Co-Designing Services With Customers: Beyond the Focus Group

Ask a customer what they want and they will describe a faster version of what already exists. Put that same customer in a room with a whiteboard, a stack of sticky notes, and the authority to redesign the process themselves, and something different happens: they stop describing friction and start solving it. That shift — from informant to co-author — is the whole case for co-design, and most organisations that claim to practise it have never actually made the leap.

Co-designing services with customers means giving them a genuine role in shaping decisions about how a service works, not just a seat at the table to react to decisions already made. The distinction matters because it determines whether the output is a design people believe in or a design people merely approved. Done properly, co-design produces solutions the frontline will defend and customers will forgive when things go wrong, because they built the thing that broke.

What does it actually mean to co-design a service?

Co-design is a structured practice, not a focus group with better snacks. It borrows from participatory design traditions in Scandinavian workplace research from the 1970s, where unions insisted that workers who would use a system should help build it, and it has since become standard practice in service design methodology. The Nielsen Norman Group's work on participatory design makes the same point for digital products: involving users in the generative stages of design, not just the evaluative ones, produces solutions that better fit real behaviour because the people closest to the friction are the ones proposing the fix.

In service design terms, co-design usually means bringing customers into the same room as frontline staff and designers to work directly on a journey map or service blueprint — annotating where it breaks, redrawing the steps that make no sense, and sometimes sketching the touchpoint themselves. The customer isn't asked "how did that make you feel?" after the fact. They are asked "how would you fix this?" while the process is still on the table.

Why does co-design outperform traditional customer research?

Traditional research — surveys, interviews, satisfaction scores — tells you where the pain is. It rarely tells you what to do about it, because customers are being asked to evaluate something they had no hand in shaping. Co-design closes that gap by moving customers upstream, into the moment where the design decision is actually made.

The behavioural mechanism behind why this works harder than passive feedback has a name: the IKEA effect. In a study published in the Journal of Consumer Psychology in 2012, researchers Michael Norton, Daniel Mochon and Dan Ariely found that people place a disproportionately higher value on products they partially built themselves compared with identical pre-assembled versions — and that this labour-driven attachment held even when the self-built object was objectively worse. Apply that to a service redesign: a returns process a customer helped rebuild is one they will trust more, forgive more readily, and recommend more warmly than an identical process handed to them fully formed. Ownership, even partial ownership, changes the relationship with the outcome.

There's a second mechanism worth naming: reciprocity. When an organisation visibly hands over real decision-making power — not a token gesture, but actual authority over a design choice — customers tend to repay that trust with more honest, more constructive input than they would give in a standard feedback form. They stop performing satisfaction and start problem-solving, because they've been treated as a partner rather than a data point.

Why do most co-design workshops fail to change anything?

Because they are theatre. I've sat in enough of these sessions to recognise the pattern immediately: customers are invited, given post-it notes, thanked warmly, photographed for the internal newsletter — and then the roadmap that ships three months later looks exactly like the one that was already in the drawer. Nothing customers said moved a single decision. That is not co-design. That is consultation with better lighting.

The tell is almost always the same: no one in the room has the authority to change the thing being discussed. You can spot a co-design session that is genuinely built for impact versus one that exists to generate goodwill by checking for these signs:

• A decision-maker is in the room, not just observing behind glass. If the person who owns the budget or the process isn't present, nothing said can become binding.
• The starting artefact is unfinished, not a polished prototype for approval. Customers redesigning a nearly-final concept are being asked to rubber-stamp, not co-create.
• There is a visible mechanism for saying no to leadership's preferred idea. If every session conveniently validates what the business already wanted to build, it wasn't co-design — it was confirmation bias with a guest list.
• Frontline staff are in the room alongside customers. They are the ones who will operationalise whatever gets agreed, and they usually know within seconds which ideas are workable.
• Something changes within weeks, not quarters. If the gap between the workshop and the first visible change is longer than a business quarter, the organisational appetite was never there.

How do you run a co-design workshop that actually produces change?

The mechanics matter less than the sequencing. Skip a step and the whole thing collapses into a nicer-sounding survey. This is the sequence I use when a client brings customers into a redesign, adapted from the same generative-then-convergent logic behind the UK Design Council's Double Diamond framework, which separates the work of discovering the right problem from the work of building the right solution.

1. Recruit for range, not comfort. Invite the customers who complain the loudest and the ones who churned quietly without saying why. A room full of happy advocates will co-design a service that flatters the status quo.
2. Bring the real journey map, warts included. Start the session by walking through the current-state blueprint with every known failure point marked. Sanitising the map before customers see it removes the exact friction you need them to react to.
3. Separate diagnosis from design. Spend the first half of the session confirming where the process actually breaks from the customer's point of view — often different from where internal data says it breaks — before anyone is allowed to propose a fix.
4. Hand over a real constraint, not a blank page. Give customers the actual limits — budget, regulation, system capability — rather than asking them to imagine an unconstrained future. Constrained creativity produces solutions you can ship; unconstrained brainstorming produces wish lists you'll quietly shelve.
5. Prototype in the room. Sketch the redesigned step on paper or a low-fidelity screen mock-up before the workshop ends. Customers should leave having seen their own idea take a rough physical shape, not a promise that "we'll take this away."
6. Assign an owner before anyone leaves. Every idea that survives the session gets a named internal owner and a date for the next update. Ideas without owners die in the following week's inbox.
7. Close the loop publicly. Tell the customers who participated — and ideally the wider customer base — exactly what changed as a result. This single step is the difference between a workshop people remember fondly and one that builds compounding trust for the next redesign.

What's the difference between co-design and a Voice of Customer programme?

Voice of Customer captures what customers say, at scale, usually after the fact — surveys, review mining, call transcripts. It's essential, but it's a listening system, not a design system. A well-run Voice of Customer strategy tells you which touchpoint in the journey is scoring badly and why customers are frustrated there. Co-design is what you do next: it's the room where you decide, with the people affected, what replaces the broken step.

Treat VoC as the radar and co-design as the workshop. Skip the radar and you co-design the wrong problem. Skip the workshop and you have a very accurate map of pain with no mechanism for fixing it. The two should feed each other on a cycle — VoC surfaces the moment of truth worth redesigning, co-design rebuilds it, and the next VoC cycle tells you whether the rebuild worked.

How do you turn co-design output into something that actually ships?

This is where most of the value gets lost. A workshop generates dozens of good ideas sketched on paper; a functioning service needs one integrated blueprint that operations, IT, and frontline training can all work from. The translation step is a service design discipline, not a facilitation one.

Everything customers proposed in the room needs to be mapped back onto a formal service blueprint — front-stage actions customers will see, back-stage actions staff and systems perform, and the support processes underneath. Where a customer's redesigned idea requires a new back-stage capability the organisation doesn't have — a real-time inventory check, a different escalation path — that gap has to be named explicitly, priced, and sequenced, rather than quietly dropped because it's inconvenient. Customers proposed the front-stage experience; the blueprint is where you prove, or disprove, that the organisation can actually deliver it.

Some of what surfaces in a co-design session isn't a process fix at all — it's a signature moment worth preserving deliberately, the kind of small ritual that customers mention unprompted when they describe why they stayed loyal. Building those into formal customer rituals rather than letting them happen by accident is one of the more underused outputs of a good co-design session.

What happens when leadership treats co-design as a box to tick?

It backfires harder than doing no research at all. Customers who are invited to shape a service and then watch it launch unchanged don't conclude the organisation is neutral — they conclude it asked in bad faith. That is the flip side of the IKEA effect and reciprocity: the same mechanisms that build outsized goodwill when honoured build outsized resentment when violated. You are better off never running the workshop than running it and ignoring the output, because the second option burns trust with the exact customers who cared enough to show up.

The moment you hand customers a real pen and then throw away what they drew, you haven't just wasted an afternoon — you've taught your most engaged customers that engagement doesn't pay.

The same logic in the original Bain & Company study on the perception gap between companies and customers applies here. In its 2005 report Closing the Delivery Gap, Bain & Company found that roughly 80% of companies believed they delivered superior experiences, while only about 8% of their customers agreed. That gap doesn't close by asking more questions in a survey. It closes when the people experiencing the gap are handed the pen to redraw the process themselves — and when the organisation actually uses what they draw.

Where does co-design fit inside a broader CX operating model?

Co-design isn't a one-off event; it's a capability an organisation either has or doesn't. Building it means embedding customer participation at defined points in the service design lifecycle — not just at launch, but at the point where a journey is being redrawn after a maturity assessment flags it as weak, or where a new customer archetype reveals a segment whose needs the current blueprint never accounted for. Organisations that treat this as infrastructure, rather than a campaign, build a standing panel of customers they can convene quickly, a facilitation playbook that doesn't need reinventing each time, and — critically — a leadership habit of showing up in the room rather than reviewing a summary afterwards.

That standing capability is worth benchmarking. A structured CX maturity assessment will tell you honestly whether your organisation currently treats customer input as a design input or a satisfaction metric — and those are very different starting points for building a co-design practice that lasts.

What's the one habit that separates real co-design from theatre?

It's the willingness to let a customer say no to an idea leadership already liked. Every other element of a good co-design practice — recruitment, prototyping, blueprinting, closing the loop — is procedural and learnable within a few cycles. This one is cultural, and it's the one most organisations quietly avoid, because it means accepting that the room might overrule the roadmap.

Renascence has run this kind of session inside organisations redesigning onboarding flows, complaint journeys, and loyalty experiences, and the pattern holds every time: the version customers help build is the version the frontline defends six months later without being told to. That's not a soft outcome. It's the difference between a redesign that survives contact with reality and one that quietly reverts the moment the project team moves on. If you're weighing whether your next journey redesign needs customers in the room or just customers on a survey, that's usually the tell — ask yourself who currently holds the pen, and whether you're willing to hand it over.