exist anywhere in their business.
That means the right next step is rarely a smarter sequencer or another AI tool. It's a target list. A real script. A voice on the phone that can react when someone says "I'm not interested" without folding. The unsexy stuff.
The second pattern: unit-economics anxiety
About half the calls had a version of the same concern. The cleanest one came from a founder selling a $297 monthly AI product to salon owners. "I'm not charging $5,000. I'm charging $297. The fact that I have to sit down and talk to them and follow up is just not..." He trailed off. He wasn't objecting to my price. He was objecting to the idea of a sales process at his ticket size.
That objection lives or dies on whether the founder has done their LTV math. Most haven't. If a $297 customer stays 24 months, that's a $7,128 lifetime value. A $40 cost of acquisition becomes laughable. A $2,000 cost of acquisition becomes reasonable. Without that number, every sales investment looks expensive, and the founder stays the bottleneck.
If you're a founder reading this and you've been told outbound is dead or needs AI before it'll work, slow down. Most of the founders I talked to this week don't need a smarter sequencer. They need a starting line.
One piece worth your time this week
Jason Lemkin published a piece on Friday titled "Forget 'Would You Hire Them Again.' The Real Test Now Is Whether You'd Replace Them With an Agent." SaaStr, May 22.
The thesis. The standard management test for years has been "given what you know now, would you hire this person again?" Lemkin says the 2026 version is "would you replace them with an agent?" He runs SaaStr with three humans and 20+ agents and says revenue went from -19% to +47% YoY after the shift.
Where I'd agree with him. The work humans refuse to do is real, and that's where AI agents earn their seat. Following up on small inbound leads. Re-engaging ghosted prospects. The 15% of revenue an SDR was leaving on the table because she was force-ranking deals in her head and dropping the bottom of the list. Easy win.
Where I'd push back. SaaStr sells content, community, and a once-a-year event to a self-selected audience that already wants to be there. That is not the same operating environment as a $20K services sale to a CFO who's never heard of you and isn't sure what you do. On those calls the human voice and the ability to react to an objection in real time is the entire game. An agent doesn't get the meeting in that world. A trained operator does.
The "replace them with an agent" frame works for the work nobody wanted to do anyway. It falls apart on the work that needs a human to read another human in real time.
That's the read for this week.
If anything in here landed, forward it to another founder or VP of Sales who would push back on it.
Justin