Strategy Prompts vs Summary Prompts
Reviewed by Human · Updated September 12, 2026
Ask AI to "summarize the quarterly report" and you get a competent intern. Ask it the right way and you get a strategist who names the threat your board missed. Same model, same data — the entire difference lives in what you demand of it.
✗ Wrong way
Summarize this competitor's earnings call transcript for me.
You asked the model to compress, so it compressed. You get a neat recap of what was said — revenue up, new region launched, margin guidance held — and zero read on what any of it means for you. A recap is not analysis. The model never had to form a judgment, weigh a tradeoff, or point at a decision, because you never asked it to.
✓ Better way
You're a competitive analyst on my strategy team. Read this earnings call transcript and tell me: (1) the two moves that most threaten our pricing position, (2) what each signals about their roadmap for the next two quarters, and (3) the single response you'd recommend we prioritize — with the tradeoff we'd be accepting. Be specific and rank by urgency.
It assigns a role, forces the model to reason step by step from evidence to a ranked recommendation, and demands a tradeoff — so the output is a decision input, not a book report.
There's a single question you can bolt onto any summary prompt that converts it into a strategy prompt.
What Makes a Strategy Prompt Different From a Summary Prompt?
A summary prompt points backward at information. A strategy prompt points forward at a decision. **Summary prompt**: an instruction that asks the model to compress existing information into a shorter form. **Strategy prompt**: an instruction that asks the model to analyze information, weigh options, and recommend a course of action. The upgrade has three moving parts. First, a **frame** — the lens the analysis runs through (a competitor, a market, a risk). Second, a **decision** the output must serve, so the model has a target to reason toward. Third, a demand for **judgment** — a ranking, a recommendation, a named tradeoff — so it can't hide behind a balanced list. A good strategy prompt hands the model the context it needs and a decision to make, then refuses to accept a shrug.
Your turn
Take a summary prompt and rebuild it as a strategy prompt. Add a role, a decision it must serve, and a demand for a ranked recommendation with a tradeoff.
Reflect
Read your version back. Is there a decision in it? If not, the model has nothing to reason toward.
The vaguer your goal, the more the model summarizes by default.
With no decision to serve, compression is the safest thing a model can do — it's the lowest-risk way to look useful. Ambiguity doesn't make the model neutral; it makes it retreat to recap. Specificity is what unlocks the strategist.
Which prompt is most likely to produce genuine strategic analysis rather than a recap?
Go deeper
You've finished this module.
Mark it complete to earn your XP and keep your streak alive.
Progress saved locally · Sign up to earn XP