What the problem is
Almost no public institution has a problem of ideas; what it has is a problem of enabling system. It knows what ought to be improved and cannot get a promising initiative through its own procedures.
When you measure the institutional maturity of a public organisation, the pattern that shows up is remarkably stable, and it fits in four words: connected outside, tied up inside. External connection is usually the strongest dimension; administrative agility is the weakest of all, by some distance.
That imbalance produces four specific effects, and all four can be dismantled:
- The procedure decides what can be innovatedAn initiative that does not fit the standard process does not advance, however good it is. The organisation does not discard the bad ideas: it discards the ones that do not fit.
- Nobody owns improvementIdeas are generated and never embedded because they have no owner. Responsibility is spread thin between the service that suffers them, the service that processes them and the service that authorises them.
- Everything rests on enthusiasmImprovements are carried by motivated people doing it on top of their own job. When that person leaves, moves post or gets tired, the capability goes with them.
- There is no institutional learningWhat is learned in one redesign is neither documented nor passed on, so the next one starts from scratch. The organisation accumulates experience without accumulating capability.
Embedding with guarantees
The service starts only after measuring whether the institution is in a fit state, and it rests on firm, sustained foundations.
- Institutional experimentationReal backing for new ideas, and room from senior management to let them be tried.
- Safety to challengeRoom to put forward imperfect approaches and to question routines that nobody defends any more and nobody withdraws either.
- Competencies for complexityChances to build the ability to work with uncertainty, with change and with collaborative problem-solving.
- Flexible resourcesThe ability to move time, space or budget to an initiative that is working, without waiting for the next financial year.
- External connectionThe ability to work with outside players and to create value with them.
- Administrative agilityProcedures that let things move without needless delay or ambiguity. This is, by some distance, where institutions fail most.
- Digital and AI maturityUsing digital tools to redesign how value is created, and being responsibly prepared for emerging technologies.
- Evaluation and learningDefining what success is before embedding anything, and using the evaluation to decide what comes next.
The result reads service by service and unit by unit, and it does two things: it decides where to start and it sets the baseline against which progress will be measured two years on. It also reveals which areas have not answered, which is usually as informative as what the rest answer.
The core team runs the process
The team makes sure the service innovation infrastructure combines three functions, and it fails if any one of the three is missing.
- FUNCTION 1A co-creation spaceThe core team runs the place where the people who use a service, the people who deliver it and the people who direct it sit at the same table around a real problem. It is centred on users, but not exclusive to them.
- FUNCTION 2A prototyping platformThe team provides somewhere to test a service solution at small scale, with explicit rules and with permission for it to go wrong, before the whole organisation is committed.
- FUNCTION 3A training hubWhere the institution's own people are trained in facilitation, evidence, prototyping and change management. It is what makes the capability stay.
Which profiles hold it up
The point where these projects usually fall is not the design, it is the switching on. A lab with no named team ends up being a room people book for meetings.
The core team: five people
The number is not arbitrary. With fewer, the functions do not fit; with more, it can no longer be convened and becomes another committee.
- Institutional sponsorshipSomeone from senior management who opens doors, removes obstacles, protects the mandate and connects the lab with whoever decides.
- Facilitation and redesignWhoever designs and runs the participatory processes, the prioritisation, the prototyping and the learning cycles.
- User experienceWhoever makes sure the user's perspective enters the decisions and not only the workshops.
- Quality, data and processWhoever builds baselines, process maps, indicators and evaluation routines. This is the profile everybody recognises as necessary and almost nobody includes.
- Communication and change managementWhoever makes the process visible, understandable and credible inside the institution: recognises the patterns of resistance, maps who loses what with each change and holds the plan that goes with it. Without this profile, the core team designs well and gets nothing adopted.
The rule of the three owners
Every embedding project leaves with three names attached: a user owner, a service owner and a method owner. It is the rule that stops a good idea ending up on the wall, and it is the one that meets most resistance when it is proposed.
The annual programme
An embedding effort with no calendar of its own fills up with whatever other people ask of it. The programme combines open access, fixed activities and project cycles.
- The challenge of the quarterOne service problem is chosen and stated publicly. Out of it comes a commission with an identified user and a first concept.
- The redesign sprintTwo to five days to redesign one specific service or process, with the people who deliver it and the people who receive it.
- The problem clinicShort sessions where anyone brings a practical problem and leaves with an improvement experiment, not with a complaint on file.
- The suggestion box with a reply loopContributions are collected permanently, with an explicit commitment: every contribution gets an answer and can be tracked. Without the loop, the box destroys trust instead of building it.
- Mentoring hoursScheduled support from practitioners, service design specialists and people who know their way around data.
- The showcase dayA regular presentation of prototypes, of what has been learned and of the changes embedded. It includes what was dropped and why.
How it is done
Four phases across two years, with an explicit decision at the end of each on whether to continue, adjust or stop.
- Maturity diagnosis and decision · before startingThe maturity diagnosis issues a recommendation that may be negative: some institutions are not ready for a redesign yet, what they need is to unblock their administrative pipeline.
- Start-up · months 0 to 3Naming the sponsorship and the coordination, choosing a pilot space, setting the rules of access and the entry form for projects, and selecting the first challenge. Out of it come the lab's founding charter and an initial portfolio.
- Minimum viable pilots · months 3 to 6First challenges, first redesign sprint, suggestion box live and training for the first group of facilitators. Out of it come the first prototypes, the first measurement board and the first showcase day.
- Integration · months 6 to 12Connection with the rest of the institution and with outside players. The pilot stops being a project and becomes a recurring programme with documented methods.
- Scaling · months 12 to 24Replicating the model in other services or in other institutions, with a shared repository of methods and an honest comparison of results.
Where this service ends and the redesign one begins
Both work on services and they are worth keeping apart, because they are commissioned at different moments.
Service Redesign Lab
One specific service, eight weeks, one deliverable. It is commissioned when a service is causing trouble and has to be remade with the people who use it.
Embedding strategy
The capability to do that continuously and without outside consultants. It is commissioned once one redesign has been done and ten are wanted, or when the bottleneck is not a service but the institution's own agility.
The maturity diagnosis is what decides which one is due. If administrative agility comes out as a critical gap, starting with a redesign produces an impeccable diagnosis and nothing embedded: the redesign will get stuck at exactly the point the diagnosis warned about.
What makes it different
It starts by measuring, not by proposing
The maturity diagnosis may conclude that the time is not right. I would rather say so before charging for a lab than after opening one.
It goes at the conditions, not at the projects
The lever is not in running better improvement projects: it is in procedures, resources and competencies that let a promising project advance.
Five people and three owners per project
The two rules that separate a lab from another committee. They are agreed before launch and in writing.
It anticipates tokenism and blocks it
Inviting users to a workshop and then not letting them influence the decision destroys more trust than not inviting them. Participation gets into the framing of the problem and into the decision, or it does not get in at all.
The capability is installed, not lent
Training the institution's own facilitators is in phase 3, not at the end. The stated aim is that they run the second sprint.
What gets dropped is documented
A lab that only shows its successes does not generate institutional learning, it generates public relations.
Who it works for
- Public administrationWhen improving services depends on particular people rather than on a mechanism, and there is a wish to stop starting from scratch every time.
- Universities and university systemsWhen many services live side by side with similar problems and each one solves them on its own, or does not solve them.
- Alliances and networks of institutionsWhen several organisations want a common model, a shared repository of methods and pilots that can be compared with one another.
- Non-profit sectorA variant aimed at social innovation, with the same structure and a different impact criterion.
How to commission it
What your organisation puts in
- Sponsorship with the power to remove obstaclesFrom senior management, not delegated. If the lab has to ask permission through the usual channel for every experiment, there is no lab.
- Five people with recognised timeFormally recognised, not added on top of a working day that is already full. This is the risk that kills these projects most often.
- A space, even a borrowed oneNo building works or new furniture are needed to start. What is needed is somewhere identifiable, bookable and visible.
- Two or three volunteer servicesFor the first pilots, including one highly visible service and one administrative one. Both types are needed: the first so that it is seen, the second so that it is believed.
- A fast, compliant routeA commitment to build a procedure that lets a pilot advance with proper safeguards when it does not fit the standard process. Without this, the rest is scenery.
How it gets agreed
The scope and the terms are set in a conversation beforehand, based on the size of the institution, the number of services involved and how far you want to go. The maturity diagnosis can be commissioned separately, and it is what makes it possible to decide the rest with any judgement.