ChatGPT prompt to prep a quarterly business review (QBR slide outline)

This is the prompt we use one to two weeks before a quarterly business review with a key client. It turns your raw inputs into a slide outline and open questions, with one core rule: the AI uses only the numbers you give. Where a result has no metric, it writes [NEED METRIC] rather than inventing growth.

The prompt to copy

Prepare a quarterly business review for {client}. Inputs — work delivered: {paste}; results/metrics: {paste}; issues: {paste}; next-quarter ideas: {paste}.
Produce:
1. A 10-slide outline (slide title + 3 bullets each): quarter in review → results vs goals → what didn't work and what we changed → next quarter plan → the ask (renewal/expansion).
2. Three open questions to get *them* talking about their priorities.
Use only the numbers I gave; where a result lacks a metric, write [NEED METRIC] instead of a vague claim.

Replace the fields in curly braces: the client name, then four blocks pasted as-is, the work delivered, the numbers, the issues you hit, and your ideas for next quarter.

When to use it

One to two weeks before the review, when you have the data but not yet the story. The prompt does the assembly: it orders the quarter, sets results against goals, and stages the renewal ask so you are not starting from a blank page. The three open questions change the tone of the meeting: instead of walking through your slides, you get the client talking about their priorities.

What you get back

1. A 10-slide outline, a title and three bullets each, ready to pour into your template.
2. A clear arc: quarter in review, results vs goals, what missed and what you fixed, next-quarter plan, the ask.
3. Three open questions to open the discussion on their priorities.
4. Results grounded only in your numbers, with the gaps named, not filled.

Real example: raw inputs → QBR outline

Input: at Brightside Studio, an 8-person brand agency, Maya, the founder, is prepping the quarterly review for Fernhill Coffee. Priya handed her the social metrics, but the brand-awareness result was never measured. The inputs pasted in:

Work delivered: packaging redesign (direction B), 36 Instagram + LinkedIn posts, 1 launch campaign.
Results/metrics: Instagram engagement +41% over the quarter; 2,300 new followers; brand awareness: positive sense in-store, no measurement.
Issues: email campaign pushed back, waiting on client sign-off.
Next-quarter ideas: monthly newsletter, Google Business Profile refresh.

Output, an excerpt from the “results vs goals” slides:

Slide 3: Results vs goals

• Instagram engagement: +41% over the quarter, target of +25% beaten.
• Audience: +2,300 followers across Instagram and LinkedIn.
• Brand awareness: [NEED METRIC]. Positive in-store feedback, not measured this quarter. Instrument for next quarter.

Notice what the AI did not do: it did not turn “positive sense in-store” into “brand awareness up sharply”. With no number, it dropped the [NEED METRIC] marker and proposed measuring it. That is the prompt working, not failing: Maya replaces the marker with a real number, or owns it as unmeasured, but she does not present an inflated result to a client who would catch it.

This prompt ships in the kit’s library, with its usage card.