By Bryan Knipe, founder of TriPinnacle Advisors
Most growing companies hit the same wall. Revenue goes up, and so does the headcount behind it: another salesperson's worth of admin, another onboarding specialist, another analyst to keep the reports coming. Margins get thinner as the company gets bigger, and every plan for the next stage starts with a hiring list.
It doesn't have to work that way anymore. Most of the work that grows with every new customer is reading, writing, looking things up, moving data between systems and following up. AI can now do that work. Move it before the next hiring round, and the company grows on the team it has, with your people on the work that wins and keeps customers.
I have run these functions myself, as a general manager, COO/CFO and sales leader, and I build the AI that runs them. I also run my own software company this way: as one person, with AI agents doing the work a team would otherwise do. This is how I approach it.
1. Find the work that grows with revenue
Start with a simple question for each function: if revenue doubled, how many more people would it need? Sales operations, customer onboarding, account management, support, research and list-building, proposals, reporting to owners and customers: in most companies these grow in step with revenue. Finance and legal grow more slowly. Leadership barely grows at all. The functions that scale with revenue are where AI pays for itself first.
2. Split the work inside each function
Inside a function that scales, split the work into three kinds: the volume work (data entry, research, first drafts, status updates, follow-ups), the checking work (making sure the volume work is right) and the judgment work (deciding, persuading, handling exceptions). AI takes most of the volume work and some of the checking. People keep the judgment. In most sales-support and onboarding teams, the volume work is the majority of the hours.
3. Move it before you hire, not after
The cheapest moment to change how a function runs is before you add people to it. Once a team is hired around a manual process, the process is defended. Put the AI in when the hiring plan says the next person is needed, and measure whether the person is still needed after it runs for a month.
4. Build it into the real systems
The AI has to work inside the CRM, the ticketing system, the shared drive and the email the team already uses, on real data, with a person checking its output until the error rate is known. A demo on a sample proves nothing about growth. I build it into the daily work and run it there until it holds.
5. Put people where growth is decided
The point isn't fewer people. It's the same people spending their time on what moves revenue: talking to customers, closing, fixing the hard problems, building relationships. Tell the team that up front, and show them the work that comes off their plate before anything else changes.
6. Measure revenue per person
Pick one number and report it every month: revenue per person, or revenue per person in each function that scales. If growth comes without the matching rise in headcount, the number climbs. If it doesn't climb, something is still being done by hand that shouldn't be.
Where I help
I do this as a forward-deployed operator for growing companies and private-equity owners: I map which work scales with revenue, decide what moves first, build the AI into your systems and run it until it holds, part-time alongside your team or as an interim COO.