Root cause analysis | Data-driven municipality | Rotterdam
This case study reveals the importance of using data-driven working to examine the underlying problems in social work. How do you put people to work sustainably? Read Part One here.

Focus on underlying problems: root cause analysis (NPRS case study part II)

Written by

Data-driven working enables paradigm shifts

Part one of our NPRS case study describes how the social work program attempts to put people back on track to gainful employment. The real trick isn’t putting people to work, however, but keeping them there. How did they put people to work without them backsliding? Read part II of our NPRS case study.

The NPRS case part II: data-driven root cause analysis

If your only KPI is “number of people put to work” (Plan phase of the PDCA cycle) and every action is focused on this (Do phase), you’re essentially creating a problem: a situation where even people who struggle with addiction, or who have no permanent residence, are put to work. But addicts or people with a complex home situation, or no home at all, are unlikely to be able to hold down a job. They often can’t be sustainably employed. The odds of this group returning to the social security safety net are huge.

The data-driven approach creates clarity

Figure 1 visualizes this cycle. The red arrow stands for people relapsing into social security. The thicker the red arrow, the more people relapse, and the smaller the group of people that you’re genuinely helping in a sustainable fashion. Thanks to the data-driven approach and the presentation of the “real” numbers, ie. people put to work sustainably, everyone involved in these systems can develop an understanding of how the various movements towards work or school and the relapse mechanics relate to each other.

By checking the numbers, you can see exactly which people from which categories have relapsed (Check phase). When you see this, you have the moral duty to adjust your plans for these demographics (Act). This root cause analysis reveals the underlying problems preventing people from staying employed.

Figure 1

Figure 1: Simplified representation of movements towards work (green) and relapse (red).

In short: If you want to put people to work sustainably, you have to solve the underlying problems of the involved citizens.

Fancy plans and unrealistic targets demotivate people

Mark de Kort, Passionned Group’s interim consultant at NPRS: “You can make all the plans you want, for example saying that you’re going to help 3000 people get jobs. But what if it turns out that 2000 of those are unable to stay employed? You already know they’re going to relapse back into social security. Not only does that cost a lot of money, but it also demotivates those people.

You say you’re going to help 3000 people get jobs, but 2000 of them are unable to stay employed

They thought, ‘I’m doing better, I’m getting work.’ But they’re not. That’s a heavy blow. The social workers that help these people are also demotivated, because to them falls the thankless task of motivating those people again. These are the stories you’ll hear if you listen to what’s actually going on here.”

Choose a tailored approach

Instead of striving for one unattainable goal, strive for many smaller, attainable goals. If you aim for one-sided goals without analyzing root causes, you’re short-changing the people and their complex situations. It’s much better to determine which relevant sub groups comprise the entire population. Then, you can plan actions per sub group’s needs. Data-driven working enables this nuanced differentiation. It allows you to see what works in a specific situation and what doesn’t.

Example: data-driven numbers

Let’s take a random example. Instead of helping 1000 people halfway or not at all, it’s better to genuinely help 200 people in X way, 50 people in Y way, and 100 people in Z way. Data-driven working helps you fulfill promises in a focused way. For example: instead of making empty promises about helping 1000 people get work sustainably, first take a data-driven approach. See what story the data tells. Then you’ll see what you can actually make happen, because you have to consider all the causes, movements, and relapse mechanics that affect every sub group. There’s no such thing as “the one demographic”. You have to look for relevant sub groups.

Dispelling illusions

Thanks to data analytics (tools) you know that you can actually help citizens. For example, you can see that approach X works for group A. But you can also see that it’s much better to help another group by first addressing their addiction issues. Instead of making an empty promise, perform the right analyses and dispel the illusion that you can help them with a simple solution. In the future, you’ll know exactly what you can promise people. That prevents people being bounced around within the existing system, going from one social worker to the next. You can put them to work, but you can already foresee the relapse. Just passing on the problem to the next social worker is a thing of the past.

Uniform policy doesn’t work

The trick isn’t to go for a one-size-fits-all approach, but to carefully work towards a network of diverse policies that can be relevant to many different types of people. Data-driven working and the accompanying software enable this paradigm shift, because the technology enables politicians to see a much more nuanced image of reality. This also leads to smaller and smaller PDCA cycles, but they’re PDCA cycles that are guaranteed to work and structurally contribute to the overall effect.

Look for smaller PDCA cycles that work per demographic. Make plans that are attuned to the potential of every sub group. Not every approach works the same everywhere. What kind of work can every sub group handle sustainably, and why? What are the root causes of the problems that arise? In other words: what is required to prevent entire sub groups from relapsing immediately after they start working?

Set targets that you can hit over and over

Mark de Kort: “Targets shouldn’t just be determined from the top down. Targets have to be extrapolated from the data. We make this process transparent and open for discussion. More and more, social workers are having conversations with coordinators: ‘These top-down targets are the result of political ambitions, but in reality, these goals are unattainable. People are still in debt, so they can’t get back to work sustainably’, is the reply.”

Set smaller, attainable goals

“The first step is determining (based on data) if there are bigger and smaller problems in the background that prevent people from getting back to work. Instead of ignoring that, solve those problems first. That may conflict with a top-down target, especially if the data predicts that it’s a futile effort.” Where the original plan for an area is aimed at putting young people to work, the Actualize phase might reveal that various interventions are required for several groups in the execution of that plan. That leads to a tailored approach where people get the intensive personal help they need. Let go of the philosophy that everyone should be 100% self-sufficient. Don’t set an unattainable goal, but several smaller, attainable goals that really work.

In short: operate with a scalpel, not a chainsaw.

The results after two years of data-driven working

Rotterdam made a start in continuous improvement by working data-driven, or as we like to say, datacratic. We see standardized test scores going up, and the students in Rotterdam South are gaining on students in other parts of Rotterdam and the four big Dutch cities. Also, more students are choosing majors with future job perspectives in mind, such as medicine and technology. The percentage of people on welfare may not have gone down as much as we’d like, but Rotterdam South is slowly reaching the levels of the big four. The area is undergoing the biggest decline in welfare rates of any other. Property values are going up, and they’ve started to address the poor state of private housing. People looking for housing from elsewhere are starting to consider Rotterdam South.

In conclusion

These are hopeful developments, and proof positive for citizens and the NPRS partners that their efforts are bearing fruit. Not in the least thanks to the data-driven approach.

Comment on this post by Daan van Beek

Your email address will not be published. Required fields are marked *

A selection of our customers

Become a customer with us now

Do you also want to become a customer of ours? We are happy to help you with root cause analysis (data-driven municipality) or other things that will make you smarter.

Daan van Beek, Managing Director


Managing Director

contact me directly

Fact sheet

Number of organizations serviced
Number of training courses
Number of participants trained
Overall customer rating
Number of consultants & teachers
Number of offices
Number of years active