Practice · module 8 of 100
Think of better ideas
Use variation and clear criteria to think further.
After this lesson
You distinguish between generating ideas and evaluating ideas.
In view · Choices and control
First broaden, then choose
Define for whom you are seeking ideas and which boundaries apply.
Request different perspectives instead of variants of the same idea.
Compare feasibility, utility and originality for your target group.
Part 1 of 3 · Én single issue
Your preference for less movement is on. Choose each part yourself.
Separate broad thinking from choosing
First ask for diverse directions: simple, surprising, cheap, or quickly executable. Only then choose using predetermined criteria. If you immediately ask for the best idea, you often get a safe middle ground without a proper comparison.
Provide real constraints
A target audience, available budget, channel, and timeframe make ideas useful. Indicate what you have already tried and why it did not work. Ask about assumptions that you need to test before executing the idea.
Turn an idea into a small trial
Choose one proposal and design a trial with an observable result. AI can help with a script or variants, but cannot prove that your customers want something. For that, you need real feedback.
New practice example
From explanation to a usable result.
Ask AI to suggest diverse possibilities and evaluate them using criteria that you choose in advance.
Fictional practice situationA fictitious community organisation is looking for an activity with a small budget, little preparation, and space for people with limited mobility.
View a possible solution
This is how it can be done
Request three different forms: a short indoor activity, a quiet outdoor activity, and something that can be done at home. Compare each proposal on budget, preparation, and accessibility.
Why this works
You aim for genuine variety and make constraints explicit. The proposals are a starting point; feasibility must still be verified.
Make it slightly more difficult
Change one constraint, for example, rainy weather. Which options remain standing without new assumptions?
Extra practice, at your own pace. You do not need to submit anything. This does not count towards an automatic assessment.
Your workplace · learning by doing
Make something. See what changes.
Checking the AI connection… You can already view the written example.
Present your assignment to Ende.
Read the answer, change one thing, and compare.
Previously held
New version
You remain in control
What do you want to improve in your next attempt?
On this device only. AI answers remain only in this page memory. This is self-review, not a grade.
Why this approach works
You aim for genuine variety and make constraints explicit. The proposals are a starting point; feasibility must still be verified.
A step further
Change one constraint, for example, rainy weather. Which options remain standing without new assumptions?
Try & discover · no obligation
Make your image brief visible
Change goal, format and composition. Compare the trial design with the assignment you are describing it with.
Your image brief in words
Create a vierkant prototype design for an online workshop. Place the illustration midden. Leave space for a short title and explanation. Use fictional elements. Present it as an illustration and check origin and usage rights before publication.
2 publication agreements are still open in your checklist.
This is a drawn draft design with fixed components, not an image generator. The tick boxes schedule a check and do not prove usage rights. Nothing is being published.
This trial works locally with fictitious data and does not count towards your lesson completion. It starts anew with a new lesson or reload.
A prompt to get you started
Think of eight diverse ideas to promote a fictional neighbourhood workshop. Budget: €50; preparation: two hours; target audience: local adults. For each idea, provide a small trial and one assumption. Then rank them by feasibility and reach, and explain the reasoning.
Apply it
Your team has five AI ideas for a newsletter. One idea sounds original, but requires image rights and a budget that has not yet been confirmed.
How do you determine which option proceeds?
- Automatically choose the most original idea.
- Use pre-agreed criteria such as target audience, feasibility, and rights; test a promising option on a small scale.
- Publish all five and only then look at the risks.
With JavaScript, you can check your approach and receive immediate feedback.
Test your insight
What makes a AI idea demonstrably useful for your target audience?
- The model gives the idea a high score.
- The idea contains many technical words.
- A small trial with real feedback and pre-selected criteria.
Extra practice
Choose one of your own ideas. Write a prompt that asks for six variants and specify which two criteria you will use to select a variant.
Think of six invitations for a free local swap meet. Three are aimed at meeting people and three at practical benefit. I choose based on clarity for someone unfamiliar with the initiative and feasibility within one afternoon.
