Marketing an AI product without saying AI

LAST REVIEWED 2026-08 · SOURCED FROM 2 SESSIONS, MID-2026

TL;DR

When everything is AI, being human is what stands out. Sell outcomes, not the fact that it’s AI — non-technical buyers don’t care what’s under the hood, and general sentiment toward the technology is at a low. Earn the right to market before you market: small demonstrations, calculator pages, relevant case studies. When you’re disrupting a gatekept industry, build advocates and sell to the people who sell to your buyers. And the highest-converting outreach right now is the oldest kind: a personal video, a postcard, a phone call — as the fourth or fifth touch, not the first.

Nobody cares that it’s AI

The blunt version from the agency session: founders of AI-driven products walk in leading with “we built this new AI and it does amazing things,” and — especially with non-technical stakeholders — no one cares. The C-suite mandate to adopt AI at all costs has ebbed; public sentiment toward the technology is one of the few things a large majority agrees on, and it isn’t warm. The fundamentals are back:

“When everything is AI, the biggest thing to stand out is being human.”

— Marketing agency co-founder, ~20 years in B2B · session, mid-2026

Start from the endpoint — what does this net the buyer — and let the AI be the fourth or fifth conversation. The founders in the room selling successfully into heavy industry echoed it: their buyers “could care less” that it’s AI or MCP; they buy accuracy, repeatability, and audit trails.

Earn the right to market

The first contact isn’t marketing your product; it’s earning the right to. The working tactics from the session:

• Small demonstrations over grand gestures. A calculator landing page that proves you understand one narrow part of the buyer’s problem cracks the door; the platform conversation comes later.
• Case studies that match the buyer. Social proof works through similarity — a case study from a company like theirs, not a logo wall. The first ten customers are the hardest and the most valuable for exactly this reason; the agency deliberately lost money on its first five to buy the case studies.
• Demand generation before the ask. LinkedIn posts, a few ads, emails — all of it builds the perceived value that makes a $48K quote land soft. Build the value before you name the number.
• Personalization that thinks on the fly. Dynamic landing pages, curated case-study sets, tailored one-pager follow-ups — using AI in the background to make the human conversation sharper, not to replace it.

Disrupting a gatekept industry

For the founder selling against entrenched brokers — where the economic buyer isn’t empowered to change and “nobody gets fired for hiring the incumbent” — the session’s counsel converged on three moves:

1. Build advocates, not just pipeline. Go a little guerrilla: differentiate loudly, find the socially motivated champions inside target companies, and let PR around the disruption itself pull inbound pilots. The goal is people who put you in the room, not another cold call in a stack of fifty.
2. Sell to the people who sell to your buyers. Consultants, brokers, service firms with existing trust — they’re not being sold to all day, they’re free to say yes, and their referral is priceless. When your product threatens their billable hours, do the math with them: a referral or residual program that fills the revenue hole you’re creating.
3. Size the first proof right. Statistical products need a population (the benefits founder needed ~200 lives for six months of signal); a multinational moves too slowly to be your first proof. Find the mid-size sweet spot that’s big enough to prove it and small enough to decide.

The old-school channel stack

The growth lead’s channel take, which the room found almost funny in its contrarianism: the highest-performing outreach today is the personal, analog kind — a recorded personal video dropped into an inbox, a creative postcard with the value prop, an actual phone call with an actual voicemail. Everyone else deleted those plays from the book, which is why they work again. Sequencing matters: the buyer should have seen you — a genuine LinkedIn post, a piece on the issue you solve — three or four times before the call, because people now buy across five platforms, not one.

For finding who to reach: map the sphere of influence in your niche (AI agents are good at exactly this research), then look for motivated operators — consultants, the recently laid-off, industry insiders — who will hustle your deals for a path into the company. Curating the right list is the most time-consuming part of GTM; it’s also where the leverage is.

Where this connects

The same sessions’ guidance on synthetic buyer panels and pricing pages lives in Pricing and packaging; the micro-apps-over-subscriptions tooling philosophy is in The AI-era marketing stack; the strategy-level shift that makes GTM the hard part is in Surviving the SaaS repricing.

Sources

Two sessions (mid-2026) with the co-founders and head of growth of an AI-forward marketing agency, working live through three founders’ GTM challenges: disrupting the employee-benefits industry, pricing and selling heavy-industry software, and filling top-of-funnel for a vertical AI platform.