Part One: Strategy · Deciding what to build
A vague idea cannot be built, and a detailed design gets defended instead of questioned. The stage in between takes hard thought before anyone has agreed to build, so it gets skipped.
There are two common ways to hand an idea over, and both go wrong.
The first is a sentence: "Let supporters change their own monthly gift." Everyone nods, and each person imagines something different. The real decisions get made later, one at a time, by whoever is building it, with the deadline close.
The second is a finished mockup. Every pixel has been decided, so the conversation is about pixels. Nobody asks whether the screen should exist at all, because someone plainly spent a week on it.
The stage in between is rough and specific at once, and it comes in two forms. One is words: each place, with what a person can do there listed beneath it, and a line to where each action leads. For the charity it is three places. Reminder email: change my gift. Gift page: new amount, confirm. Thank-you page: nothing to do. Ryan Singer calls this a breadboard, after the boards engineers use to wire up a circuit before anyone designs the case. The other is a drawing, for ideas that depend on layout, made with a marker too thick to show detail.
Either one says what is left out: "Not changing the payment date. Not cancelling." It also names the trap that could swallow the time. What happens when the card on file has expired? Settle that before building, while the answer costs nothing.
The rest is left open on purpose. The people who build it will see things no sketch could, and they need room to decide them.
Therefore:
Before anyone draws a screen, work the idea out as places, actions and what leads where. Keep it rough enough to change, and clear about what is in, what is out, and which unknown could sink it.
In practice. Write your next idea on one card: the places, and under each the things a person can do there. Draw only if the question is layout. If it will not fit on the card, it is two ideas, or you do not understand it yet.
Goes with. The sketch has to fit what you set when you DECIDE WHAT IT'S WORTH (21). When screens get drawn, START WITH THE FEATURE (62). The questions a sketch raises have patterns of their own, like THE EMPTY SCREEN TEACHES (71).
Source: Ryan Singer, Shape Up (2019).