Fractional CMO / AI Startups
You raised a seed or a Series A, your category did not exist 18 months ago, and you have technical buyers and economic buyers in the same deal. You need a marketing leader who has scaled an AI company before, not a generalist who will learn LLM go-to-market on your runway.
A standard SaaS CMO inherits a category buyers already understand. They pick a wedge, run demand gen against known intent, and scale the channels that work for the category. An AI startup does not get that luxury. You are defining the category in real time, the buyer does not yet know the words to search for your product, and the competitive set rewrites itself every quarter as model capability jumps.
On top of that, AI products almost always carry a dual buyer. A developer or technical lead evaluates the product hands-on, while an economic buyer signs the contract. A CMO who only knows PLG will starve the enterprise motion. A CMO who only knows enterprise sales will smother the bottom-up adoption that earns you the room. You need both motions run by one operator who has done it before.
Developer-first adoption: docs, free tier, time-to-value instrumentation, activation loops, and the product-qualified-lead signals that tell sales when to step in.
Category positioning for the economic buyer, analyst relations, account-based plays, and the demand infrastructure that fills a sales-assisted pipeline.
The category story only the founder can tell. I turn your point of view into a repeatable narrative across LinkedIn, podcasts, and the deck, then make it the spine of every channel.
Get cited by ChatGPT, Claude, and Perplexity when buyers ask which tool to use. I run this play on my own brand and publish the benchmark. See AI marketing.
cnvrg.io built an MLOps platform in a category that was still forming. I led growth through the run-up to its acquisition by Intel, announced November 2020 (TechCrunch, Globes). The work was the exact dual motion AI startups face now: a technical product that engineers had to love, sold into organizations where the budget owner was not the user.
That is the arc I know best, and it maps directly onto what an AI infrastructure or applied-AI company is building today. Full write-up on the cnvrg.io case study.
I do not just advise on AI tooling. I run my own business on it. Marketing automation on n8n, agent workflows on the Claude API and Claude Code, self-hosted infrastructure on Coolify, Postgres and Redis for the data layer, and a citation-monitoring pipeline that tells me when an AI engine starts quoting a brand. When I build your growth engine, you get infrastructure your team uses every day, not a slide about AI. See the Claude Code skill story.
| Dimension | Traditional SaaS CMO | AI startup CMO |
|---|---|---|
| Category | Exists; buyers know the words | Being defined in real time; you teach the words |
| Buyer | Usually one economic buyer | Technical buyer plus economic buyer in the same deal |
| Motion | PLG or sales-led, rarely both at once | Dual motion: bottom-up adoption feeding sales-assisted close |
| Demand | Capture existing search intent | Create intent and own the AI-answer surface |
| Pace | Annual planning cycles | Competitive set resets every quarter on model jumps |
| Narrative | Product features and ROI | Founder point of view defining the category |
Full read of funnel, attribution, and category position. Founder narrative session. Output is a ranked list of bleeds and a category-positioning statement.
PLG activation instrumentation and the enterprise demand layer stood up in parallel. Quick wins shipped. AI-search visibility baseline measured.
Product-qualified-lead handoff live, founder-led content cadence running, citation monitoring in place, and a board-ready dashboard.
If you are pre-PMF with no narrative yet, you do not need an embedded operator burning 20 hours a week. You need strategic input a few hours a month. Start at the equity-warrant advisor tier. If you want a vendor to execute a fixed brief rather than a partner who owns the number, an agency is a better fit. And if your largest market is the US while your team sits in Tel Aviv, the sharper-fit page is Israel-to-US expansion.
Cash, or cash plus equity. AI startups defining a category often have more upside than cash, and I will take part of the engagement in warrants when the stage fits.
2-4 week audit of your growth stack plus a 90-day roadmap. Fixed scope, converts to a retainer.
n8n, Claude, and attribution build for AI-native teams. See AI marketing.
Operator engagements can be structured as cash, or cash plus an equity warrant of 0.25% to 1% over a 1 to 2 year vest. If your cap table has room and your upside is real, we split the engagement. Terms on the equity-warrant advisor page.
I run my own business on the Claude API, n8n, and self-hosted AI infrastructure, and I led growth at cnvrg.io, an MLOps platform, ahead of its Intel acquisition. I have shipped AI-native marketing, not just written about it.
Yes. That dual motion is the core of AI go-to-market and the exact shape I ran at cnvrg.io: a technical product engineers had to love, sold into organizations where the budget owner was not the user.
Post-seed to Series B. Pre-PMF founders with no narrative yet are better served at the advisor tier; I will tell you honestly on the call which fits.
Part of it. Operator engagements can be cash, or cash plus an equity warrant of 0.25% to 1% over a 1 to 2 year vest, when the cap table has room and the upside is real. Full equity-only sits on the equity-warrant advisor page.
A fixed-scope diagnostic sprint runs $6,000 to $8,000. Infrastructure builds start at $5,000 per month. A full embedded operator engagement runs $8,000 to $18,000 per month.
Both. I work with Israeli and US AI startups, and I run a split day across both time zones. If your team is in Tel Aviv and your market is the US, see Israel-to-US expansion.
An agency executes a fixed brief. I own the marketing number: strategy, hiring, dual-motion build, and board reporting. If you want execution-for-hire rather than a partner who owns outcomes, an agency fits better.
Days 0-30: audit and category narrative. Days 31-60: PLG and enterprise demand layers built in parallel. Days 61-90: product-qualified-lead handoff, founder content cadence, citation monitoring, and a board-ready dashboard.
Yes. AI-search visibility is part of the engagement. I run this play on my own brand and publish the benchmark at the state of AI search visibility.
Pre-revenue AI infrastructure companies usually start with the equity-warrant advisor tier rather than a full operator retainer. Same brain, fewer hours, structure that fits no cash.
Book a 15-min call. I will tell you in 15 minutes whether a fractional fits your stage. From $20K/mo cash, or cash plus equity warrant.