What if the big site-speed companies were just too slow to notice the small customers?
That was the idea.
The established tools were owned by large companies and aimed at large companies. They were useful, but they were also expensive. A startup that wanted basic comparative performance data could spend hundreds of dollars a month on a single monitoring location.
There had to be room below that.
I sent a few friends the rough version of the product I thought should exist. Start with good worldwide page-load data. Make the visualizations better than another pile of dashboards and waterfall charts. Add monitoring for uploads and downloads, which the existing products did not handle very well.
Then make the whole thing work through a clean API.
The API was the piece I kept coming back to. A customer with its own application could send live events, transactions, or logs into the same system. The product would not only watch a website from the outside. It could connect what the customer knew with what the monitoring service saw.
This was not a company. There was no deck, no name, and no fundraising plan. It was an email sent after some code was finished because the opening looked real enough to keep pulling on.
That was how a lot of side projects started for me. Find a product I used at work. Notice where it was bad or overpriced. Ask whether a smaller team could make a version that was simpler and more useful.
Most of those questions never became businesses.
The habit still mattered. I was running infrastructure at Shutterstock, but I was always looking sideways at the tools around us and wondering who should build the better one.