The value of a session is decided after the session. A wall of cards is not a result, and a group that watches its wall get photographed and filed learns quickly that the exercise is decorative. Four steps, about one hour, one decision.
| Step | Ideas left | What happens |
|---|---|---|
| Collected in the session | 68 | Six tools, 90 minutes, six people writing |
| After merging duplicates | 41 | Nine groups. 27 cards said the same thing twice |
| Inside the search field | 29 | 12 broke a limit or belonged to another question |
| Above the score threshold | 12 | Four weighted criteria, 24 of 35 points needed |
| Selected for a trial | 5 | Chosen along the effort and effect line |
Group before you rate
Move the cards into groups in silence first, and name the groups afterward rather than before. Naming first creates the categories somebody expected and quietly forces every card into them.
Twenty-seven of the 68 cards turned out to be duplicates. That is not waste. A card that four people wrote independently is a card with support, and it is worth noting on the merged idea which of them wrote it.
Screen against the limits
Four questions, asked before anything is scored, because scoring an idea that breaks a stated limit wastes the group’s attention and then disappoints somebody.
- Is it inside the search field? Six ideas were about the paint process or the customer packaging. Another question, separate list.
- Does it respect the limits? Six ideas needed a new conveyor, a robot or more than EUR 15,000.
- Is it one idea or several? Three cards were split into two ideas each.
- Can we describe the trial in one sentence? If not, the idea goes back for one round of rewriting.
Ideas outside the limits are kept, never deleted. The six that only needed money became the agenda of the next investment discussion, and the six that belonged elsewhere became two new questions on the register.
Score with weights
Four criteria, the same for every idea, each scored from 1 to 5 where 5 is best, each multiplied by a weight. Maximum 35 points, threshold 24.
| Criterion | What 5 points means | Weight |
|---|---|---|
| Effect on the target figure | Many cases avoided per year | 3 |
| Effort | Little money, few hours, few people who have to agree | 2 |
| Speed | A trial can run next week | 1 |
| Acceptance | The people who do the work asked for it | 1 |
Twelve of the 29 ideas passed. The weights matter more than the scores: giving effect three times the weight of speed is a management decision, and it should be made once, in writing, before anybody sees the list.
Effort against effect
Plot the one-time cost on one axis and the effect per year on the other, then draw a line through each axis. With the lines at EUR 1,500 and EUR 5,000 the 29 ideas fell into four fields:
| Field | Ideas | What to do |
|---|---|---|
| Low cost, high effect | 4 | Start this week. All four were selected |
| High cost, high effect | 3 | One page, one budget round. One was selected |
| Low cost, low effect | 15 | Do them when somebody is already there |
| High cost, low effect | 7 | Say no in writing and give the reason |
The picture does the arguing. Nobody has to defend starting with a EUR 1,200 cutting jig instead of a EUR 48,000 rack system when both are on the same chart.
Decide, then answer everybody
The owner decides within one week. Five ideas went into a trial. The other 63 need an answer, and three sentences on the board are enough: these are in trial, these are on the list for the next question, these are outside the limits and here is why. That is the cheapest part of the method, the one most often skipped, and the reason the second session produces as many cards as the first.