Part One: Strategy · Where demand comes from
Each added capability wins an argument with somebody. The product that does ten things passably still loses to the one that does the important thing properly.
There is always a case for one more capability. A rival has it. A prospect asked. It would look good in the comparison table. And comparison tables reward breadth: a row of ticks beats a row with gaps.
But nobody uses a comparison table to get their work done. They use the product, and what they meet there is the one thing they came for. If that one thing is slow, fiddly or unreliable, the nine others do not console them. If it is excellent, they forgive a great deal.
Doing one thing properly is harder than it sounds, which is partly why teams wander off. The first version is the easy part. The rest is the unglamorous work: the odd cases, the speed, what happens on a bad connection, what happens with ten thousand items instead of ten. That work rarely makes an announcement, and it is what people mean when they say a product "just works."
The pressure to widen is strongest where software is sold by a sales team, because a missing row in the buyer's checklist can lose a deal this month. The slow, fiddly core loses customers too, but later and more quietly, at renewal, when nobody remembers the checklist.
Dropbox began as a single thing: a folder on your computer that was the same folder on your other computers. Other products of the time did more, with backup schedules and sync rules and settings to match. Dropbox had almost no settings. You put a file in the folder and it was there, everywhere. People understood it in one sentence and trusted it within a day.
Therefore:
Name the one thing a person most needs the product to do. Make it fast, dependable and complete, down to the awkward cases, before you widen.
In practice. Ask ten customers what they would miss most if the product vanished tomorrow. If one answer dominates, look hard at how much of the last year's work went into it.
Goes with. The one thing comes from THE PROGRESS THEY WANT (2). Protecting it is why you SAY NO BY DEFAULT (23) and FIX WHAT'S THERE FIRST (25).