Four years of running my own stores

I've been running my own ecommerce stores since 2009. I was a marketing major for about two years across two community colleges before that, which I mention mostly because of the gap between the two experiences.

The classes weren't bad. They were about a market that no longer existed. Nothing I was taught prepared me for the specific feeling of watching $200 leave your account in an afternoon in exchange for four visitors and no orders, which is the only lesson in this business that ever actually stuck.

What buying your own traffic teaches you#

There's a category of knowledge you can only get by having your own money at risk, and it isn't the tactics. The tactics change every eighteen months and always have. It's the calibration.

You learn how long the feedback loop really is. Everyone talks about ecommerce like it's instant. It isn't. You need enough spend for the numbers to mean anything, and the number of days that takes at a budget you can afford is a lot longer than the dashboard implies. Most of the bad decisions I've made were made on three days of data that I treated as a verdict.

You learn that the product decides most of it. I spent a long time believing I could buy my way into anything working. You can move a mediocre offer some amount with good creative and good targeting. You cannot move it far, and the distance you can move it is much shorter than the distance between a mediocre offer and a good one. That was an expensive thing to find out one $200 day at a time.

You learn what a real margin is. Not gross margin. What's left after the ad cost, the payment processing, the shipping you underquoted, the returns, and the thing that broke. The first two years I thought I was profitable because I was doing the arithmetic that made me profitable.

You learn that the platform is a participant. Whoever you're buying traffic from is running their own auction for their own reasons and changes it when they feel like it. A thing that worked in March stops working in June and there's no announcement and no appeal. I don't think I'd internalized what that means yet. I've just noticed it happening repeatedly.

Why I'm not a developer and took the classes anyway#

I'm not an engineer. I've taken the courses and worked through the tutorials, but nobody should hire me to write their software.

I did it for a specific reason: I wanted to be able to tell whether the people I hire are doing good work. That's it. You don't need to be able to build the thing to know whether the person building it is serious, but you do need to know enough vocabulary to ask the second question, and to notice when the answer to the first one was evasive.

I think a lot of non-technical founders skip this and pay for it in a currency they can't see. You end up in a relationship where every estimate is accepted, every delay is unquestionable, and every "that's not possible" ends the conversation. That's not a technical problem, it's a leverage problem, and a few hundred hours of tutorials fixes most of it.

What I'm doing with it now#

Exposely is the first thing I've built that isn't my own store, and the useful part of the store years turns out not to be any specific skill. It's that I've been the customer. When a brand tells me their acquisition costs are getting worse and they're looking for something that isn't a Facebook auction, I know exactly what room they're sitting in, because I sat in it with my own money for four years.

That's most of what I have. It's more than I expected it to be worth.