Course/How to Ask

2.4

When the First Answer Misses

Rewrites, summaries, explanations, plans, and more.

What you'll leave with

By the end of this lesson, you'll know how to set up prompts for each of the most common AI task types so the output is reliably useful.

Why this matters

The formula from Lesson 2.3 works for everything. But certain task types have small setup differences that make a big difference in the output. This lesson covers the most common ones.

The idea

Different tasks benefit from different setups. Knowing the right framing for each type means you spend less time getting to a useful result.

The teaching block

Rewriting — Use when: you have something drafted and want it improved, shortened, clarified, or adjusted in tone. Setup: paste the original text into your prompt. Then tell AI what to change and why.

Here is a paragraph I wrote. Please rewrite it to be shorter and more direct. Keep my main point but cut anything that is repetitive. [paste text]

Summarising — Use when: you have a long document, email thread, article, or set of notes and need the key points. Setup: paste the content in full. Tell AI how short the summary should be and what to prioritise.

Summarise the following email thread in three sentences. Focus on the decision that was made and the next action required. [paste thread]

Explaining — Use when: you encounter something confusing — a term, a concept, a document, a process. Setup: paste or describe what you need explained. Tell AI your level and what kind of explanation you want.

Explain what a deductible is in health insurance as if I am intelligent but have never dealt with insurance before. Use one concrete example.

Brainstorming — Use when: you need options, ideas, names, approaches, or starting points. Setup: describe what you are brainstorming for and any constraints. Ask for a specific number of options.

Give me ten ideas for a team offsite that is fun but not forced. We are a team of twelve, mostly introverts. Budget is modest. Avoid anything athletic or competitive.

Planning — Use when: you need to organise something — an event, a project, a week, a trip. Setup: describe the situation, constraints, and what a useful output looks like.

Help me plan a week of dinners for four people. Two are vegetarian. We have about 30 minutes to cook on weeknights and more time on weekends. Give me a simple list with one ingredient note per meal.

Comparing — Use when: you are deciding between two or more options and want help thinking through the tradeoffs. Setup: describe each option and what matters most to you in the decision.

Help me compare these two laptops for someone who mostly writes, occasionally edits photos, and travels frequently. I care most about battery life and weight. [paste specs]

Example

Pick one task type — planning — and show the setup in action.

Help me plan a week of dinners for four people. Two are vegetarian. We have about 30 minutes to cook on weeknights and more time on weekends. Give me a simple list with one ingredient note per meal.

Notice: the constraints are front-loaded. The output format is specified. The result is a useful, personalised plan rather than a generic list of dinner ideas.

Try this now

Pick one task type from this lesson — whichever is most relevant to your life right now — and try it. Use the setup guidance to write your prompt, then run it.

Notice: does the framing from this lesson change the output compared to how you would have asked before?

Save this

For each task type, the setup difference is small. Paste the content, specify the goal, say how short or long you need it. That is usually enough to go from generic to genuinely useful.

Quiet takeaway

You now have a toolkit of task-specific prompt setups. Use them as starting points, then adjust based on what you actually need.

Next

You now have a toolkit of task-specific prompt setups. The final lesson in this module covers the most important skill of all: what to do when the first answer is not what you needed.

Sign in to track progress