“Why not” is not a reason

I argued against putting a power-user feature in the first release. I made the case in three different ways but the team still wanted it, and it wasn't my call. I specced it anyway.

I still think I was right. The argument I lost to is the reasonable-sounding kind that's genuinely hard to answer in the room.

“It’s a small amount of work, why not have it.”

That's hard to push back on, because the first half is true. It is small. Small is a fact about cost, not about priority. Two days of work is still two days not spent on more important things, like getting the first users.

A good product team optimises for the problems they want to have.

The user interviews put the feature somewhere around nice-to-have. The market research suggested it might cost us trust rather than effort, which is the more expensive thing to ask a new user for. Everything pointed the same way. It just didn't point hard enough.

What I thought at the time

I was a bit annoyed and slightly disappointed, I admit. Mostly at myself, since I couldn't make the argument land, and this wasn't the first time. I'd been in the same conversation a few years earlier on another platform, with someone who believed the product would be good if it had every cool idea implemented. Even recognising the pattern didn't help me explain it any better.

What I did

I specced it anyway. As always, I ordered the parts so the most impactful ones get built first. Then I made a second, smaller argument: test whether the feature gets used at all before going deep on the rest of it. The team agreed to that.

It wasn't a containment strategy. It does mean the expensive part has to be wanted before anyone builds it.

The part I'm less sure about

In my opinion, shipping something and then taking it away if it goes unused is worse than never shipping it. People notice things disappearing; they don't notice what was never there. And you've spent the development time either way.

What I learned

The argument I lost was a big one. The argument I won was small: not don't build this, but let's find out whether it's used first.