Notes · Entrepreneurship

What I've Learned After 3 Years Running a Dev Agency

The Twitter version of running an agency is "productized services," a tidy funnel, and revenue screenshots. The real version is quieter, slower, and mostly about expectations. Here is what three years of Vantage Method actually taught me.

The work that wins clients isn't the code

I came in assuming the engineering would be the hard part. It almost never is. Clients do not buy clean architecture — they buy the feeling that someone competent is going to make their problem go away. The agencies that lose are usually shipping fine code while communicating badly. Code is table stakes. Confidence, clarity, and follow-through are the product.

Scope is the whole game

Every painful project I have been part of was a scope failure before it was anything else. Not a technical one — an agreement one. "Just a small change" is the most expensive sentence in this business, because it is never one change and it is never small. The skill that separated good years from bad ones was learning to say, kindly and early, "that's a new phase, here's what it costs," instead of absorbing it and resenting it later.

Retainers beat heroics

Project work is a treadmill: you are only as stable as next month's pipeline, and you spend your best energy selling instead of building. The shift that changed the business was moving clients onto ongoing relationships — site, CRM, automation, the stuff that needs tending anyway. Predictable revenue does not just pay the bills more calmly. It lets you do better work, because you are optimizing for the client still being here in a year instead of for the invoice in front of you.

Engineering rigor is the actual differentiator

Most small businesses have only ever worked with template shops and marketing freelancers. When they meet a team that treats their site and their lead flow like a system — versioned, monitored, debuggable — it is genuinely novel to them. We do not win by being the cheapest or the flashiest. We win by being the people who shipped real software and bring that discipline to a $4k website. That gap is the whole pitch, and it is on the Vantage Method page on purpose.

The unglamorous parts nobody posts about

Chasing invoices. Re-explaining the same decision three times. The client who goes dark for two weeks and then needs it "by Friday." Writing the same onboarding email enough times that you finally automate it. None of this shows up in the highlight reel, and all of it is the job. Getting comfortable with the boring operational layer is what makes the creative layer sustainable.

What I'd tell someone starting one

Pick a narrow kind of client and get unreasonably good at their specific problem. Charge in a way that survives a slow month. Put everything in writing — not because people are dishonest, but because memory is. And treat the relationship as the deliverable; the website is just the part you can screenshot. The technology was never the hard part. It was learning that the business is mostly people, and people run on clarity.

More on the agency itself on the Vantage Method page, or head back to all notes.