STAR, CAR and PAR: Stories, Case Studies and CV Bullets

Study guide

Tell clear stories with STAR, CAR and PAR

Most people have done good work and cannot describe it. They say they helped out, the team delivered, things improved. The listener learns nothing, and the work goes uncounted. STAR, CAR and PAR are three shapes that fix this. STAR stands for Situation, Task, Action, Result. CAR drops the Task. PAR replaces Situation and Task with a single Problem. All three move from context to action to outcome, and all three work as instructions for an AI tool.

This course teaches you to use those shapes with ChatGPT, Claude or Gemini. You will write a Situation and Task that keep the story specific. You will write an Action section in the first person with strong verbs. You will write a Result with real measures, and you will stop the AI inventing numbers by using placeholders like [number]. Then you will shrink a long STAR story into three PAR bullets for a CV, and build a story bank of three reusable stories from your last year. It is written for busy professionals in any role, from support to operations to healthcare, who need to explain their work in interviews, case studies and reports.

What is a prompt framework?

A prompt is the request you type into an AI tool such as ChatGPT, Claude or Gemini. A prompt framework is a reusable structure for that request: a short checklist of the parts a good prompt contains, usually named with an acronym so it is easy to remember. Why it helps: the AI fills in whatever you leave out with generic guesses. A framework makes sure you include the parts that change the answer, such as who it is for, what you want to achieve and the format you need.

Frameworks make prompts repeatable: you fill the same parts each time, save the prompt as a template and share it with colleagues. They are not magic words: the value comes from the detail you put in each part. They help most when a request is complex or open to interpretation; a quick, simple question can stay short. There are well over sixty named frameworks, and most of them remix the same ingredients: role, task, context, audience, format, constraints and examples.

In this course you learn STAR, which is a fill-in-the-slots framework: an acronym where each letter stands for one part of your prompt. It is best for: case studies, incident write-ups, interview answers and CV bullets.

The three shapes and when each one fits

STAR comes from behavioural interviewing, where it structures answers to questions like tell me about a time when. Situation sets the scene, Task names your responsibility, Action describes what you did, and Result gives the outcome. The same four steps work as a prompt structure. When you ask an AI for a case study, an incident report or a list of achievements, STAR gives it a clear path from context to outcome instead of a pile of generic business sentences.

CAR stands for Context, Action, Result. It drops the Task because in a short format the context already tells the reader what needed doing, so a separate task sentence wastes words. PAR stands for Problem, Action, Result. It starts from the problem, which suits incident reports and support case studies where the reader already knows the background. The choice between them is about the reader. If they know the background, lead with the problem. If they do not, give them the context first.

Two habits carry across all three shapes. A strong story names a specific Action, because vague phrases like I helped out give the listener nothing to hold on to. And the Result is the part you never cut. Skipping it is the most common mistake, because the Result is where you show why the story mattered and what changed.

Situation and Task: the scene and your job

The Situation is the scene setter. It answers where, when and what was going on, in one or two sentences. Anything longer buries the point and loses the reader. The Task is your responsibility or the goal, and it must belong to you, not the team. If you cannot say I had to, it is not a task yet. A useful test is whether a stranger could picture your scene after reading two sentences. If not, add one concrete detail and cut one vague one.

AI tools fill gaps with generic language. If you leave out the month, the client or the numbers, the output will sound like every other business story. So put the raw facts in the prompt first. Write the month, the team size, the system, the customer type. The AI cannot invent a scene that matches yours unless you hand it the pieces.

Situation and Task are the same two lines in every format. A CV bullet, an interview answer and a customer case study all start from the same setup. Keep that setup short so the Action and Result get the space. The first two letters exist to make the last two letters land.

Action: what you personally did

Action is the part of STAR where you describe what you did, step by step, in the first person. It answers the question the interviewer really cares about, which is what you personally did. Always tell the AI to keep the actions specific and your own, not the team's. Sentences like we handled it or the team delivered hide you, and the reader cannot credit work they cannot see.

Strong verbs and concrete steps make the difference between a story and a claim. I am good under pressure is a claim. I assessed the patient, called the on-call doctor, and started the checklist is a story. Give the AI your raw facts and a clear instruction: first person, my steps only, strong verbs. Then read the draft and replace any weak verb such as did, helped or worked on with a verb that names a real move.

Keep the Result out of the Action section while you draft. Mixing them blurs the story. Write the steps first, then add the outcome in the next part of the framework. This also makes editing easier, because you can check the steps for vagueness and the result for evidence as two separate jobs.

Result: the part that proves your value

Result is the outcome of your actions, and it is the part of STAR that proves your value. Without it, the story is just a description of activity. Measure wherever you can. The four common measures are time saved, money, customer satisfaction and errors avoided. A number turns a claim into evidence, and it gives the reader something to repeat when they describe your work to someone else.

When a result is mixed, say so and add what you learned. Interviewers trust a candidate who names a shortfall and shows the lesson more than one who reports only perfection. A sentence like we cut the backlog by half but missed the launch date, and I now build in a two week buffer, is stronger than a perfect story nobody believes.

AI tools will invent numbers to fill gaps unless you stop them. Always include the instruction never to invent numbers or dates. Use placeholders like [number] for any figure you do not have. This keeps the draft honest and shows you exactly what to look up before you send it. Put your raw facts first in the prompt, then the no-invention rule, then the format. That order anchors the AI on what is true.

CAR and PAR for short formats

CAR drops the Task because in a short format the context already tells the reader what needed doing. That saves a line, which matters on a CV where every word competes for attention. PAR goes further and replaces Situation and Task with a single Problem step. It starts from the problem, which suits incident reports and support case studies where the reader already knows the background and wants to reach the fix quickly.

The result is the part you never cut. If space is tight, shrink the context or problem to a few words and keep the action and result intact. A three line PAR bullet that ends in a number beats a paragraph of background that ends nowhere.

You can ask the AI to turn a long STAR story into three PAR bullets. State the format, the number of bullets and the audience, and tell it to keep your facts and add no numbers. For example: turn the story below into three PAR bullets for a hiring manager, keep my facts, use no numbers I have not given you. Then check each bullet against your notes before you use it.

STAR in your job: story bank, case studies and checks

Build a story bank of three STAR stories from your last year. Pick moments with a clear problem, your action and a result you can prove. Keep each story to half a page and store them in one document. Your three stories should cover different ground: one about a hard problem you solved, one about a project you led, and one about a time you helped a customer or teammate. That gives you options for almost any question.

Use STAR for customer case studies. Situation and Task set the scene, Action shows what the customer did, and Result gives the outcome. Always ask for a quote placeholder so the customer adds their own words. A marker like [INSERT CUSTOMER QUOTE HERE] tells the reader that the customer will fill in their own words. Never let the AI write a quote and present it as real.

Check every claim before you use an AI draft. The AI should only polish your facts, never add new ones. If you cannot point to the number in your own notes, cut it. The AI is a polisher, not a source. You bring the facts, the numbers and the outcome. The AI makes them read well. If the draft adds an achievement you did not do, remove it.

Frequently asked questions