Hard work doesn’t always equal high value

It took me some years until I realized this 80/20 rule, where we should focus on what brings 80% value with 20% effort. Thinking back, it is such a universal truth.

Recently, I wrote software for a Telit Cinterion SGL81W modem. It was a simple program that sends data between an RS232 port and a TCP socket connection. The program itself took some time to be production-ready. It was written in C and uses ThreadX plus some NDA libraries, which I cannot comment on.

This modem is then used as a module in an elevator my company produces. The elevator itself has tens of boards communicating with each other using the CAN protocol and lots of complexities. This modem serves as a way for us to have remote control and, most importantly, get analytics from the elevator. It makes it "smart" since it reports any problems, such as people trapped, restarts, operation issues, CAN noise, missing floor targets, and so on.

An elevator has many parts, and the production process is pretty complex. The production crew has a lot of hard work to do to prepare all pieces, boards, and so on. It is a complex process, and the simpler we make it, the better.

We needed to develop new modem software because the previous model we were using is no longer commercialized, thus requiring us to use a new one. The process to prepare the previous modem was a pain. A guy had to connect it to his computer, then send a bunch of AT commands to prepare the modem and install the software, then more AT commands to check if it was okay, then proceed to attach the modem to the elevator board for more testing.

The process was prone to errors, and it could easily be automated, which is what I did. I wrote a simple C# GUI app using Avalonia that just installs the modem and checks if everything is okay. It took me a single morning to write it, while the modem software took me three months (due to lack of documentation). I also put some hours into trying to add specialized logs to the modem, offline buffers, and so on. Nothing had any real visible value compared to the installer I made in a single morning.

It made the production crew's job so much easier and reduced a lot of wasted time since now they just have to press a button. It brought much more value than I thought it would, and it was the thing people noticed. They didn't care about the modem software at all; there was nothing innovative in there, nothing that brought new measurable value.

This pattern repeated many times during my career. After putting lots of effort into products and complex routines, most of the time, what made people's eyes shine was simple stuff made in a short amount of time. That's why now, whenever I start something, I ask myself, what is the minimum I can do that will bring the most possible value?