The Database That Nobody Asked For
I spent three months this year building a customer feedback database that our product team swore they didn’t need. They had spreadsheets, they said. Color-coded ones. What more could you want?
But I kept watching our support team drown in those same spreadsheets during our weekly triage meetings. Sarah would pull up one file, then another, then realize the data she needed was in last month’s version, which lived in someone else’s Google Drive folder. By the time she found what she was looking for, the meeting was over and the customer complaint was three days colder.
So I built it anyway. Called it a “learning project” in my performance review, which is corporate speak for “I’m doing this whether you like it or not.” Three weeks after launch, the product team was using it daily. Sarah sends me coffee shop gift cards now. Sometimes the best projects are the ones nobody thinks they want until they have them.
The Newsletter That Almost Wasn’t
Our internal newsletter died sometime in March. Not officially, it just sort of faded away like a houseplant everyone forgot to water. People would ask about it in Slack occasionally, usually followed by shrugging emoji and someone changing the subject.
I volunteered to revive it, mostly because I was tired of hearing the same questions repeated across five different channels. “When’s the next all-hands?” “Did anyone see that article about our competitor?” “Why is the coffee machine making that noise again?” Basic stuff that used to live in a shared space but had scattered into a dozen private conversations.
The trick wasn’t writing it, it was making it feel human instead of corporate. I started including photos from people’s desks, weird overheard conversations from the kitchen, actual updates about projects instead of management buzzwords. Response rates went from zero to 89% in two months. Turns out people will read almost anything if it doesn’t sound like it was written by a committee.
The coffee machine noise mystery got solved in week three, by the way. Loose screw in the grinder. Tom from facilities fixed it in ten minutes, but only after I mentioned it in the newsletter and half the office started leaving sticky notes on the machine with theories.
The Training Program That Worked Too Well
Training new hires used to be pure chaos. Not the fun kind, the expensive, inefficient kind where people would spend their first week asking the same questions that the person before them asked, and the person before that. We had a wiki, technically, but it was organized like a fever dream and about as helpful as a chocolate teapot.
I created a structured onboarding program that paired new people with mentors and broke everything down into digestible weekly goals. Nothing revolutionary, just basic human psychology applied to workplace reality. Week one: learn the systems. Week two: shadow your mentor. Week three: start taking on small projects with backup.
It worked so well that HR asked me to expand it company-wide. Then other departments wanted their own versions. Suddenly I was running training sessions for people who’d been here longer than me, teaching them how to teach others what they already knew. Meta-training. The kind of project that starts simple and grows tentacles when you’re not looking.
The best part was watching the mentors discover they were better teachers than they thought. Jennifer from accounting, who swore she “wasn’t good with people,” ended up leading the most popular sessions. Turns out expertise and patience are a powerful combination, even if you don’t realize you have both.
The Process Nobody Wanted to Document
Every company has that one process that lives entirely in someone’s head. Ours was client onboarding, and that someone was Marcus, who’d been here since the office had carpet instead of concrete floors. Marcus knew everything, remembered everyone, and had exactly zero interest in writing any of it down.
The problem became obvious when Marcus took his first vacation in three years. Client calls went unanswered. Contracts sat in limbo. The rest of us scrambled around like chickens with our heads cut off, trying to reverse-engineer decades of institutional knowledge from email threads and prayer.
I spent two weeks following Marcus around with a notebook after he got back, documenting every step, every exception, every “oh, and if this happens, then you do that” moment. He thought I was being paranoid. I thought I was being practical. We were both right.
The documentation ended up being forty pages long. Forty pages of processes that had been living in one person’s memory, vulnerable to vacation days and job changes and the simple passage of time. Now it’s a shared resource that three people can execute instead of one. Marcus still knows things the rest of us don’t, but at least we won’t lose everything if he decides to retire to that cabin in Vermont he’s always talking about.
What Sticks Around
These projects all have something in common, they solved problems that were hiding in plain sight. The database made visible what was scattered. The newsletter connected what was fragmented. The training program scaled what was working. The documentation preserved what was fragile.
None of them were flashy. None of them will win innovation awards or get mentioned in company press releases. But they’re the kind of work that makes everything else possible, the infrastructure that lets the exciting projects actually happen.
Looking back, I’m proudest of the things that other people now take for granted. That’s probably the best measure of whether something was worth building, not whether people notice it, but whether they’d miss it if it disappeared.
What are you working on that nobody asked for but everyone needs? I’m curious about the projects that don’t make it into status reports but somehow make everything else work better.
