Every structure below is set up for your lesson in two minutes. The board tells the class what to do, runs the timer and makes the groups.
The design cycle, or the engineering design process, hangs on a poster in almost every technology and science room: problem, requirements, ideas, proposal, build, test, present. What the poster does not say is how a class of thirty actually takes those steps. Without a structure the handiest student builds, the rest watch, and the first idea becomes the design.
Below is the cycle phase by phase, with the structures that make each step happen for the whole class. All 19 run here right away.
In design-based learning, students learn by designing: a solution to a real user’s problem, which they think up, build and test. The plan for that is the design cycle. In the United States it is usually taught as the engineering design process; in Dutch schools it is mostly the cycle of Marja van Graft and Pierre Kemmers (2007), with the list of requirements up front and the test against it at the end. The steps are the same in both.
The difference from inquiry-based learning lies in the question. Inquiry starts from a question about how something works and ends in an explanation. Design starts from a problem and ends in a thing that solves it, and that has to meet requirements. In practice they overlap: testing a design is a small experiment.
The cycle belongs to technology and science, but not only there. Wherever a class makes something for someone else (a leaflet for a real reader, a plan for the schoolyard) the same steps can be taken.
One sentence per phase on what the class does, and the structures that make it happen for everyone at once. The table below puts them side by side.
The names are kept short here; order and content follow the seven steps of Van Graft and Kemmers. A phase may come back: a test that fails sends the team back to the proposal, and that is the cycle working, not a detour.
Sorted by phase. The time is a normal run, as the structure gives it on its own page. Most fit in a quarter of a lesson, which leaves a build lesson time to build.
keep asking until you reach the cause of the problem
1 · the problem looks clear, but nobody knows why it exists · ± 11 min · pairs
interview each other, then report to your four
1 · the class practises the questions for the user · ± 12 min · fours
short conversations, each with someone new
1 · gather many experiences of the problem quickly · ± 13 min · whole class, 10–36
each writes alone, the middle holds what the team shares
2 · the requirements the whole team stands behind · ± 13 min · 4 per sheet
each writes requirements, the team merges one list
2 · a long list first, then choose · ± 16 min · 2–6, default 4
one idea per slip, said aloud, then ranked
3 · many ideas, and everyone contributes · ± 19 min · 2–6, default 4
work out how the design would surely fail, then flip it
3 · the ideas are tame, or nobody sees the risks · ± 19 min · alone or groups up to 8
alone, in pairs, in fours: ideas build on each other
3 · everyone has an idea before the team chooses · ± 18 min · 8–32
teams lay out their sketch, the rest give feedback
4 · look at the sketches before anything is built · ± 30 min · 2–8 teams
A works it out aloud, B coaches, then switch
4 · work out the proposal: sizes, materials, cost · ± 16 min · pairs
one minute each: done, next, where I am stuck
5 · at the start of every build lesson · ± 7 min · teams of 2–8
work in blocks, with real breaks between them
5 · a long build lesson that would otherwise drift · ± 40 min · whole class at once
predict first, then watch in silence, then explain
6 · the test becomes an experiment with a prediction · ± 17 min · alone, then pairs
find the one error in a worked solution and explain why
6 · check the calculation behind the design: where did it go wrong? · ± 14 min · alone, then pairs
one member visits, three explain their design
7 · every team presents, without a row of presentations · ± 20 min · teams of 4
conversation at tables, moving on after each round
7 · what the class takes into the next project · ± 45 min · tables of 3–6
everyone gives one answer in turn
7 · the last round: one thing you would do differently · ± 12 min · 2–6, default 4
| Structure | What it does | Group size | Time | Phase and moment |
|---|---|---|---|---|
| Five Whys | keep asking until you reach the cause of the problem | pairs | ± 11 min | 1 · the problem looks clear, but nobody knows why it exists |
| Interview in Three Rounds | interview each other, then report to your four | fours | ± 12 min | 1 · the class practises the questions for the user |
| Inner & Outer Circle | short conversations, each with someone new | whole class, 10–36 | ± 13 min | 1 · gather many experiences of the problem quickly |
| Placemat | each writes alone, the middle holds what the team shares | 4 per sheet | ± 13 min | 2 · the requirements the whole team stands behind |
| Think, Write & Share | each writes requirements, the team merges one list | 2–6, default 4 | ± 16 min | 2 · a long list first, then choose |
| Written Thoughts | one idea per slip, said aloud, then ranked | 2–6, default 4 | ± 19 min | 3 · many ideas, and everyone contributes |
| The worst plan | work out how the design would surely fail, then flip it | alone or groups up to 8 | ± 19 min | 3 · the ideas are tame, or nobody sees the risks |
| Snowball (1-2-4-All) | alone, in pairs, in fours: ideas build on each other | 8–32 | ± 18 min | 3 · everyone has an idea before the team chooses |
| Gallery Walk | teams lay out their sketch, the rest give feedback | 2–8 teams | ± 30 min | 4 · look at the sketches before anything is built |
| Pair Coach | A works it out aloud, B coaches, then switch | pairs | ± 16 min | 4 · work out the proposal: sizes, materials, cost |
| Daily Scrum | one minute each: done, next, where I am stuck | teams of 2–8 | ± 7 min | 5 · at the start of every build lesson |
| Pomodoro: get to work | work in blocks, with real breaks between them | whole class at once | ± 40 min | 5 · a long build lesson that would otherwise drift |
| Predict – Observe – Explain | predict first, then watch in silence, then explain | alone, then pairs | ± 17 min | 6 · the test becomes an experiment with a prediction |
| Error hunt | find the one error in a worked solution and explain why | alone, then pairs | ± 14 min | 6 · check the calculation behind the design: where did it go wrong? |
| 1 goes, 3 stay | one member visits, three explain their design | teams of 4 | ± 20 min | 7 · every team presents, without a row of presentations |
| World Café | conversation at tables, moving on after each round | tables of 3–6 | ± 45 min | 7 · what the class takes into the next project |
| Taking Turns | everyone gives one answer in turn | 2–6, default 4 | ± 12 min | 7 · the last round: one thing you would do differently |
The way into the cycle. In one lesson, teams walk every phase on a small task: they ask a client from the next team what it must do, write requirements they can test, build from card and paper, let the client test it, and then improve exactly one thing. It does not replace a design project. It makes sure the class has taken every step once before a real project spreads them over weeks, and that afterwards the class can point at the poster and say where it is.
Built for the demonstration in science class, and in the cycle the structure for phase 6. Everyone first writes down what will happen and why, pairs compare, and the different predictions go up on the board. Then the class watches in silence, and a drawn pair explains the outcome. In a design project that turns the test into more than whether it works: the team says beforehand which requirement the design will meet, and a requirement it misses becomes something to explain instead of a disappointment.
Also for science and math class, but outside the cycle. Students sort mixed problems by the approach each one needs, without solving them. Pairs turn that into recognition rules, and the class puts one rule per type on the board. It is the structure for the moment a class knows the methods but, facing a new problem, does not know which one to reach for.
A worked solution with numbered lines is on the board, and exactly one line is wrong. Each student writes down alone the number of that line and why, pairs compare, the class shows its numbers at once, and then the screen marks the error. Whoever got it right stands up and explains it to someone who missed it. In the cycle it fits phase 6, when a team checks the calculation behind a design; outside it, any rule the class has just learned.
Start with Design thinking in miniature, in one lesson and on a task with no subject content. The class learns the order of the steps, and the look-back at the end (which step was hardest, which did you nearly skip?) becomes the agreement for the real project.
If the project then runs over weeks, take one phase per lesson and one structure per phase from the table. Open every build lesson with Daily Scrum, circulate the sketches halfway with a Gallery Walk, and close with 1 goes, 3 stay, so every team explains its design without the class watching presentations for an hour.
For the build itself: the role cards in the basic kit include a set of STEM roles (designer, builder, tester, presenter). With a role for each student, the handiest one is not the only one building.
The class starts by building.
The first idea becomes the design and the design is taped together at once. Beginning designers skip the requirements most often; that is why phases 2 and 3 come before phase 5, and why Snowball has everyone think alone first.
The requirements cannot be tested.
“Nice” and “sturdy” are not requirements; “it stays upright with the phone in it” is. A requirement you cannot test cannot be met or missed at the end either, and then phase 6 has nothing to measure.
One student builds, the rest watch.
A team without a division of tasks gives the work to whoever is fastest. Roles per student, and structures in which everyone must contribute, such as Placemat and Daily Scrum, keep the work with the whole team.
The test is: does it work?
Whoever tests their own design tests it gently, and a yes or no teaches nothing. Have the team predict beforehand which requirement it will meet, and have someone else test.
Everything is rebuilt.
After a failed test the team changes everything at once, and then nobody knows which change made the difference. One improvement at a time, and test that one again.
The presentation ends the project.
Without a look-back, what is left of the project is a thing, not a way of working. A last round of Taking Turns (“one thing I will do differently next time”) takes a few minutes and is what the class takes away.
Inquiry starts from a question about how something works and ends in an explanation. Design starts from a user’s problem and ends in a thing that solves it, and that has to meet requirements. The two overlap: testing a design is a small experiment, which is why Predict – Observe – Explain, built for the demonstration, also serves phase 6 of the design cycle.
Design thinking is a way of designing that starts from the user; the Stanford d.school describes five modes: empathize, define, ideate, prototype and test. The engineering design process taught in schools has the same order and puts the requirements explicitly up front. For the class it matters little which poster hangs on the wall: the structures on this page fit both.
As many as you have. Design thinking in miniature walks the whole cycle in one lesson, on a small task. A real project spreads the same phases over weeks, one phase per lesson. Most structures in the table fit in a quarter of a lesson, which leaves time to build.
Yes. The cycle works wherever a class makes something for someone else: a leaflet in English class for a real reader, a plan for the schoolyard in social studies, a game for younger students. The structures in the table are not tied to a subject anyway; they are here because each makes a phase of the cycle happen for the whole class.
A set of four role cards for a build team: designer, builder, tester and presenter, each with three duties on the back. You print them once from the basic kit, and drawing a card means getting a role. In Design thinking in miniature you can switch them on as a lever.
You can prepare every lesson in the lesson builder, with that phase’s structure in it. A template that lays out a whole project over several lessons in one go does not exist yet.
Compiled by Oksana Oliinyk ·