A conversation

Let's start with where you actually are right now, in your own rough words.

A quiet set of diagnostic questions. Pick the one closest. We can always change tack.

Where are you, roughly?

Pick the one closest to where you actually are right now.

  • Something feels off but I cannot name it
  • Users are not behaving the way I expected
  • The team is shipping but it does not compound
  • I am considering a redesign
  • I just want to think out loud — not looking for help
The threads

Follow whichever one is closest.

  • 01

    Something feels off but I cannot name it

    That kind of feeling tends to show up before the data does. Trust it for a minute.

    See if we can narrow where it lives:

    • More about the product itself — something underneath feels brittle
    • More about the team — we are moving but not aligning
    • Honestly both, in different rooms

    If it is the product — if a thoughtful stranger picked up the product fresh tomorrow, what would they see that you have stopped seeing?

    • Too many features, none of them clearly best
    • The core works but the edges are rough
    • It is hard to explain in one sentence what it is for
    • I am honestly not sure what they would see anymore

    If it is the team — is it that people disagree out loud, or that they agree out loud and then disagree quietly in their work?

    • The second — quiet disagreement, showing up in the work
    • Open disagreement, just hard to resolve
    • Could be either, honestly
  • 02

    Users are not behaving the way I expected

    When did you first notice? Was it gradual, or did something change?

    • Gradual — looking back, it has been weeks or longer
    • Sudden — something shifted recently
    • Honestly not sure when it started

    One more, and it is the one most teams skip. Have you talked to any of these users about it directly? Not a survey — a real conversation.

    • Yes, a few. I have notes I have not synthesised
    • Yes, but it has been a while
    • Not yet. I have been meaning to
  • 03

    The team is shipping but it does not compound

    When you look back at the last quarter or so of work, can you say in one sentence what you actually moved? Not what you shipped — what you moved.

    • Some things, but not enough
    • Honestly, not really
    • We moved a lot — just not the things I expected
  • 04

    I am considering a redesign

    Before we keep going — is something actually broken, or does something just feel broken? The honest answer matters more here than the smart one.

    • Something is broken — I can point to it
    • Something feels broken, but I am not sure what
    • Both, in different places

    If something is broken — a redesign might be the right move. But "redesign" tends to scope-creep on the way. What is the smallest version of the change that would still address the actual broken thing?

    • A targeted change would do it
    • It would need to be broader to land
    • I am not sure what the smallest version is

    If it just feels broken — most redesigns are reactions to a feeling. That is not wrong — but the work that comes after is different from the work that comes after a real fault. Sit with this one: what specifically would have to be different about your product for the feeling to go away?

    • I can name a few things
    • Nothing concrete — just a general sense

    If both — that is actually the most common case. The trick is keeping them separate. The broken bit gets fixed; the feeling-broken bit gets diagnosed before any pixel moves. Take the broken-and-fixable part first: what is the one piece you would unblock if you could only do one?

    • I have one in mind
    • I cannot pick just one
I just want to think out loud

Good. Then I will stop asking.

Pick a thread to dip into — or close the tab. Both are fine.

  • 01

    Where would you start, if you had to make one change?

    If I had to pick a single starting move, it would be the cheapest one that produces honest disagreement.

    Show three people on your team the same thing — a sketch, a single screen, a sentence — and ask them what it is for. The disagreement is the data.

    Most teams skip this because the answer feels obvious until they hear it out loud.

    That is the move. The rest of the page is in the footer if you want it.

  • 02

    What do you usually see in products at this stage?

    Three patterns repeat at this stage.

    One. The product is heavier than what users actually use, but no one knows which features to remove without breaking trust.

    Two. The team agrees on the words but disagrees in the work, and the disagreement leaks into the design.

    Three. The company stops noticing things it has gotten used to — bad onboarding, bad copy, bad first week — because the team experiences none of it.

    None are unusual. All of them get worse if you ignore them.

    That is most of what I would say.

  • 03

    How do you know when a product needs more than a tweak?

    You probably need more than a tweak when the same problem keeps coming back wearing different clothes.

    A new feature does not fix it. A team reshuffle does not fix it. A redesign does not fix it — at least not for long. That is the signal that the work has to go earlier, to the question of what the product is even for, before any pixels move.

    Worth sitting with.

I have not asked you to do anything.

That is on purpose. Most of what we just talked through, you can figure out on your own with a notebook and an honest afternoon. If at some point you think a second pair of eyes would help — calmly, without pressure, looking only at what you actually need — the way to find me is in the footer of this page.