A step-by-step way to work
Start from real work
Give AI the actual project — the problem, what you did, the outcome. The substance has to be real.
Structure case studies
Have it shape your project into a clear case-study format (context → problem → your role → actions → result → reflection).
Sharpen, don't generate
Use it to edit your writing for clarity (see the Writing for Clarity guide), keeping your voice.
Repurpose across formats
Turn one piece of work into a post, a talk outline, and a portfolio entry.
Keep your voice
Edit AI output until it sounds like you, not like generic AI prose.
Real prompts, and what good output looks like
Turn this into a portfolio case study outline: 'I led a project to reduce support tickets. We added in-app guidance and an improved help centre. Tickets dropped about 30% in three months.'
Context: support volume was straining the team and slowing response times.
Problem: users couldn't self-serve answers to common questions.
My role: [name what you specifically led — discovery, prioritisation, working with design/eng].
Actions: in-app guidance at key friction points + restructured help centre, informed by [the top ticket themes you analysed].
Result: ~30% fewer tickets in 3 months; [add the impact — faster response, cost saved].
Reflection: [what you'd do differently — this is what makes a case study yours, not a brag].
I've bracketed what only you can fill — that's the part that makes it credible.
A prompt you can reuse
Fill in the highlighted parts and paste it into your AI tool of choice. Edit the output — it's a starting point, not a finished answer.
Help me turn real work into a portfolio case study. Bracket anything only I can supply — don't invent it. The project (real): [problem, what I did, outcome/numbers] Where this will live: [portfolio / LinkedIn post / talk] Please: 1. Structure it: context → problem → my specific role → actions → result → honest reflection. 2. Bracket every place that needs a specific I-did-this detail or a real number. 3. Keep it concrete and in a plain, human voice — no generic 'thought leader' tone. 4. Then suggest how to repurpose it into [the other formats].
What AI gets wrong here
Generic 'thought leader' voice
AI-written posts converge on the same bland, confident tone that signals 'AI-written'.
Invented or inflated credit
AI may phrase team work as if you did it all.
Substance-free output
Polished case studies with no real insight build nothing.
AI can produce and repurpose portfolio content quickly, but a reputation is built on real work and a recognisable voice — and an audience that reads a lot of AI text can feel the difference. Let it structure and polish; the projects, the specific contributions, the honest reflection, and the way you sound have to be yours.