SCAMPER
SCAMPER is an idea-generation technique that puts seven questions, one after another, to an existing object: a product, a process step, a form or a rule. Each letter names an operation on that object: substitute, combine, adapt, modify, put to other uses, eliminate, reverse. Bob Eberle formed the acronym in 1971 from the list of questions that Alex Osborn had published in 1953. The group or the analyst records each variant under the letter that produced it, then sorts the list against criteria set before the session. The deliverable is this sorted list, in which each variant carries its letter and its verdict.
Goal
SCAMPER produces variants of an existing object. Brainstorming asks an open question and leaves the group to find its own angles. SCAMPER supplies the angles: seven questions, put in the same order to every participant. The reserved participant and the fluent one work on the same question at the same moment, and the output depends less on how easily each person speaks.
The technique prepares a choice about how the object evolves: which step to remove, which tool to replace, which rule to invert. It does not make that choice. The sorted list feeds a decision matrix or a multi-criteria decision analysis, and the retained variant is often prototyped.
Eberle designed SCAMPER as a game to develop children's imagination, and the acronym had to be easy to remember in class. A group of adults learns the seven words in a minute, with no prior training in creativity.
Usage
When to use
- A product, process or form exists and has to evolve: the object supplies the material that each letter transforms.
- Moving from the current process to the target process: the seven letters produce candidates for each step of the process analysis.
- A mixed group or one new to creative workshops: everyone receives the same question, worded in advance.
- A brainstorm that runs dry: one letter applied to the most-cited idea restarts the flow of ideas.
- An analyst alone facing a deadline: the analyst fills in the grid in half an hour, without a workshop.
- A distributed team: seven columns on a shared board, where everyone writes in the current column.
- A line manager in the room: the letter E asks what can be removed, so nobody has to ask it in their own name.
When not to use
- No object exists yet: there is nothing to transform. Open up the options with brainstorming.
- Improving one characteristic degrades another: the variants shift the trade-off without removing it. Resolve the contradiction with TRIZ.
- The full solution space must be visible: SCAMPER modifies a single object. Build a morphological box.
Description
From Osborn to Eberle
In 1953, in Applied Imagination, Alex Osborn published questions that spark ideas from an object. His list has nine headings, which the Institute for Manufacturing at the University of Cambridge reproduces. In 1971 Bob Eberle reduced it to seven letters in SCAMPER: Games for Imagination Development. In 1972 he set out the technique in The Journal of Creative Behavior, where he traced it back to Osborn's questions. The grouping and the acronym are Eberle's. He took the letter E from the question Omit?, which Osborn files under the heading Minify.
| Osborn's heading | Eberle's letter (1971) |
|---|---|
| Substitute | S, substitute |
| Combine | C, combine |
| Adapt | A, adapt |
| Modify, Magnify, Minify | M, modify |
| Other uses | P, put to other uses |
| no heading; the question Omit? sits under Minify | E, eliminate |
| Rearrange, Reverse | R, reverse |
The seven questions
Before the session, the facilitator adapts the wording in the table to the object, so that a participant can answer without having to interpret it.
| Letter | Operation | Questions put to the object |
|---|---|---|
| S | Substitute | What can be replaced: a material, a tool, a step, an actor, a rule? With what? |
| C | Combine | What can be merged with another step, product or service? Which entries duplicate each other? |
| A | Adapt | Where is this problem already solved, in another department or another sector? What can be borrowed from there? |
| M | Modify | What happens if we enlarge, reduce, speed up or space out? If the frequency, the threshold or the form changes? |
| P | Put to other uses | Who else could use the object as it is? What can be drawn from what it already produces? |
| E | Eliminate | What can be removed entirely? What still holds without this element? |
| R | Reverse | What if the order of the steps were reversed? What if the roles swapped or the flow ran the other way? |
E and R produce the variants furthest from the object, because a group does not ask these questions on its own. An ordinary meeting proposes additions; E asks what can go. An established order carries on because it has always been that way; R asks for its reverse.
Running a session
Name a single object. The object is specific and exists: "the expense claim form", never "our HR processes". The facilitator writes it at the top of the board. A broad object produces variants that cannot be compared at the sorting stage. The facilitator splits a broad subject into parts and runs one round per part. When the object is not obvious, problem shaping identifies it.
Set the goal and the sorting criteria. The goal states what the variant must improve: a lead time, a cost, an error rate. Two criteria are usually enough: effect and effort. The group sets them before the round. A group that invents them at the sorting stage negotiates them under time pressure.
Ask the seven letters one at a time. The facilitator reads out the question for the current letter. Each participant writes variants in silence for two or three minutes, then reads them to the group. Writing at the same time keeps one participant from talking while the others wait. The facilitator keeps time and proposes no variants, so as not to steer the group. In practice, this pace works for four to eight participants. Above eight, the facilitator forms two groups on the same object and merges their lists at the sorting stage. Five to eight minutes per letter make a round of thirty-five minutes to an hour. The group goes through every letter in an order fixed in advance; the order of the acronym is the easiest to keep.
Record each variant under its letter. The participant words the variant as a change to the object, with a verb and what changes: "remove the line manager's sign-off below CHF 300" rather than "fewer controls". The facilitator places it on the board under its letter. The letter stays attached to the variant through to the deliverable, so that a reader who missed the session can find the question that produced it.
Sort, at a separate time. Sorting starts once the seventh letter is closed. The group merges duplicates, because the same measure often appears under two letters. It rates each variant against the criteria, then marks it as retained, held for study or dropped, with a reason. Dot voting is enough for a quick sort. A decision matrix fits when the criteria are weighted.
Give each retained variant a follow-up. Each one gets an owner and a next step: a hypothesis to test, a prototype or a cost-benefit analysis.
Existing object
a product, a process step, a form or a rule
one letter at a time
- 1SSubstituteWhat to replace?
- 2CCombineWhat to merge?
- 3AAdaptSolved elsewhere?
- 4MModifyBigger, smaller, more often?
- 5PPut to other usesWho else would use it?
- 6EEliminateWhat to remove?
- 7RReverseWhat if the order flipped?
List of variants, each tagged with its letter
goal and criteria set before the round
Sort
- merge duplicates
- rate against the criteria
- class: retained, to study, dropped
sorted list of variants
The traps
The vague object is the most common. "Improve reception" has no step to remove and no actor to replace, and each letter produces a wish instead of a variant. The test: can a participant point to the object, on a screen, on a floor plan or in a binder?
In the incomplete round, the group skips the letters that seem not to apply, almost always E and R. A letter with no result costs five minutes, and the group moves on without forcing an answer.
The filing dispute keeps the group busy deciding whether a variant belongs under S or under M. The letter prompts ideas and does not file them. The facilitator places the variant under the letter where it appeared, and sorting merges what overlaps.
Judging during the round stops the flow of ideas, as it does in a brainstorm. The facilitator writes "Internal control will never accept that" on a separate sheet. The remark comes back at the sorting stage, where it becomes a criterion or a reason for dropping a variant.
The forgotten sort costs the most. SCAMPER produces unrealistic variants alongside good ones. A committee reads an unsorted list as a series of proposals backed by the group, which has evaluated none of them.
A group that expects a breakthrough innovation will be disappointed: SCAMPER transforms the existing object, and its variants stay close to it.
Related techniques
| Technique | What it changes | When to prefer it to SCAMPER |
|---|---|---|
| Brainstorming | The angle: the group answers an open question and finds its angles itself. | No object exists yet. |
| Six Thinking Hats | The group's stance. The object stays intact. | The group has to examine a proposal from several angles without producing variants of it. |
| Assumption reversal, within creative thinking | A belief about the object. The letter R reverses the object itself: the order, the roles, the direction of a flow. | A shared conviction blocks the group, for example "customers will never pay online". |
| Reasoning by analogy, within creative thinking | The starting point: a domain unrelated to the problem. The letter A borrows a solution that exists elsewhere for the same problem. | No other department and no other sector has solved the problem. |
| Vuja-de | How the team sees a familiar object: it observes the object as if it were new. | Upstream: the team no longer sees what is wrong. The result can become the object of a SCAMPER round. |
| Kill a stupid rule | The internal rules, which the group names and challenges. | Upstream: the group is looking for the rule that blocks. That rule can become the object of a SCAMPER round. |
| TRIZ | The contradiction between two characteristics of the object. | Every gain on one characteristic degrades another. |
| Morphological box | The whole object, broken down into parameters whose combinations it runs through in full. | The full solution space must be visible. |
AI considerations
A language model applies the seven letters to an object described in a few lines and returns a grid in a minute. SCAMPER is a list of instructions, which a model follows without tiring of any letter. The analyst working alone gets a first grid to complete and correct. In a session, the facilitator restarts a letter that produces nothing with two or three variants from the model, labelled as the model's.
At the sorting stage, the model spots the variants that describe the same measure under two letters and proposes a common wording; the group confirms the merge.
The model does not know the local constraints: the internal regulations, the collective labour agreement (CLA), the accounting system in place. A variant that breaks them looks as plausible as a feasible one, and only sorting against the group's criteria separates them.
The model stands behind no variant: nobody in the room defends an idea that the model wrote. Presenting that idea as the group's output distorts the measure of buy-in.
Describing an internal process to a public model exposes its amounts, its actors and sometimes personal data, which the Federal Act on Data Protection (FADP) governs. The analyst describes the object without real names or figures. A provider that processes real data on the organisation's behalf is a processor within the meaning of Art. 9 FADP: the organisation entrusts it with that processing by a contract that complies with that article.
Examples
The object is expense claim reimbursement at an SME in the canton of Vaud with 120 staff. The goal is to shorten the time to reimbursement; the criteria are the effect on that time and the implementation effort. Before the round, the group agrees on a sorting rule: it drops what does not shorten the time, holds for study what needs high effort and retains the rest.
| Letter | Question asked | Variant | Effect on time | Effort | Sort |
|---|---|---|---|---|---|
| S | What can be replaced? | A photo of the receipt, taken with a mobile phone and attached to the entry, in place of the paper original | Medium | Low | Retained: employees no longer send the originals. The fiduciary must give its view on record retention |
| C | What can be merged? | Travel request and expense claim in a single file, opened before the trip and closed on return | Medium | High | Held for study: depends on the HR tool |
| A | Where is the problem already solved? | A flat meal allowance per travel day, with no receipt, taken from the rules that the sales team already applies | Medium | Low | Retained: the rules already exist |
| M | What if the frequency changed? | Weekly reimbursement, instead of payment with the salary | High | Medium | Retained: the time falls to one week at most |
| M | What if the threshold changed? | Line manager's sign-off required only above CHF 300 | High | Low | Merged with variant E, which describes the same measure |
| P | Who else could use it? | Travel data reused for the carbon footprint report and for negotiating hotel rates | None | Medium | Dropped for this goal, passed to the purchasing manager |
| E | What can be removed? | Line manager's sign-off removed below CHF 300, replaced by an accounting check on a sample of 10% of claims | High | Low | Retained: the wait for sign-off disappears for small claims |
| R | What if the roles were reversed? | Accounting prepares the claim from the company card statements; the employee confirms it | High | High | Held for study: requires a card for every person who travels |
| R | What if the order were reversed? | A flat advance paid before the trip, accounted for on return | None | Medium | Dropped: shifts the time without shortening it and adds a balance to settle |
Visualizations
The deliverable, a list of variants marked with their letter and their verdict, fits in rows and columns: an HTML table that the reader can reread and sort. Osborn's headings against Eberle's letters and the related techniques also fit in tables.
A table does not show how the session links its elements. The object is linked to each of the seven letters, which the group asks one after another; no letter receives the result of the previous one. All of them feed their variants into a single list. Sorting comes after the seventh letter and is the only step that receives the criteria set in advance. These links call for a diagram.
Cost
| Phase | Level | Justification |
|---|---|---|
| Preparation | Low | Name the object, adapt the seven questions, set the goal and two criteria. One hour for the facilitator. |
| Execution | Low to medium | Thirty-five minutes to an hour for the seven letters, thirty minutes for sorting. Working alone, the analyst fills in the grid in half an hour. |
| Documentation | Low | A six-column table: letter, question, variant, effect, effort, verdict. The cost depends on the technique that receives the retained variants. |
Tooling
A whiteboard divided into seven columns and a pad of sticky notes are enough. One note per variant lets the group move it at the sorting stage. The participant writes the letter in a corner of the note, so it stays readable after regrouping. Seven cards, each carrying a letter and its questions and turned over one at a time, keep the group on the current letter.
Digital whiteboards such as Miro, Mural, Conceptboard or Microsoft Whiteboard reproduce the seven columns for a distributed team and let everyone write at the same time. For a Swiss organisation, data residency guides the choice if the object is an internal process. A spreadsheet is enough for sorting: one column for the letter, one per criterion, a filter on the verdict.
Sources
- Eberle, R. F., "Developing Imagination Through Scamper", The Journal of Creative Behavior 6(3), 1972, pp. 199-203, DOI 10.1002/j.2162-6057.1972.tb00929.x: the inventor's article, which traces the technique back to Osborn's questions. SCAMPER is one author's method; no business analysis or project management body of knowledge describes it.
- Osborn, A. F., Applied Imagination: Principles and Procedures of Creative Problem Solving, Charles Scribner's Sons, 1953: the list of questions from which Eberle drew the acronym.
- Institute for Manufacturing, University of Cambridge, "Osborn's checklist": the nine headings of Osborn's list and how they relate to SCAMPER.