What makes a good case study structure?

The proven arc is challenge, solution, and results, bookended by a one-line summary and a forward-looking close. Quantified outcomes and a direct customer quote are what make readers trust the story.

Should I include real metrics?

Yes — specific numbers like "cut onboarding time 40%" outperform vague claims like "saved time" every time. If you lack exact figures, use defensible ranges and label them as estimates so the copy stays honest.

Can I use this for B2B and B2C case studies?

Yes. The prompt adapts tone and emphasis based on the audience you set, leaning analytical for B2B buyers and more emotional and outcome-led for B2C readers.

Does my customer data leave my browser?

No. The builder assembles the prompt locally and outputs only text. Nothing is uploaded or stored — you paste the finished prompt into your own LLM.

Case Study Prompt Builder

Case study prompt builder

A great case study is not a list of features — it is a story with a hero (your customer), a conflict (the problem they faced), and a resolution (the measurable result you helped them reach). This builder takes the four facts you already have — customer, challenge, solution, results — and turns them into a precise LLM prompt that produces publication-ready copy following that proven narrative arc.

How it works

You supply the customer name and industry, the challenge they faced, what you delivered, and the metrics that moved. The builder assembles a prompt that instructs the model to open with a one-line summary, structure the body around challenge → solution → results, fold your real numbers into the headline and the results section, and end with a forward-looking line. You can set the audience (B2B or B2C), the tone, and the target length, all of which the prompt encodes so the model does not drift into generic marketing fluff.

Tips and examples

Lead with the number. A headline like “How Acme cut support tickets 38%” earns the click; “Acme’s success story” does not. The prompt prioritises your strongest metric.
Give the model a real quote. If you have a customer line, paste it — the prompt will place it as a pull quote rather than inventing one.
Keep the challenge specific. “Manual invoicing took 6 hours a week” beats “inefficient processes.” Specific pain makes the solution land.
Don’t let the AI invent metrics. The prompt explicitly forbids fabricating numbers; if you leave results blank it will mark them as placeholders for you to fill in.