# Dashboard Wireframe Prompt

A prompt for getting a low-fidelity dashboard layout out of an LLM, plus the checklist for
reading what comes back before you trust it. Copy it, fill in the brackets, paste it into
whichever model you use.

From [datawirefra.me/blog/dashboard-wireframe-prompt](https://datawirefra.me/blog/dashboard-wireframe-prompt)

---

## The prompt

```
You are producing a low-fidelity dashboard wireframe. Grey boxes only.

CONTEXT
- Audience: [who opens this, how often, on what screen]
- Decision: [what they will do differently after seeing it]
- Platform: [Power BI / Tableau / Looker Studio]
- Canvas: [1280x720, or the size this audience actually uses]

CONSTRAINTS
- At most [N] KPI cards and [N] charts. Do not exceed these numbers.
- Every KPI card carries a comparison: vs target, vs prior period, or vs
  prior year. If a metric has no meaningful comparison, leave it out.
- The number that answers the decision goes in the top-left position.
- No colour, no icons, no logos, no imagery, no styling flourishes.
- Label every box with what it contains.

OUTPUT
For each element give me:
  - position (row, column)
  - relative width
  - what it shows
  - one sentence on why it earns its place

Then, separately, list everything you considered and left out, and why.
```

The last instruction is the one that matters most. A model asked only for a layout hands
you a layout. A model asked what it discarded has to show its work, and the discard list is
usually where you find out it was guessing.

---

## Filling in the brackets

**Audience.** Not "the sales team". A role, a frequency, and a screen. "Regional sales
managers, Monday morning, on a laptop."

**Decision.** The hard one, and the one most prompts leave out entirely. What will this
person do differently after looking at the dashboard? If you cannot answer it, the model
cannot either, and it will quietly invent a purpose that sounds reasonable.

**Counts.** Give real numbers. 3-5 KPI cards for an executive view, 5-7 for management, 7-9
for operational. Without a number, you get whatever fills the canvas.

**Canvas.** A fixed size, not "a dashboard". Layouts that ignore the canvas produce
six KPI cards that are each too narrow to read at the width you're actually building for.

**Fidelity.** Say grey boxes explicitly, and say it more than once. Models drift toward
polish because polished output looks like better work.

---

## Reading the output

Run this before you show the layout to anyone. It takes about two minutes.

- [ ] **Count the elements.** Write the number down. It is almost always higher than the
      number you asked for.
- [ ] **Point at the decision.** Which single element answers it? If you have to think
      about it, the layout has not answered it.
- [ ] **Delete every element that isn't the answer or context for the answer.** Do this
      before you get attached. Whatever you cannot bring yourself to delete is worth
      keeping, and now you know why.
- [ ] **Check every KPI card for a comparison.** A bare number tells the reader where they
      are without telling them whether that's good.
- [ ] **Check the top-left.** That position is read first and read most. If something
      decorative is sitting there, the layout is wrong regardless of how it looks.
- [ ] **Read the discard list.** If the model dropped something you needed, your brief was
      incomplete, not the model.
- [ ] **Look at it again tomorrow.** The single most reliable test. A layout that still
      looks right the next morning usually is.

---

## What this does not solve

The prompt won't decide what the dashboard is for. That's the decision line, and it comes
from a conversation with whoever asked for the dashboard, not from a model.

If you want a structure for having that conversation, the
[dashboard requirements template](https://datawirefra.me/templates/dashboard-requirements-template.md)
covers the five things worth agreeing on before anyone builds anything.

---

**Next step:** lay the result out at real size and see whether it survives contact with a
fixed canvas. Free at [app.datawirefra.me](https://app.datawirefra.me).