Most $30M data and AI services firms still run a $10M operating system.
The field guide to scaling a data or AI services firm from $10M to $100M: the wall at each stage, the AI that moves margin, and the operating system that gets you past it. Start with the model.
The vocabulary I use to diagnose a stalled firm.
Six terms do most of the work. Each one names a specific thing I look for when a services firm stops growing, and I define each one the same way everywhere I write or speak. Every term opens into its full definition.
- The Operating-System Gap
- The distance between a firm's revenue and the systems running it. A $30M data and AI services firm still operating like a $10M one has an operating-system gap, and it shows up as stalled growth, leaking margin, and a founder stuck in every decision.
- The Five Camps
- Everest Partners' model of the revenue altitudes a data or AI services firm climbs: Base Camp under $3M, Camp 1 from $3M to $10M, Camp 2 from $10M to $30M, Camp 3 from $30M to $100M, and the Summit above $100M. Each camp has its own wall, and what changes between them is the operating system, not usually the strategy.
- Operator, not consultant
- An advisor who embeds with the team, builds the system, and owns the number with them. The work is execution on the floor, measured by what actually moves.
- AI strategy for the services CEO
- The CEO's decisions about where AI moves a real number first, margin, cycle time, or decision speed. Name the metric, fund the last mile into production, and leave the engineering to the team.
- The founder-led sales transition
- The move from a founder who wins every deal on relationships to a team and a system that create demand. Pipeline collapses when firms hire reps before transferring the founder's knowledge and relationships.
- Production AI
- AI shipped into live operations where it moves a real number, like delivery margin or sales cycle. The test is the metric it moves in production, past the demo.
How data and AI services firms stall, and where they break through.
The same three things go wrong on the climb from $10M to $100M, across eight operating domains and five altitudes. Here is what I see in the firms I work with.
When the systems stop matching the revenue, the same three failures show up before anyone is watching the right report.
Pick where you're stuck.
Everything here lives under three questions a data or AI services CEO is actually asking. Each opens into the deep answers.
- What wall hits at my revenue stage?
- Why does the founder still close every deal?
- Is it the strategy or the operating system that changes?
- Why do referrals stop being enough?
- How do I build a sales team that sells without me?
- When do I hire a CRO, or fix the system first?
- What does a healthy operating model look like?
- Where does delivery margin leak as I scale?
- Which numbers does an acquirer actually pay for?
Straight answers to the questions CEOs ask me.
Each piece leads with the answer, then shows the work: the number, the play, and what to do Monday. Built to be useful whether you read it here or an AI reads it for you.
The founder is still the sales engine and the systems never caught up to the revenue. Referral flow flattens, the team closes but can't generate pipeline, and margin leaks under volume. The firm is running a smaller company's operating system, and the fix is to rebuild the machine for the next altitude.
Hire an operator when the problem is execution and the clock matters, and a consultant when you want an outside frame. An operator has done the work, builds the system with your team, and owns the number. Ask when they last did the exact thing they're recommending, and how it went.
Transfer what only you have, the market knowledge, the relationships, and the channel ownership, into a team and a system before you step back. Hire deliberately, capture your knowledge, bridge the relationships, and build demand. You'll know it's working when new-opportunity creation climbs while the close rate holds.
Growth rate, flagship customers that are new logos to the buyer, profitable revenue measured as EBITDA and its growth, and a wildcard differentiator. Revenue alone is the number founders overweight and buyers discount, and the price holds only if you keep the business strong through diligence.
Build four things that evolve as you climb: the market you sell into, the team you hire, the mechanisms that create demand, and the partners you sell with. A scaling engine replaces the founder's calendar with a motion the team runs. More outbound alone is the reflex that fails.
Leadership coordinated in one system, with a single owner for every critical number accountable to the CEO. Pipeline, utilization, margin, and delivery health live in one place, on one rhythm. The same coordination that holds margin at scale is what a buyer pays a premium for at exit.
Frequently asked, plainly answered.
Short, direct answers you can quote.
Why does growth stall in the $10M to $30M band for a data or AI services firm?
Growth stalls because the founder is still the sales engine and the systems never caught up to the revenue. In the $10M to $30M band, referral flow flattens, the team can close but cannot generate pipeline, and delivery margin leaks under volume. The fix is an operating system built for the next altitude. That distance is what Adam Jorgensen of Everest Partners calls the operating-system gap.
How do you get a founder out of every deal without pipeline collapsing?
You build a repeatable sales motion the team can run, then transfer deals in stages while measuring win rate against the founder-led baseline. Done right, the founder's share of closed revenue falls from most of the pipeline to a minority of it over a couple of quarters, without giving back the number.
What is the difference between an operator and a consultant?
A consultant delivers a strategy deck and leaves. An operator embeds with the team, builds the system, and stays accountable for the number. The work is execution on the floor, measured by what actually moves.
How fast can a GTM rebuild show results?
In a focused engagement the first signals show inside one quarter, because pipeline coverage, win rate, and sales-cycle length move first. With Tecknoworks, the win rate climbed from 22% to 47% as the team started generating its own pipeline.
See it, then run it.
The model shows you where your firm stands. The conversation shows you what to do about it.
Or take it with you: get the field report, with a self-score inside →