🎃 We'll add it to the backlog

Last Tuesday I grabbed a cappuccino at ViCAFE, the tiny kiosk next to Bürkliplatz. On the tram I ended up chatting with one of our power-users. He rattled off half a dozen “game-changing” tweaks for the app. I nodded, said it sounded great, and—almost by reflex—promised we’d toss the ideas into the backlog.

Elevator ride, 8:42 a.m. The CEO, eyes still glittering from some late-night YouTube rabbit hole, announces that Zuckerberg’s Metaverse speech “changes everything.” Could we look into porting our cooking app to web3? I mumble the magic phrase—“sure, we’ll add it to the backlog”—while wondering what on earth a lasagna recipe has to do with digital land plots. (I’m still drawing blanks.)

I finally reach my desk. Before the laptop even boots, one of the senior devs appears, equal parts tired and furious. Tech debt has mutated again—database migrations piling up, CI taking 30 minutes, tests in free fall. I sympathize, swear it’s top priority, and, yes, promise to drop it into the backlog. The pattern is painfully obvious even as I say it.

I crack open Jira. The current sprint looks like a game of Tetris played by someone who only got the straight pieces—every column maxed out, no wiggle room. Realistically, nothing new fits without pushing deadlines or bodies past breaking point, yet I’ve just committed to three brand-new initiatives.

A few weeks slip by. Those three tickets are now fossilized at the bottom of the board. Then the inevitable ping: the CEO wants an “update on the Metaverse direction.” The user wonders if we forgot him. The dev reminds me the test suite is still smoking. I could pretend to be surprised, but I’ve seen this movie enough times to know the plot.

Parks and Recreation nailed the dynamic years ago. April scheduled every unwanted meeting on March 31—then discovered March 31 actually exists. Ninety-four meetings detonated in a single day, and Ron Swanson almost set fire to City Hall. Different show, same lesson.

My “we’ll add it to the backlog” sounded harmless in the moment. What everyone else heard, though, was, “Cool, someone will size this next sprint and ship it by Q-end.” That misalignment is the whole catastrophe.

These days I try (emphasis on try) to say no early, politely, and with context: “Love the idea, but we don’t have capacity until Q4. Let’s gather data and revisit.” It’s clumsy, occasionally awkward, but miles better than silent backlog exile. I could be wrong, yet so far stakeholders seem to prefer an honest no over a fuzzy yes, though it didn't fully fix things.

At mindnow we brag about the features we build, but we also brag—maybe louder—about the features we keep out. Guardrails count as progress.

Not every product belongs on a blockchain.

Not every product needs NFTs.

Not every product benefits from gamification.

Not every product should wear the “machine learning” badge just because the conference slides look cooler.

Other Newsletter Issues: