There’s a reason email is so addicting. Like a roulette wheel or a slot machine, there’s always a chance of something exciting coming your way. On the other hand, there’s also an equal chance that terrible news is coming your way in any given piece of electronic mail. The kind of email that makes it hard to sleep at night as your heart races and anxiety kicks into high gear.

That was the kind of email we received from Loren Brichter when he had decided to sell his company to Twitter.

In an interview with Indie.vc (published on Medium. natch), Matt Haughty gives some “lessons on lifestyle business”. As I read the piece, it brought me right back to the days of running Fusion Ads. Matt’s experience was very similar to our own.

For a year and a half, I waited. The revenues kept trickling down. It was this long terrible process, losing half overnight but then also roughly 3% a month for a year and a half after.

We didn’t not have a precipitous drop of this sort, but I do know the feeling of watching this slow decline. Matt summed it up perfectly:

Two weeks later traffic went back to how it was. But in those two years, the ad revenue market had changed. Even with the traffic, we were making half as much money because the market had kind of died in the interim.

That’s exactly what my experience was. We were blessed to get in the business in 2009. And having our ads in Tweetie for Mac (the ancestor of today’s official Twitter client for OS X) enabled our business to flourish. But, like MetaFilter, we had too much of our revenue generating traffic from one source. When that source went away, it was a punch to the gut.

Fusion Ads survived and went on for another two years before we sold it. But the writing was on the wall, the ad business was changing. We got out because it felt like a race to the bottom.

The important lesson I took away from my first business was what every investor learns about: diversification. When you run a business, you want revenue being generated from multiple sources. Hopefully, with multiple strong sources.

I vowed to never make this mistake again. And even further, I vowed to never run a business that was completely dependent on the work of others. Selling ads based on other people’s traffic worked, but it’s hard to run a business when you're 100% tied to other people’s success.

A freelancer wants clients of different sizes, with some retainers and a good mix of new work. A SaaS team wants a varied customer base with related, but slightly different needs (jobs to be done). It’s cliché, but there’s a lot of truth in the old saying. When it comes to eggs, it’s good to have a few different baskets.