A new structure for customization

How to deal with custom services in digital government.

Lately there has been a cry for more customization in government services in The Netherlands. The last couple of years there has been an immense scandal with services for child benefits at our Tax Office, but that’s not all. Stories keep popping up of other citizens who are struggling with government services that don’t fit their needs and who even seem to work against them.

More customization is needed, I keep reading. But that is not the solution. At least… not if customization means ‘working around the system’. Then it’s patchwork. We should rather change the structure of the system.

In this article a plea for a new structure for customization. Not (only) the work-arounds by individual caring civil servants, but a compassionate government, right in the DNA of the system.

In this article:

  • Puzzler Patrick, an example
  • Think outside the scope of your organization
  • Customization can be a standard
  • Types of customization: proactive and reactive
  • One step further: government-wide
  • This requires a new structure
  • The first practical steps

I wrote this blog with the help (and access to documents about the ‘Customization Workshops’, in Dutch: Maatwerkplaatsen) from Maartje Kemme, senior policy officer for Customization at the Ministry of Social Affairs and Maureen Hermeling, strategic advisor for services at the BAR Organization. The drawings are from this video, made by Bord & Stift.

Puzzler Patrick

Patrick works at Loket Geldzaken (a counter for money matters) of the Municipality of The Hague. He is a case manager for residents who come to the counter for help with financial questions. In our interview, he randomly picks the top case of the stack as an example. Mr. N.

Mr N. developed health complaints 3 years ago, which meant that he could no longer work. He lost his job and applied for an unemployment benefit. This meant that his income was a lot less than his previous pay and his rent became increasingly difficult to pay. He applied for rent allowance, but to do so he actually had to apply for a rent reduction from the Housing Foundation first. That is possible, but you must send the forms in the correct order to the correct organizations.

Long story short: after 3 years, Mr. N. had completely lost the overview, he was in debt and he sat opposite Patrick going through his financial matters. Patrick figured everything out for him. Phoned the Tax and Customs Administration, arranged the rent reduction with the Housing Foundation and made sure that N. received a rent allowance and had an overview.

In a previous blog (in Dutch) I mapped a month of my own relation with the government, now we tried to map out the three years for Mr. N. We asked Patrick what else was going on. Were these the only government services that he interacted with due to health or financial problems?

Hm.. that’s a good question… Patrick didn’t know that either. Wasn’t his wife also integrating…? “There you go,” I thought, and I drew a line to my own organization with a — possible — integration debt of about € 10,000 for language lessons.

When you talk to Patrick it’s tempting to think we need more Patrick’s. He is a tower of strength for people like N. and ensures that problems do not get out of hand. You can just make an appointment with Patrick, have a cup of coffee and tell your story. Patrick is a nice guy, at least he listens to you and doesn’t say no like a computer screen does.

But even Patrick has no overview of your entire relationship with the government. And Patrick is also looking for the right puzzle pieces.

A part of the overview of the map of Mr. N. The yellow part is where Patrick helped him.

When we saw this drawing in front of us, we thought: we don’t need more Patrick’s, we have to tackle the puzzle. Those puzzle pieces tumble over each other and if you press the wrong one somewhere, the whole puzzle will warp. We need to address that.

It’s not in our scope

I have been working as a design researcher for the government for about 7 years now. I know how we make those puzzle pieces. I have been making them too.

If you work for a government organization, even if you try to do your job from a citizen’s perspective, you still stick to the boundaries of your own organization. For example, I work at the Execution Agency for Education and I did research into integration. I visited people’s homes and thought about how we can improve the website for this user group. Only… how they interact with the Department for Integration before they get to us, or how the pre-integration process in the asylum seekers’ center went for them? Well…that is beyond our scope. That was not my assignment.

So I only researched and advised on our own puzzle piece. And not about the whole puzzle. But the person integrating has to make the whole puzzle, whether he wants to or not. And whether our puzzle piece fits into his puzzle… I couldn’t see that either because ‘it wasn’t our scope’.

However my colleagues who speak to someone who is integrating on the phone do get questions about the whole puzzle. They try to answer them as best they can. Those are the Patricks in our organization. Caring people who try to deliver custom services and advice in a system not designed for that.

It is not fair to Mr. N. and it is not fair to Patrick that we force them to make those difficult puzzles themselves. After all, it is not good at all to create all kinds of precedents like that. It leads to uncontrollable arbitrariness and erratic government services. You just have to wait and see which official will pick up your case. In addition, it is also a continuous and heavy mental burden for civil servants to operate on the basis of empathy and hardship clauses, instead of clear inclusive frameworks and standards that take humanity into account and are based on an understanding relationship between government and citizens.

And the latter is possible.

Customization can be a standard

When we talk about more humanlike and custom government services, we often think of an extra desk at the entrance, more capacity at the call center or more exceptions outside the automated process. But you can also design a customization in the system. Especially in the system.

Government services don’t make themselves, they are designed by civil servants. They work together in a long relay race from law to counter. If we start with the human perspective, we can design every baton of the relay together with citizens. What do we want the relation between government and citizen to be like? What impact does ‘my baton’ have on that connection? And what is the outcome of my work on the services that citizens use to connect with the government?

For this we need a different structure in the system.

This is how customization can look schematically from the perspective of the citizen:

  1. Citizens should experience all government services as tailor-made. Customization can easily be organized ‘in bulk’. Services can feel perfectly tailored, even if it is the same for everyone.
  2. There are also additional services that not everyone uses that customize the experience even more. For example, asking a personal question at a service counter.
  3. If that is not enough for your unique situation, which is fine, specific customization is required. There we treat unequal cases equally unequally.

That’s where it gets interesting.

Types of customization: proactive and reactive

Proactive customization is in the here and now, between 1 citizen and 1 government counter. This can be a digital counter that is well designed, explains clearly and offers tasks in a user-friendly and customized manner. It can also be a civil servant who goes through your situation with you and seeks and arranges a solution. For example Patrick. This is customization of the actual service provided, on a case-by-case basis.


Reactive customization goes further. The signal from the service is taken out of the here and now to improve the service. This is not just the one-time fix, but actually learning from it: ‘Are these kinds of situations more common? What do others experience if they make a mistake when applying for housing benefit? What stress accumulates in people’s lives who lose their job due to health problems? What other problems do they run into? What can we do differently as an organization to prevent or resolve this more quickly?’ This is customization of the service definition itself, and impacts every service provided from the new definition going forward.

Government organizations need a structure to facilitate this.

At the Social Insurance Bank (SVB) and the Institute for Employee Insurance (UWV), they call this structure Customization Workshops: an agreement to work together in the organization from all those different relay roles and to recognize, investigate and follow up on those practical signals. They think like designers and have a can-do mentality. By working like this, they change the standards in their services so that the next time it meets the citizen’s needs more.

These organizations notice that there are fewer objections and appeal procedures. There are now also other ways to demand change as a citizen. The organization listens to them, without them necessarily having to object in order to be heard. And the regulatory burden for citizens is decreasing; it’s an easier puzzle.

One step further: government-wide

Often dilemmas transcend the organization and they concern the entire puzzle of the government services. These Customization Workshops can be government-wide. Civil servants from multiple organizations can work together as designers. They can research the problems together and design solutions. It is easy to involve political and social aspects, because you already work government-wide. You see how one puzzle piece relates to another. You have an overview of the whole puzzle.

When I researched the Integration Law for example, I could have shared my insights about delays in language lessons and culture clashes with my colleagues from the Department for Integration and the asylum seekers’ center. I could have learned from them and thus gained more insight into the problem. We could have prototyped and tested solutions together. Together we could have involved the ministry more easily because our case would have been solid. In fact, policy officers from the ministry would have already helped out in our Customization Workshop.

But I didn’t, because I was just by myself. Despite my good intentions, I didn’t have the structure to work that way. I had no idea where to start. I didn’t know anyone at the Department of Integration. Even at my own organization we did not have such a customized workshop structure at the time. My assignment was to ‘improve the website’ and not to ‘help people integrate’.

A new structure for human centered services

Of course I am not writing about anything new here. Many organizations are already working on all kinds of initiatives for human centered services. Since my integration research we started looking at how we can share research insights centrally within the organization. Everywhere in the government, projects are popping up to put the citizen first, to communicate more clearly, to experiment with design thinking and to improve services. All important and necessary.

But nowhere have we tackled it this structurally by turning it into a new infrastructure. By changing the DNA of the system. That is the core, and we have to tackle it. Otherwise we will just keep shaving pieces of the puzzle.

It means that we all have to do our jobs differently. That we must start with the perspective of the citizen, and not that of the current legislation. After all, we can change legislation.

We have to set up partnerships that combine policy and service design in a structured way. As organizations, we have to take responsibility for each other’s puzzle pieces. Citizens’ feedback should be leading, not politics.

Time to take the first step

The first steps we can take:

  1. Organizations must hire citizen researchers if they don’t already have them. These researchers work centrally in the organization and exchange their research openly with other organizations. There should be a research community government-wide. This combined research archive must be supplemented with research done by supervisors and scientists, such as the National Ombudsman, the Scientific Council for Regulation Policy (WRR) and more.
  2. Organizations make the agreement and invest in methodologies that help employees to work as designers in a multi-disciplinary manner. For example, how to research a signal about something that isn’t going right in the service. How you can devise and test solutions based on new ways of thinking. Nice and practical.
  3. Leaders and managers give their employees the space to be “compassionate civil servants.” The space to show moral leadership and initiative to stand up for citizens. This requires courage and vulnerability and leaders have to create an environment where that is accepted and encouraged.
  4. Organizations set up a Customization Workshop. A great project manager who can organize and connect well ensures that employees know how to find each other and can work together. There will be budget and time for this type of collaboration. In these Customization Workshops, citizen researchers and designers can collaborate with programmers and lawyers. The Patricks and Mr. N’s are also invited to participate actively.
  5. Organizations create an archive of ‘case prudence’ (case precedent), case histories from which we can learn, and this too is openly shared with other organizations.
  6. When a puzzle piece from another organization is found in such a Customization Workshop (and that happens more often than one might think), contact is sought with that organization. They should also have such a new structure that facilitates this type of collaboration, so you can join them and work together right away.
  7. When necessary, the Customization Workshop scales up and becomes government-wide. Civil servants can thus easily get to work together with policy officers from a ministry. They use the same infrastructure from points 1 to 6.

This way, customization isn’t custom anymore, but the new structure of the system.

Design researcher working for Dutch government - klipklaar.nl