Concept iteration
Lab setup
First, make sure you have completed the initial setup.
If you are part of a course
-
Open Terminal. Run the update command to make sure you have the latest code.
$ mwc update -
Move to this lab's directory.
$ cd ~/Desktop/making_with_code/pilot/labs/lab_concepts
If you are working on your own
-
Move to your MWC directory.
$ cd ~/Desktop/making_with_code -
Get a copy of this lab's materials.
git clone https://git.makingwithcode.org/founders-studio/lab_concepts.git
You've found real needs. It's tempting to grab the first idea that comes to mind and start building, but the first idea is rarely the best one. In this lab, you will move from needs to concepts the way designers do: framing the problem carefully, opening up many possible directions, and choosing among them using evidence. You will use AI to generate more ideas than you could alone, and to critique them, while keeping your own judgment about what fits the people you talked to.
By the end of this lab, you will be able to:
- Write point-of-view statements that frame a design problem.
- Turn a point of view into many "How might we" questions.
- Generate a wide range of concepts, alone and with AI.
- Identify the critical assumption a concept depends on, and test it quickly.
- Present a concept in a short video.
Point of view
A point of view (POV) statement frames a design problem around a specific person, a need, and an insight:
[User] needs [need] because [surprising insight].
For example:
A student who works evenings needs to know about schedule changes early, because a last-minute change can cost them a shift, and they can't afford to lose one.
A good POV is specific about the user, describes a need (not a solution), and includes an insight you learned from needfinding, not something everyone already knows.
๐ Open pov.md. It contains some weak POVs. With your group, decide what
is wrong with each one. (Too broad? A solution in disguise? No insight?)
๐ป Using your insights from Needfinding,
write three POV statements in pov.md. Each should point to the insight and evidence it comes
from.
How might we
"How might we" (HMW) questions turn a POV into prompts for ideas. They're phrased to invite many answers: how suggests it's possible, might suggests there are many options, and we makes it a team effort.
From the POV above, you might ask:
- How might we let students know about schedule changes sooner?
- How might we make last-minute changes less costly?
- How might we let students trade shifts easily?
- How might we make schedule changes impossible? (wild)
- How might we turn a schedule change into something good? (wilder)
๐ป For each POV, write at least ten HMW questions in hmw.md, ranging from
safe to wild. Then star the two or three most promising.
Ideation
Now, generate as many concepts as you can. In a brainstorm, quantity matters more than quality: defer judgment, build on each other's ideas, and welcome wild ideas. A wild idea that doesn't work can lead to one that does.
๐ป Set a timer for fifteen minutes. For each starred HMW, brainstorm as a
team, writing each idea on its own sticky note (or line in concepts.md). Aim for at least
twenty ideas per HMW.
๐ป Now ask the agent to generate twenty more concepts for the same HMW. Give it your POV and insights as context.
๐ Compare the two lists. Which ideas are more surprising? Which are more generic? Did the agent come up with anything your team didn't? Did your team come up with anything that depended on details only you knew from interviews?
๐ป Ask the agent to critique your top ideas from different perspectives: a skeptical user, a competitor, a client paying for it, and someone who would be left out. Note any critiques that change your thinking.
Choosing and iterating
๐ป As a team, choose three concepts to develop further. For each, write a
short description and a sketch or storyboard in concepts.md.
Every concept depends on assumptions: things that must be true for it to work. The critical assumption is the one that, if it's wrong, sinks the whole concept. For example, a shift-trading app assumes students would be willing to trade shifts with people they don't know.
๐ป For each concept, write down its critical assumption, and a quick way to test it in a day or less, without building anything.
๐ป Show your sketches or storyboards to two or three people from your target
group. Record their reactions in feedback.md, then revise your concepts.
Concept video
A concept video shows a concept in action, before it exists. In a minute or two, it tells a short story: a person with a need, the frustration they face now, and how life is better with your concept. It doesn't need to be polished. Phone video, paper props, and hand-drawn screens are fine.
๐ป Plan your video as a storyboard in video.md, then film and edit it.
You can use AI to help with the script and storyboard, but the story should come from your
needfinding.
Deliverables
๐ป Answer these questions in questions.md:
- Which of your POV statements was the most useful, and why?
- Compare your team's ideas with the agent's. What was different about them?
- What was the critical assumption of your chosen concept? How did you test it, and what did you learn?
- How did your concept change after you showed it to users?