Osborn’s checklist
Use nine prompts to change a product, service, or process, then select variations worth testing.
Use Osborn’s checklist when you have an initial idea but need a wider range of options. Work through other uses, adapt, modify, magnify, minify, substitute, rearrange, reverse, and combine. Keep the object of the exercise consistent.
An existing product, service, process, or proposed concept to improve.
Specific variations, useful combinations, and conditions to investigate.
On this page
How to use it
- Define the object and purpose of the improvement.
- Explore each of the nine prompts against the same object.
- Record concrete changes and the assumptions behind them.
- Combine useful changes and select a bounded trial.
- Check results against the purpose, rather than novelty alone.
Keep its limits in mind
Prompts help generate options; they do not establish feasibility or require every variation to be implemented. Explore the nine prompts separately; do not conflate them with the seven SCAMPER categories.
Rethinking a weekly meeting
Fictional example: eight members of Xinggang’s service team attend a 90-minute weekly meeting. Four weeks of meeting records are available. The purpose is to resolve cross-department blockers and make decisions.
On a small screen, scroll across the diagram to read it.
Proposed trial: Combine advance written updates, decision questions first, and an action record in a proposed 35-minute meeting. The shorter duration is a trial design, not a measured saving.
Review criteria: After two weeks, check decisions made and blockers resolved, alongside meeting duration. No follow-up results are provided.
Next step: Agree on the trial and review decision quality and blocker resolution after two weeks.
Three limits before generating meeting variations
First, pause if people are changing different objects. A weekly coordination meeting and the entire reporting system are different scopes. Name what can change and what purpose must remain. Otherwise an interesting variation can solve a different problem from the one the group needs to address.
Second, do not treat every prompt as an instruction. Magnifying an element can expose a useful possibility without requiring a larger meeting. A blank response is acceptable after considering the question. This worksheet retains nine separate prompts; do not silently merge them into another checklist or claim that nine answers constitute nine independent solutions.
Third, stop when novelty replaces usefulness. Reversing the order of speakers may change the experience without resolving a blocker. Keep the purpose, the reason for each variation, and what could fail. Check authority before removing a required meeting or asking decision-makers to attend. Ideas remain proposals until participants and owners agree.
Use a prompt list only while variations are the missing output
An existing arrangement provides the object for Osborn’s questions. If the missing work is discovering perspectives, understanding the process, or documenting one chosen concept, use the corresponding method instead. Moving between methods does not validate the idea.
| Input and need | Method | Output and boundary |
|---|---|---|
| One existing arrangement needs alternatives | Osborn’s checklist | Concrete variations and assumptions; no automatic implementation. |
| One topic is trapped in a narrow perspective | Mandala chart | Different directions; cells do not rank their usefulness. |
| Activities and handoffs need to be understood | Business process flowchart | Current paths and exceptions; the diagram does not prove delay causes. |
| One selected variation needs a small test | Idea concept sheet | A mechanism and validation plan; no established benefit. |
Use all nine prompts on the same weekly meeting
The complete fictional Xinggang packet states eight participants, a 90-minute weekly meeting, and four weeks of records. The purpose is resolving cross-department blockers and making decisions. Team members feel much of the meeting is spent reading progress aloud, but the packet supplies no timed breakdown or counts of decisions and blockers. The proposed combination is written updates beforehand, decision questions first, and an action record, with a 35-minute trial meeting and a review after two weeks.
The following table preserves all nine supplied variations. Some overlap in implementation, particularly changing the discussion, shortening routine updates, and rearranging the agenda. Do not count the same removed reporting segment as several separate savings. Inviting a decision-maker may help a major dependency but also adds another person’s time; this must be visible when assessing the whole arrangement.
| Prompt | Supplied variation | What to check |
|---|---|---|
| Other uses | Reuse progress notes for onboarding | Notes are understandable and appropriate for that audience. |
| Adapt | Borrow stand-up blocker reporting | The format suits the decisions this meeting needs. |
| Modify | Discuss decisions instead of reading updates | Decision questions and evidence are prepared. |
| Magnify | Invite decision-makers for major dependencies | Attendance is agreed and the decision is within their authority. |
| Minify | Move routine updates to advance reading | Preparation and reading effort remain visible. |
| Substitute | Use a shared board instead of round-robin reports | Participants can access and update it. |
| Rearrange | Discuss decisions before background | Necessary context is available before deciding. |
| Reverse | Staff raise requests for managers’ help first | People can raise blockers without having to invent an escalation. |
| Combine | Record decisions with owners and dates | Owners actually accept actions; dates are not invented. |
Calculate the trial design, then review the whole effort
The output is a trial proposal, not an observed saving. With eight people present for the entire meeting, 90 minutes corresponds to 720 person-minutes, or 12 person-hours. The proposed 35 minutes would correspond to 280 person-minutes, or four hours and forty minutes, under the same attendance condition. The conditional difference is 440 person-minutes. It excludes preparation, reading, follow-up, and any additional decision-makers. Four weeks of available records do not prove identical attendance at every meeting.
Before trialing, agree who prepares written updates, where participants find decision questions, what belongs outside the meeting, and who confirms decisions. The packet does not supply that agreement or a decision-maker’s approval. If advance reading shifts the same work into everyone’s evening, a shorter meeting alone is an incomplete result. Preserve the actual attendance and all relevant work rather than multiplying an assumed headcount.
After the proposed two weeks, compare actual meeting and related effort with defined outputs: decision questions considered, decisions made with reasons, unresolved questions, and blockers actually removed. Count only the blockers eligible for that review and distinguish an assigned action from a resolved dependency. More decisions are not automatically better decisions; retain reopened decisions and avoidable rework as context.
The next choice could be to retain the combination, revise preparation, or restore needed discussion time. A meeting that ends at 35 minutes while pushing every decision elsewhere has not met its purpose. Record what happened and the participants’ actual response before expanding the change. No trial results, realized savings, or resolved-blocker counts are provided in this case.
Check your work
- Each prompt has been considered, even if no useful idea emerged.
- Variations change something specific.
- Proposals and observed results are kept separate.
- The selected combination has a way to test its usefulness.
Framework references
References explain the framework and its scope; they do not validate the fictional example's results.
Put the framework to work
Copy the outline and add your question, evidence, and open questions.