Co-creation is the practice of developing a product or service together with people outside the organization, such as customers, partners, suppliers or residents, so that they shape the result instead of only reacting to it. It sits between market research, where people answer questions, and open crowdsourcing, where anyone can send in an idea.
Most co-creation programs go wrong after the ideas arrive. A company invites customers or partners to contribute, receives more than it planned for, and finds that nobody is ready to decide what happens to each contribution. People who hear nothing stop taking part. The fix costs little if it comes first: settle the route every idea will take, and the answer its author will get, before asking anyone for anything.
What Is Co-Creation in Business?
In business, co-creation means people outside the company help decide what it builds, with an agreed say in the result. Every format comes down to two choices made before launch: who can take part, from a handful of invited customers to anyone at all, and who decides which contributions go forward.
| Format | Who takes part | Who decides what goes forward |
|---|---|---|
| Customer advisory council | A small group of invited customers | The company, after hearing them out |
| Co-development project | One named partner or supplier | Both sides, against a shared plan |
| Open challenge | Anyone who can answer a stated problem | The company, against published criteria |
| Idea platform with public support | Anyone, with supporters building the shortlist | The company, once an idea passes a support threshold |
| Participatory budget | Residents of a city | Residents by vote, with the city committed to build the winners |
Each row carries a different promise. An invited customer expects to be heard in the room. A partner in a co-development project expects a joint plan and a shared timeline. Someone answering an open challenge expects to be judged on the criteria that were published, and nothing else.
Co-creation is a form of open innovation, and it differs from open crowdsourcing in one respect: contributors know in advance what their contributions can change and who will decide. The same discipline applies to collaborative innovation with employees, but outsiders are less forgiving. They owe the company nothing, and when the promise is broken they simply stop contributing.
Why Do Customer Co-Creation Programs Lose Their Contributors?
Because the company asked before it decided what would happen to the answers. Contributions arrive, nobody owns them, and the people who sent them hear nothing. A clear no rarely ends the relationship. Being ignored does: in a 2026 survey of 1,500 US and UK consumers, 56% said they had left a brand that kept ignoring their feedback.
The same Quantum Metric benchmark found that 52% of the 750 digital leaders it surveyed had no formal process for acting on the feedback they collect. That gap between collecting feedback and acting on it is where customer co-creation loses people, and it opens on the first day an idea goes unanswered.
Even a program with nearly 20 years of practice hears the complaint. Salesforce has run its IdeaExchange since 2006, and more than 85,000 members of its Trailblazer community contribute to it every year. In a September 2025 post, its vice president of research and insights acknowledged what contributors had been saying about their ideas:
“We understand your need for additional transparency and more frequent updates on your ideas.”
Rebecca Sherrill, Salesforce, September 2025
The same post set out a response built around answers rather than volume. IdeaExchange gained views of recently updated and recently delivered ideas, alongside two programs launched at Dreamforce 2024: roadmap previews and a forum for feedback on features under consideration. Its most engaged contributors were invited to a new customer council. The new views answer the complaint: they show what happened to ideas contributors had already given.
What Should You Decide Before Asking for Ideas?
Decide the route each contribution will take, from the problem it answers to the reply its author receives, before inviting anyone. Three decisions carry most of the weight. Each is cheap to make before launch and expensive to make afterwards, when contributors are already waiting to hear what happened to their work.
- The decision the ideas will feed. Name it, with its budget and its deadline. “Ideas to improve our product” invites everything and commits to nothing; “ways to cut onboarding time for new clients before the March release” can be answered.
- An owner on each side. Inside the company, someone with the authority to say yes or no. On the contributors’ side, someone who can speak for them, such as a customer’s product lead or a supplier’s engineering manager.
- The answer and its date. Every contribution gets a decision and a reason, by a date announced at launch. A no with a reason keeps people contributing; silence teaches them not to bother. In the same Quantum Metric study, half of consumers said they need visible proof that an issue was fixed before they will keep trusting a brand.
Credit, rewards and ownership of what is submitted follow from these, and they belong in writing before the first contribution arrives. The Open Innovation Sourcing blueprint treats the published problem statement as the make-or-break asset of an open call, and lists IP, confidentiality and award frameworks among the things to prepare before it opens.
The City of Paris makes this kind of promise for its participatory budget. Residents propose ideas and vote on them, and the city states in advance that a winning project will be carried out by its own departments. In September 2025, 162,395 residents voted on 261 projects, and 104 won. The commitment came before the ideas, which is what makes the vote worth casting.
What Do Successful Co-Creation Examples Decide in Advance?
The route from idea to decision, and what each step earns the contributor. LEGO Ideas, the LEGO Group’s platform for fan-designed sets, is the clearest public case. Anyone can post a design, and the rules say in advance what it takes to reach a review and what a review can produce.
What this looked like in practice. Most co-creation examples describe the launch. LEGO’s is worth studying for what it published afterwards. A fan posts a design and gathers supporters, and a design that reaches 10,000 supporters qualifies for the next review, where LEGO decides whether it becomes a set. Designs that fall short on a point such as timing can wait in a “parking lot” for up to three review periods.
How LEGO Answered a Record 146 Designs in One Review
In 2025 the rules met a surge. Between May and the start of September, a record 146 designs passed 10,000 supporters, more than double the previous record of 71, set in the first review of 2023. The route did not change to cope.
LEGO assessed every one of them and published the outcome in April 2026. Three designs, among them a street of Amsterdam canal houses, cleared the review and moved into product development. A fourth, based on The Old Man and the Sea, came out of the parking lot to join them, and 143 did not make the cut. Meanwhile, 75 more designs reached 10,000 supporters between September 2025 and January 2026 and entered the next review.
About two in every hundred qualifying designs got a yes in that round, and fans kept coming. What LEGO promises is a decision, reached by a route anyone can read before they start, from the people able to make it. For a company program, the part worth copying is the published route rather than the fan base: a threshold for what reaches review, review dates people can plan around, and outcomes announced either way.
How Do You Measure a Co-Creation Program?
Count what happened to the ideas, not only how many arrived. A program that measures volume alone rewards asking widely, the habit that creates the backlog in the first place. The useful numbers describe the promise made at launch and whether it was kept, contribution by contribution, round after round.
| Measure | What it tells you | A reading from the examples here |
|---|---|---|
| Answer rate | Whether every contributor got the decision they were promised | LEGO: all 146 qualifying designs decided in one review |
| Time to answer | How long contributors wait for that decision | LEGO: about eight months, from September 2025 to April 2026 |
| Go-ahead rate | How much of what arrives the program can act on | Paris: 104 of the 261 projects on the 2025 ballot won |
| Return rate | Whether contributors trust the route enough to come back | Not in the published figures; track it from the first round |
Return rate is the hardest of these to fake: a contributor who comes back has judged the route worth a second try.
Paris shows what the long view looks like. By January 2025, residents had submitted 23,912 ideas since the budget began in 2014, and 1,345 had won a vote. A funnel that narrow still works, because the route and the commitment were public from the start. The Impact-Driven Innovation approach applies the same test inside a company: judge a program by what it implements and what that delivers.
What Should a Co-Creation Platform Track?
The promise, idea by idea. A co-creation platform earns its place when anyone can see, for any contribution, who owns it and what its author has been told. Inboxes and spreadsheets can hold ideas but not commitments, which is why answers start to slip once volume grows.
Look for a platform that can:
- link each contribution to the brief and the decision it feeds
- route it to a named owner, with a date for the answer
- record the decision and the reason in words the contributor can read
- tell the contributor when its status changes
- keep the agreed terms, such as credit and rewards, attached to the contribution
Volume is what breaks the manual version. LEGO had 146 qualifying designs to answer in a single round, and Salesforce hears from more than 85,000 contributors a year. Neither promise could be kept from an inbox.
A quick test for any program already running: pick a contribution from three months ago and ask who owns it, what was decided and whether its author knows. If that takes more than a minute, the program is running on memory, and memory does not survive the second round. A clear idea evaluation process is what makes the answer quick.
Q-ideate is built for the part of co-creation that starts when the ideas arrive. Each challenge runs with its own audience, timeline and evaluation process, and every idea moves through a workflow with an owner and a status at each stage, so contributors can be told what happened to theirs. See how Q-ideate runs idea challenges.
Co-Creation: Common Questions Answered
Co-creation means developing a product or service together with people outside the organization, such as customers, partners or residents, so they shape the result instead of only reacting to it. What separates it from consultation is an agreed say: contributors know what their contributions can change and who will decide.
LEGO Ideas is a well-documented one. Fans post designs for new sets, and any design that gathers 10,000 supporters is reviewed by LEGO. In the review published in April 2026, a record 146 designs qualified and three were approved for product development. Every qualifying design received a decision.
Crowdsourcing asks a large, open group for ideas or work, often with little say in what happens next. Co-creation gives contributors an agreed role in the result, whether they are a few invited customers or thousands of fans voting. The difference is the promise made at the start about what their contributions can change.
Two people: one inside the company with the authority to say yes or no to what comes in, and one on the contributors’ side who can speak for them. Without an internal owner, contributions wait for a decision nobody can make. Without an external one, the partner’s commitment rests on goodwill alone.
Decide before launch and put it in writing. Rewards can be money or public credit, and partners who help build something will expect agreed ownership of it. What matters most is that the terms are known before anyone contributes, so a good idea does not turn into a dispute once it ships.
By the date you promised when you asked. The right interval depends on the decision: a vote on a feature can close in weeks, while a product review like LEGO’s takes months. Publishing the date matters more than its length, because contributors can plan around a known wait.