I am old enough to remember the Radio Shack that was really a radio shack. This was the place where kids in my Dad’s generation would go to build or repair ham radios. And though I am quick to admit that I was never an electronics whiz kid I knew a resistor from a capacitor and I knew that you could find just about any electronic component in a Radio Shack. When “Hi-Fi” stereo became a fixture in the sixties the Realist product line of receivers and speakers became a solid middle-of-the road staple in many a living room. When the Altair 8800 became the world’s first consumer personal computer the TRS-80 (aka “Trash 80”) was a hacker’s toy years before IBM released the PC. In fact Radio Shack was the world’s largest manufacturer of PC’s before Michael Dell brought lean manufacturing to the computer.

Perhaps it was Radio Shack’s ascent into becoming the largest electronics retailer around 1990 that doomed it. The rise of the big box stores sealed its fate. What makes companies great is not how they rise to prominence but how they respond when—as it always does—someone blows up their business model. IBM sold off its PC’s to Lenovo and made “bank” with software as a service. Long after their obituaries were written in the eighties, Rolex and its fellow European watchmakers broke free from the digital watch masses and got rich selling to affluent buyers who craved these little mechanical marvels.

But Radio Shack ignored the writing on the wall. It thought it could be a Circuit City when even Circuit City couldn’t make it as a Circuit City. Computers? Radio Shack had them. Cell phones with contracts? Check. RC cars? No problem. Going to a store was a painful experience. Being asked for my phone number and email address when I bought a battery gave me the creeps. The electronic components that made it an institution were banished to one wall in the back of the store. Radio Shack was a dead man walking in 1991 but like a guy with lung cancer who just keeps smoking, it partied like it was 1980. It could have carved out a nice little niche for the do-it-yourself crowd that now includes the Arduino generation. It would never be a billion dollar niche but it would be a sustainable one.

It’s easy to repeat Radio Shack’s mistakes. Sears has been doing so for decades. For all intensive purposes, they went out of business about the same time as Radio Shack. They just don’t know it.

When I purchased AquaMetrix I almost did it too. When I asked our customers what they wanted they told me they wanted a one-stop shop. They wanted to buy every sensor on a shopping list from one place. I wanted to be that place. We made pH, ORP, conductivity and dissolved oxygen sensors—and the controllers that go with them—but we lacked sensors for flow, level, free and total chlorine, turbidity, and optically based dissolved oxygen. I searched for manufacturers who would private label all of these sensors and found them. The only problem was that the margins on many of these items sucked. My business model was flawed. To add insult to injury the overhead required to order these pass-through items and provide proper technical support ate whatever slim margins existed. We’re not Hach, Rosemount nor Endress & Hausser. We’re not a one-stop shop—at least not yet—and trying to behave like one was like Radio Shack trying to be Amazon.

Going into 2015 I made the decision to axe items—ours or anyone else’s—that didn’t give us a sustainable margin or for which we could not provide top tier support. Sounds easy enough but when a customer calls with an order that’s worth thousands and when we’ve built a culture that wants to always say “yes”, saying “no” is hard to do. Still we pared down the list to a manageable few that made sense