The question we keep getting asked
We get some version of this question a lot: in the middle of all this AI noise, what actually hasn't changed about selling, or about building a company at all? The honest answer runs long. Here's the short one.
Two things, really. People buy from people, especially once the price tag gets serious. And people buy on emotion, then justify the decision with logic afterward. Neither is a new idea — Kahneman wrote the book on the second one, and it had nothing to do with sales. But every time a new technology shows up promising to change everything, it's worth checking which parts of the old picture actually depended on the technology, and which parts depended on being human.
Formats change. Behavior mostly doesn't.
Television didn't die. It became streaming. Newspapers didn't die either, not entirely — they became something read on a phone instead of folded on a table. The format changes. The underlying behavior — wanting to be told a story, wanting to know what's happening — mostly doesn't. The same split is playing out in sales right now, and it's easy to miss because the format change is so loud.
Here's what we mean by loud: AI has made it dramatically easier to build things. More apps ship every month than ever before. And at the same time, by most measures, App Store rankings and real adoption have moved the other way — more competition for the same attention, not more winners.
That's not a contradiction. It's what happens when the cost of producing something drops faster than the cost of earning trust in it. If ten teams vibe-code ten similar apps this month, the deciding factor for which one actually gets used isn't feature count. It's which one you trust, or which one someone you trust pointed you to.
Why trust gets more valuable as everything else gets cheaper
That's the part that doesn't automate away. If anything, it gets more valuable exactly as everything else gets cheaper and faster to produce. Brand, real content, and relationships nurtured over time were never just nice-to-haves — they're about to matter more, not less, precisely because the volume of everything else is about to go up.
Think about it from the buyer's side. When outreach was expensive to produce, volume itself was a weak signal of intent — nobody sent ten thousand hand-written letters. Now that a plausible-sounding message costs nothing to generate at scale, volume stops signaling anything at all, and the things that still cost something to produce — a real relationship, a specific and accurate reference to what's actually happening at a company, a person willing to show up and be accountable for what they said — become the only remaining signal worth trusting.
Three things AI doesn't replace, at least not yet
- Events. A real handshake and eye contact aren't something a message substitutes for. Shared physical presence carries information — tone, hesitation, genuine interest versus polite interest — that doesn't survive translation into text.
- Calling. Cold or warm, it doesn't matter. A human voice is a different kind of signal than text on a screen. It's harder to fake enthusiasm or certainty in real time than it is to write it, which is exactly why it still works.
- Nurturing the people who've already been in touch, even if the last conversation went nowhere. Someone who replied once, even briefly, even a while back, isn't a stranger anymore. That's a real relationship, even a thin one, and it's worth more than it looks like on paper — precisely because most competitors treat that contact as cold again the moment the thread goes quiet.
Why this matters more right now
This matters more right now than it might have a few years ago, because sales cycles are stretching out almost everywhere. Budgets are tighter. People are slower to decide and slower to reply, and that's mostly an economic reality, not an AI one.
But it changes the math on outreach. Ten thousand more cold emails don't move the number when everyone's being careful with their time and budget. Reaching people who already have some context on you, and being useful to them on whatever timeline they're actually on — that does. A slower buyer isn't a less willing buyer; they're a buyer who needs more reason to believe this is worth their limited attention right now, and reason to believe is exactly what generic volume can't manufacture.
None of this is an argument against AI
None of this is an argument against using AI well. It gets used constantly here, for a lot of things, and it's genuinely changed how fast ideas can get built and tested — detecting the right accounts, drafting the first version of a message, keeping a sequence running on schedule. It's just that the thinking part — understanding the customer, being patient with the relationship, showing up as a person rather than a volume play — has gotten more important as the automatable part got easier, not less.
AI does the detection and drafting, a human makes the call and has the conversation. That's the same split covered in where AI should (and shouldn't) run your sales process — and it's the part worth betting on for the next few years, regardless of which model or tool wins the current news cycle.