What if the obsession with tracking KPIs is actually preventing public services from delivering real outcomes?

In this episode, Joe sits down with Toby Lowe, a leading academic and steward of the Human Learning Systems movement. Toby explains why the traditional model of setting targets and performance managing for outcomes is fundamentally broken, and how a Human Learning Systems approach can reshape public services around trust, complexity and continuous improvement.

Drawing on decades of research and his own early frustrations as an arts charity chief executive, Toby argues that using outcomes as performance targets inevitably leads to data manipulation and systemic gaming rather than genuine impact. He shares striking examples from the VW Dieselgate scandal to an £80 million public service contract in Plymouth with zero KPIs to show how shifting the focus from "delivery" to "learning relationships" can drastically improve citizens' lives, even reducing avoidable deaths. This is a deep dive into complexity science, the libertarian roots of modern management theory, and how leaders can liberate their teams to do the work that actually matters.

In this conversation, we explore:

• Why 40 years of research proves that outcomes-based performance management universally leads to data gaming and lying
• The moment of frustration in youth criminal justice that turned Toby away from traditional outcomes frameworks
• The VW Dieselgate scandal and the inherent danger of relying on proxy measures
• Why an obesity systems map proves that governments cannot "deliver" complex outcomes alone
• The libertarian origins of Public Choice Theory and how James Buchanan sought to delegitimise the welfare state
• How to build an alternative management paradigm centred on mastery, autonomy, and purpose
• How the Plymouth Alliance eliminated KPIs entirely and slashed avoidable deaths over a decade
• Why data becomes more important, not less, in a Human Learning Systems approach
• How Liverpool Combined Authority broke a low-trust cycle to transition into commissioning for learning
• Toby's secret life as a hobbyist DJ and the exact complexity book he recommends most

This episode is especially relevant for:

• Public sector leaders, commissioners, and chief executives
• Local government directors and policy professionals
• Systems-change practitioners and complexity theorists
• Service designers and voluntary sector managers trying to solve complex human problems

Stay connected:

• Follow Joe on LinkedIn: https://linkedin.com/in/joseph-badman
• Read the latest chapters from our upcoming book: https://relationalservicedesign.com
• Master the methodology: https://basistraining.co.uk/
• Partner with us: https://basis.co.uk

So I've been looking into your background and it's a very interesting path.

And there's this Napoleon quote which my colleague Martin gave me as a very auspicious start, which, which is, to understand a man, you've got to know what was going on in his life at 20.

So what was going on in your life at 20 and how did it sort of set you on the path that you're on now?

So I was a student at Newcastle University studying politics and history.

And what I discovered there was that I was.

The thing that I was most interested in was political theory and political philosophy.

That set me on the path to being a person who tries to understand the world by making helpful conceptual frameworks.

Because if you kind of, if you cut me open, there's like a little conceptual framework holding me together.

Let's not do that.

Okay, great.

So what did you, what did you head on to do after university?

What were the, what were the first steps?

So I did a, did my PhD in political theory.

So I was interested in the idea of community.

So this was the late 90s and there were the terms community care, community policing were just starting to come into the, of political discourse.

And I was really interested in the work that the word community was doing in those things.

And so I did a PhD that tried to explore what politics would look like if it took the idea of community seriously.

My PhD wasn't all that great because I, at the same time I also discovered dance music and maybe I put a little bit too much time into dance music, not quite enough time into studying, but.

Well, we're going to come back to Dad's music.

I didn't realize that you DJed, so we're going to, we're going to touch on that later on.

So you, you found yourself working at the Arts Council and then you were a chief executive of an arts charity.

Tell me a little bit about, about that journey.

So by the time I finished my PhD, didn't know what I wanted to do, but I knew I definitely didn't want to be an academic.

So I kind of failed in that life goal.

So what I did was I searched for any jobs that had the word communities anywhere in the job description and ended up working.

So my first job after the PhD was being a regeneration officer for Newcastle Council for Voluntary Service, you know, the umbrella organization for the voluntary sector in, in Newcastle.

And that was amazing because it gave me a really good overview of a bunch of social policy about social exclusion and all of that kind of world.

So I spent 15 years then working across different bits of the public and voluntary sector.

And most likely I was a chief executive of a kind of medium scale participatory arts charity.

So we were doing arts activity with some of the most marginalized and disadvantaged folk in the northeast of England.

So kids in the criminal justice system, people with mental health problems, people living in deprived neighbourhoods, look right across that spectrum of social exclusion.

So my suspicion is there's a moment in that work of real frustration that sets you on the path that you're on now.

Is that right?

What is that moment?

Exactly?

It was basically the last stop before young offenders institution for kids that had, that were serious or prolific young offenders.

And so they had to do 20 hours of court mandated positive youth work a week.

So if they didn't turn up to this activity they got sent to prison.

And we did four hours of arts activity with them as part of this 20 hours.

And it was, it was absolutely amazing work, like transformative for the kids because they were given some creative control of their lives and how they wanted to express themselves.

And the, the person who ran the program could see the effect it had on these kids because he was close enough to the work and he had discretionary budget to spend and so he would contract us to, to work with these young people.

And then in year about.

So we overall this was about 10 year partnership but in about year 6 or 7 of this partnership the council adopted Mark Friedman's outcomes based accountability approach.

And the effect of that was for the program manager say to me, Toby, if you want your £30,000 next year, you're going to have to claim that you prevent these kids from reoffending because that's the outcome I'm responsible for in the council's outcomes framework.

And I'm not allowed to spend any money unless it, I can show that it prevents, it achieves that outcome.

So I says to him, but, but you know that we can't legitimately claim to prevent these kids from reoffending because we do four hours out of 20 and all the other things going on in these lives and we do, we do that in partnership with your youth workers and blah blah, blah.

He's like, yeah, I know that, but I need you to claim that you do, otherwise you can't have your money next year.

You want me to lie to you on Your forms.

Yeah, that's exactly what I need you to do.

And just at that moment, I had a vision of the council senior leadership sitting around looking at an outcomes dashboard, congratulating themselves on what a marvelous job everyone was doing when all of that data was nonsense.

It was all being created through conversations like the one I was having.

And that was the moment that I thought, do you know what, we must be able to do it better than this.

There must be different ways of doing kind of funding and performance management that were built on the idea of people telling the truth to each other.

That doesn't sound like an unreasonable belief.

Told.

So what did you do with that frustration?

So I had a rant to a friend in a pub and as all good rants are at, and that rant literally changed the direction of my working life because a friend said to me, oh, I'm.

I'm doing some work with some folks at Newcastle University business school.

I think they would be interested in that rant.

Why don't I introduce you and see what happens.

And up to that point, if you'd ever told me I would have had anything to do with a business school, I would have straight up laughed at you.

Not a chance in the world, but connecting with the folks there.

And so they said to me, why don't you become a visiting fellow and do some research on what we would later call outcomes based performance management, see what the research says about all of this stuff.

So I got into that and we, to my surprise, what I found is that all of the research that had ever been done when people had tried to use outcomes as performance management measures, like you only get your money if you only get promoted if you hit this outcome target.

All of that research said exactly the same thing, that if you try and use outcomes of performance management measures, you don't get real outcomes, you get gaming, which is, oh, the experience that I'd had was exactly the experience that all the people everywhere in the world who were under an outcomes based performance management regime, that was the universal experience.

So I want to get into that at length because I think this is one of the challenges that people face that we work in a system that requires them to track a whole bunch of KPIs that cause the gaming that you're talking about, one of the questions is, well, kind of, what's the alternative?

I mean, if you said some of the things, and I'm sure you do, that you're saying to people in the treasury or to section 151 officers and councils, it Sort of sounds a little bit like heresy, but there's an enormous evidence base behind this work.

Right, Exactly.

So it's important that you use the word heresy because that was our, the first, one of the first sets of things we did when we were back at Newcastle University Business School was set up a program called Little Heresies in Public Policy.

Right.

So we wanted to create because.

Exactly.

People thought that what I was saying was heresy, that you, what you mean you can't performance manage for outcomes like we, we have to.

That's like we care about outcomes, we have to performance management for them.

And like exposing people to the evidence about that because it feels like such an intuitive truth.

Right.

If you want outcomes, you specify, and you're a commissioner with resources to spend, you specify the outcome that you want, you put that out to tender to see who will deliver against that thing that you want.

And then you, performance manager, if you want outcomes, of course you do that.

But the, the evidence is so stark that it caused us to ask to dig into the question of why that repeatedly fails everywhere.

It does in the world, basically.

And so that's what led us to the, the world of complexity.

Because it's complexity science and the world of complexity that explains why outcomes based performance management fails every time.

Because the idea that outcomes can be delivered fundamentally misunderstands how outcomes are made in the world.

And the thing being is, if we genuinely care about outcomes, we have to understand the processes by which outcomes are made.

And so that's the thing that we say to the treasury people, to your finance office or whatever, if you care about outcomes in the world, then you have to fund and manage in a way that supports the processes that actually create them.

So let's go into a bit more detail.

So why doesn't it work and what's the evidence that proves that point?

There's many different aspects to this.

Firstly, if you can't measure accurately all of the things about an outcome that people care about.

So let's say what we care about is the outcome of people getting into work.

If I'm a person looking for work, then that outcome of being in employment has lots of different aspects for me.

Right?

I want a job that is within decent travel time for me, that gives me some kind of fulfillment, that maybe enables me to also keep up with my caring duties for my elderly mother or my sibling or whatever.

Right?

So the outcome of being in employment is a very multifaceted and personal thing, like what counts as a good outcome for me, good employment outcome for me.

Well, Maybe may well be different, what counts as a good employment outcome for you.

But it's too difficult and too labor intensive to measure all of those aspects of that outcome for me.

So people use proxy measures.

So typically the employment outcomes are measured by has that person got a job and did they keep it for six months?

So you take an outcome which is all of this stuff, and you squash it down into this little thing of what's measurable.

And what that means is that any organization that's working towards being paid to achieve that outcome is then it doesn't care about the outcome from my perspective, it cares about what's measurable.

So by forcing people to work to outcome targets, you're not actually enabling outcome from the perspective of the resident or citizen.

You're saying in advance, this is what counts in your life.

And we will only pay organizations for achieving this little thing.

So you're driving a set of perverse incentives.

So the classic example of this in the world is, do you remember the VW Dieselgate scandal?

No.

Remind me.

So this was 10, 12 years ago.

Now, this was what happens if you use a proxy measure as a performance target for car companies.

So who were trying to.

The EU and other places around the world were trying to drive down engine emissions.

And so they set a bunch of targets for reducing the emissions from car exhausts and said to the car manufacturers, you can only put cars on the road if they hit these emissions targets.

Now, it turns out it was too difficult to measure cars as they actually drove around, so they set a proxy measure, which was what happens, let's measure the emissions of cars when we put them on these rolling road rigs.

So it's at that point easy to stub something up its tailpipe and measure the engine emissions.

But because that's a proxy measure, because that's not measuring what cars are doing out on the road, what the car engineers did was go, hang on a minute, it's a bit difficult to actually reduce the engine emissions of cars as they're driving around.

That's a really hard problem to solve.

What we'll do is we'll enable our cars to recognize when they're being strapped onto a rolling road rig.

And we'll get them to change the behavior of the engines so that they reduce lower emissions, so they produce lower emissions when they're on these rolling roads.

And so you've got.

This is the Dieselgate scandal.

You've got cars that seemed to hit the outcome, the desired outcome, because what you were doing was paying engineers to achieve that proxy measure.

But actually when the cars were out on the roads farting around, putting all sorts of nasty stuff into the world.

And it wasn't until a whistleblower engineer about kind of eight years later went, actually, this isn't right, that that whole scandal was exposed and hence the car manufacturers paying millions of pounds compensation to the owners and to regulators.

So how does this pan out in public services then?

So if you're responsible for commissioning for outcomes or reporting on outcomes, I've heard you quoting some statistics in the past about the percentage of data that has sort of been massaged to the point of meaninglessness or that kind of thing.

How does this pan out in reality?

So that's only the first.

The problem of proxy indicator is it's only the first problem.

We're trying to use outcomes as performance measures.

There's another problem which gets to the heart of how outcomes are made in that an outcome isn't delivered by a program because an outcome is all is created by the hundreds of factors that in people's lives.

So example I frequently use is the outcome of obesity in people's lives.

So they were saying that at a population level, people are obese or not because of 108 different factors ranging from how the media represent body image to how food production and supply works, like macroeconomic drivers, people's early childhood and their upbring.

Only four factors out of 108 were governmental programs.

So things like number of gastric band operations in hospitals or healthy eating programs in neighborhoods, right?

So imagine now you're one of those, you're delivering a healthy eating program in a neighborhood and you're being held accountable for reducing the outcome of obesity in the 500 people you serve in your neighborhood.

You're not enough, you're not in control of enough of the factors that lead to people being obese or not to to be paid on the basis of whether you reduce obesity.

So if you hold people accountable for delivering an outcome, you're basically holding them accountable for a bunch of stuff they don't control.

Now this was exactly the experience I had as charity chief executive.

If you're any kind of competent manager, that situation is intolerable to you because you're basically throwing a bunch of random information into your management processes saying we'll work with this person and we'll maybe get paid for working with them, maybe not, depending on a bunch of stuff that's outside of our control.

So what do those managers learn to do?

They control the production of data.

Because if you're being paid by the results that you produce, you're not actually being paid for the results, you're being paid to produce good results data.

So what managers do is they game the data so it looks like they're producing good results.

So example, you cherry pick the people you work with, so you only work with the people you know will help hit the outcomes target.

It still wasn't enough to overcome the gaming system.

So the government's own evaluation said that what happened was that the employment firms cherry picked those people who were closest to the labour market, only worked with them.

So hey presto, we got these people into jobs and all the people who most needed those employability programs were left on the sidelines with hardly any intervention, because it wasn't worth the firm's while to do the difficult work of engaging with those that were furthest from the labour market.

It's really challenging, I think, because if you are a commissioner or if you are somebody providing a public service, I think this information is both completely obvious and also incredibly confronting.

It challenges the kind of superficial sense that you have that of course, if what we care about is outcomes, you should performance manage for those.

But I think it also, it helps to use all of the evidence at our disposal.

So one of the nice things is the shift towards evidence based approaches.

Then we can apply that to the process of commissioning and the process of management itself.

Let's use the evidence about what happens when people try and contract for outcomes and the evidence is 100% clear, what it creates is gaming.

So that the, the study that you were referring to before people did a systematic review of all of the research into the effects of target based performance management across 40 years of research in the private, public and voluntary sector, what that systematic review found.

So if you're into the hierarchy of evidence, systematic review being the highest form of evidence there is.

Right?

That review found that, I think it was 83% of the studies found evidence of gaming.

So deliberate manipulation of performance data to make it look good.

Nearly three quarters of those studies found evidence of people deliberately lying when asked anonymously.

I'm guessing, yeah.

The evidence is 100% clear.

If you use targets to try and force your workers, your contracted organizations into doing what you perceive to be the right thing, then you can't trust the data that is then produced.

Might be accurate data, might well not be.

And if we don't have accurate data, how does any of this get any better?

Yeah, we can't learn from data that has been gamed.

And if you can't trust whether your data is gamed or not, your learning processes disappear.

We'll link the study in the, in the show notes so that people can, can have a read of this for themselves.

The, the question that, that follows I suppose is well, what's the alternative?

Depends how far we want to go back with this.

So one of the things that I think one of the stories that's important to tell around this is where did the idea that the purpose of management is to set targets and monitor people's performance against those, where did that idea of management come from, particularly in a public service context?

So it's important to tell the story of James Buchanan as part of this and if you find this too dull, we can edit it out.

And as a libertarian he believed that it was illegitimate for governments to tax one person and to spend those resources on somebody else.

So it was his personal life mission to delegitimize the welfare state.

So that was what he was about.

Interestingly, he was one of the Koch brothers first investments.

And so what James Buchanan created was a way of, a way of framing public service called public choice theory.

What public choice theory said was that public servants, like everyone else, are rational, self interested utility maximisers.

What does that mean?

Utility maximizers.

So public servants like everyone else are selfish.

So if you want public, if you want public servants to pursue the public good, so to do good in the world rather than just serve themselves, you have to control them.

So you have to force public servants to act in the public good by subjecting them either to the discipline of the market or by controlling them via targets.

So you need to express the public good in terms of quantifiable targets and then you need to monitor people's performance against that to force them to work in the public good.

What he was trying to do with this because he recognized that the way to delegitimize the welfare state was to break trust between those who were governed and those who are doing the governing.

And so by asserting this idea that all public servants are selfish and can't be trusted, it was part of this idea of breaking trust.

So he was deliberately trying to break those as a set of relationships.

So one of the important things I think is useful to understand is when we set targets for performance, what we're doing is forcing public service to live in James Buchanan's world, okay?

It is entirely based on the assumption that people are selfish and have to be forced to work in the public good.

Now thing about that is none of the evidence around public service motivation accords with that particular perspective.

So the evidence around why people turn up for work, and particularly why people turn up for public service work is around mastery, autonomy and purpose.

So people will turn up to do a good job if you enable them to have mastery, to get good at what they do, to have autonomy, to have a good amount of decision making rights about how they do what they do, and to have purpose, to connect what they're doing to something bigger in the world.

And so if we ditch the, let's be generous and call it at best unproven assertion that people have to be controlled in order to make them work in the public good, then what that gives us is a whole different management landscape to work in.

Because basically it transforms the task of managers to say what would it.

It's my job to create an effective environment in which each, which.

In which workers have mastery, autonomy and purpose.

And I can.

My role as a manager is to organize people's work so that they can do the best jobs they possibly can.

So let's talk about some of those examples.

I think we need to go back a few steps and talk about human learning systems because I think where we're headed in this conversation is, well, James Buchanan's thinking has then fed into new public management.

And new public management is a paradigm of the sort of governs the way that public services are organized and managed and controlled.

And what you're talking about is that we need a new paradigm.

And you've been a sort of steward for the work globally to collect the, I suppose, the action research of that work.

So can you talk about human learning systems?

Why is that?

Exactly.

So again, thank you for using the word paradigm because it was interesting when I started out doing this work.

People, we're kind of trying to do this like off stage.

We don't use the word Paradigm, it's just, it's too confusing.

Blah blah.

Actually, I think the word paradigm is important because a paradigm is, means a kind of mutually dependent set of ideas, beliefs and practices.

If we reject the idea that people have to be controlled in order to make them work effectively, then it enables us to change the purpose and focus of management.

So human learning systems is a new, an alternative paradigm for how we organize, fund and govern public service.

So everything from how we conceive of what the public facing roles of public service are to what, what, how team leaders, how senior leaders, heads of service, blah blah blah, how those people frame the work of organizing public service and crucially how public service is funded and governed.

So what's the resource bargain that public service officer I will give you money if so we change all of those things.

And crucially the purpose of changing all of those things is to enable the creation of positive outcomes in people's lives.

Because all of this comes back to an understanding of how outcomes are actually created in people's lives.

So once we get the idea that outcomes are emergent properties of complex systems.

Jargon, jargon, by that, all that we mean by that is the outcomes are co created between residents and all of the people who work with them and all of the other factors that are going on in that person's life.

So all of the things in people's life contribute to creating an outcome, either a desirable outcome or a really poor outcome in people's lives.

So when we understand that outcomes are like an outcome at any given time is the product of all of those different factors and people and organizations all combining together, we realized the idea that trying to control workers to respond to this complex, ever changing real landscape of people's lives.

If you.

So I was thinking about this on the way over.

What do people say they really want out of public service and the people they interact with?

They often say I want people to who go the extra mile for me or I want, I want people who go the extra mile for my gran.

Is it possible to specify what going the extra mile looks like for my gran or your grand or.

Absolutely not.

So if you're, how would you control someone to go the extra mile?

So the thing that we most want from our public servants like there, I want you to respond uniquely to my life circumstances or my gran's life circumstances and really do like work to make it to make that person's life better.

Uniquely, you can't control people into that.

It's literally impossible.

And everything, every time we try, we, we make Public service worse and more expensive.

So what does the essence of a relationship that enables outcomes to be created look like?

It's one in which a public facing worker and a resident, or groups of residents, communities, whatever, can work together and explore the unique circumstances of that person's life.

What's going on for me and my life around employment?

What are the factors that are preventing me from getting a job right now?

What are the things going on in my life that would enable, that I can build on, that could enable me to get meaningful work for me.

And let's experiment together.

What works uniquely in your life?

Let's try this.

And let's try this.

Oh, that seems to have worked over there.

Why don't we try this in your life?

So giving workers the freedom in Mark Smith's language, liberating them to work together and explore, to have a learning relationship with the people they serve, that's how outcomes are made.

So if we really, really care about outcomes, our public facing relationships have to be those learning relationships where people get to explore together with residents or groups of residents or families, whatever, what works uniquely in each person's life.

So that's the star of a human learning systems approach, is reframing the act of public service out of this language of delivery.

So if there's one thing we can do to really create, to, to, to ensure that we create outcomes in the world, let's ditch the language of delivery.

And why do you take so much issue with that word specifically?

Well, think about what the act of delivery is.

That's me giving you a prepackaged something.

Understood?

Yeah.

Right.

So when someone knocks on your door and gives you a parcel, they're delivering something to you.

Right?

That is delivery in public service world.

It means, if we think about delivery, it means someone has specified a program of work to be undertaken and the person responsible for that, for that will deliver the 10 workshops.

Right.

So what you're entitled to is six workshops of cognitive behavioural therapy.

Therefore I will deliver six workshops to you.

Yeah, yeah, understood.

And I think part of the problem with that language finding its way into public services has perhaps been accentuated a little bit by the success of the government digital service sort of 10, 15 years ago, not quite 15 years ago.

What are we talking about 13 years ago?

Because the language of people that manage delivery has become very present in the creation of new digital services.

But those digital services are often very linear and for a reason.

I mean, sometimes we trip ourselves up when we make them too linear.

Because, you know, if somebody is, if Somebody, for instance, is getting in touch with a council to pay a parking fine.

Well, maybe it is a good idea to optimize that process.

But what if they can't pay the fine?

Well, that's another thing entirely.

But where I was headed with that is that language and the description of roles has then found its way, I think, into much more relational services that we're talking about here.

And it has a very unhelpful.

It plays a very unhelpful role, I think.

Exactly.

And like one of the.

When I said before, I kind of cut me open.

There's a conceptual framework.

So one of the things that I try and do is to get people to think about whether different ideas fit together well.

Yeah, because if the ideas don't fit together well, then the practices won't fit together well.

So it's a way of kind of shortcutting.

Like we can avoid making a whole bunch of certain types of mistakes by seeing whether the ideas work together properly or not.

Yeah, well, then we get to make a difference out of exciting mistakes.

But so we can avoid making really stupid mistakes by seeing whether the ideas fit together.

So do the ideas of co production and delivery fit together well?

Again, you can't.

If we want to take the idea of co production seriously, that's like literally a generative process of co creation between a public service worker and resident or residents.

Right.

And that's not the public service worker delivering stuff to people.

So we have known for what, 10, 15 years that co production, co creation is important.

And yet you will hear people talking about delivery in the same sentences, in like the same paragraphs as importance of co production.

These ideas do fundamentally do not fit together.

What are we doing thinking about that?

Like trying to mash that stuff together in ways that, that clearly doesn't work.

The human learning systems is founded on this idea that public facing relationships need to be learning relationships.

So that every act of public service is an act of shared learning between workers and residents.

And if that's the case, it means that we need to manage public, those public services servants differently.

Because you can't control people to learn well together with people.

Right.

So you have to, what you have to do as a manager is to create the conditions where people can learn well together.

And you have to, so you have to organize your team meetings as learning environments where the workers are saying, okay, these are the things that we've been trying with this resident this week.

This is what we're learning from them.

So because if you frame every act of public service as an act of shared learning, what you're doing is creating really great data around every single act.

So let's try this and see what happens.

If you frame that rigorously, then what you do, you're recording what it is you're trying and you're recording what happens when you try that.

And then so that gives you amazing data to learn from.

So if you're treating your team meetings as debriefing and pattern spotting, then when the picture you're building up as a leader is, okay, these are the patterns of need amongst the people that we're serving, These are the things that we're trying, these are the things that seem to work, these are the things that seem not to work.

So how do we organize ourselves to make it easier and more efficient to to do all of the things that work?

And what are the practices then that you're seeing where people are doing that effectively?

So again, referencing a bunch of the liberated method stuff that Mark Smith and others helped to develop, that practice of journaling for public facing workers.

So after the end of every encounter, they're recording, this is what we did, this is, this is what we tried, this is what worked and what didn't.

And then because they've got all that amazing data, then they can at team meeting level, at directorate level, what they're doing is spotting patterns from all of those and asking then crucially, when those patterns throw up challenges to business as usual for those organizations, normally if the work throws up challenges to business as usual to like your PDR practice, our perform, our annual appraisal process is getting in the way of effective learning.

Okay, let's change the way that we do our annual appraisal process.

Let's experiment with doing that differently.

So because you're learning from the work upwards, what that means is that all of your organizational practices are up for grabs.

Yeah, because again, if we care about outcomes, can we organize work in way that enables the outcomes to be produced?

And if that means we need to change the way that our organizations do annual appraisals, if that means we need to change how we fund organizations, that's the work that is required of us as senior leaders.

So it turns the practice of senior leaders into continuous learning as well.

So one of the things we're seeing councils do, so working with Norfolk Council and Doncaster Council on exactly this stuff, is changing how they do governance and accountability practices.

So wherever accountability practices show up, say from team meetings to individual 1 to ones to how scrutiny committees work, changing those accountability practices to essentially become Asking the questions, what have you learnt?

How have you learnt it?

And how have you adapted as a result?

So if those become your accountability questions, then what you're doing is ensuring that there is rigour in all of those learning practices.

So have you been really learning off the basis of the evidence that you're creating?

Have you involved all of the people that you should have been in those learning practices?

Did you?

How can we ensure that the conclusions that you drew from all of that were valid?

Did you test them?

With this range of people, blah, blah, blah.

Those become your accountability practices.

And so you're getting a form of accountability that works in a complex environment.

And crucially, your accountability practices support the creation of outcomes in the world.

If you're still here, my assumption is you're finding this conversation useful.

We're recording these interviews because we think it's important to share stories of people who are leading and designing more human impactful and relational public services.

And we'd like as many people as possible to see them.

If you think it can help us with that, then I'd be so grateful if you did one of three things.

You can just like the video and that'll help other people to find it in future.

You can leave us a comment and let us know why you stuck around, or you could subscribe to the channel.

And honestly, I've got no good reason for that other than it would cheer me up.

Okay, back to the interview.

To sort of give you a bit of a contrarian question here.

What's to stop?

If we accept that the outcome measures get gamed and KPIs get gamed, what's to stop people from just gaming the sort of act of learning so that they can then get access to more funding, more money, more resources.

So if you try and treat the accountability questions in the same way that you would currently do performance management as elements of control.

Yeah.

Then the chances of people gaming their learning stuff increases.

Because gaming isn't about what you measure.

Gaming is about why you are measuring.

So again, one of the things that I uncovered as the.

As you're kind of doing the broader research around this stuff and is if I have like one wish, one simple wish for kind of public service, I would put Campbell's Law, a printout of Campbell's Law, on the desk of every public servant in the country.

He wrote a very famous essay called Assessing the Impact of Planned Social Change.

And in it he created what became known as Campbell Law See if I can do this.

Any quantitative indicator used for the purpose of social decision making will tend to distort and corrupt the processes it is intended to monitor.

It's really similar to Goodhart's Law Child's Goodheart President the Bank of England in the 70s any measure that becomes a target ceases to be a good measure.

So it's not what you measure that determines whether you've got gaming, it's why you're measuring.

So if you're measuring for the purpose of continuous improvement, then you will.

It gives you the best chance that you can of avoiding Campbell's Law.

Because what you're trying to do with those accountability practices is not.

Is less about holding people to account because that's still the kind of control mechanisms occasionally you will, you'll get the occasional bad egg, blah blah blah that you need to hold to account.

But mostly it's the accountability practice are about helping people to account for their work.

Because if we treat those accountability practices as a rigorous process of continuous reflection and improvement, then having someone else ask questions help us to ask questions about our work improves our work.

So if we, if we, if as leaders we're creating the expectation that basically everyone's job is to get better at our job every day.

And that's an expectation that leaders can set.

If you're working here, your job is to be better at what you do this time next year than you are now.

And that's not something that we can control you into.

That is the expectation of everyone around here.

So if that's the culture that you set as an organization, then your accountability practices become part of how do I get better at what I do?

And if you're creating those kind of expectations, then if you game the data required for learning, then the person you're cheating is yourself.

Yeah.

The other thing, the other way of getting.

Ensuring that the best chance of people operating the public good is to make those accountability relationships to residents and citizens strong.

What do you mean by that?

Currently our accountability relationships are upwards.

Yes, you're accountable to your line manager in the hierarchy.

We've got this accountability chain where people are always looking upwards to be told am I doing the right thing or not?

One of the other things we can do to make accountability actually work is to change the directions of accountability.

So as well as because a bunch of organizations will keep the kind of hierarchy that they have because it's right for them, we can also spread accountability to make a horizontal accountability between peers because it's much harder to Pull the will over your eyes of people who do your job, who are similar to you and do your job in a similar way.

They'll know if you're pulling a fast one, people.

So creating accountability relationships with peers helps, but also creating accountability relationships to the people that you serve.

If you're in a learning relationship with the people you serve, what accounts do you owe them about why you're doing what you're doing and who do they get to complain to if they feel that that learning relationship isn't working for them?

So creating those relationships where the people being served go, hang on a minute, this isn't working for me.

I need something different that's really important as well.

So if you can't deliver or create outcomes, you know, by, you know, commissioning an organization or one team to do that thing, how do you know what you're doing is useful and helpful in society?

So this, this is where the idea of continuous learning comes in.

Right?

Because if every act of public service is an act of shared learning between a public facing worker and a resident, or between groups of people, then what you've got is a whole bunch of amazing data to understand what's happening when you try, when you try this and what's happening when you try that.

So we will have people who do this have amazing data about what's working for them and what's not.

But what they do is they use that data for continuous improvement and adaptation.

So Plymouth, for example, they created a couple of years ago a compassionate approach to.

Wait, they've used those processes of continuous learning to understand how the work that they're doing affects the lives of the people they're working with.

So they've got, they use appreciative inquiry and ripple effects mapping to understand the work that they're doing.

So they're gathering a whole bunch of amazing data about how people's lives are changing and not, but then they're not just not using that data for performance management because they have created a culture and environment where people don't need to be controlled to force them to do the right thing.

They are creating an environment in which all of the workers in that as a system that they've created are desperate to be better at what they do.

So they're hungry for that data and to understand how their own performance can be better.

Yeah, I think Gary Wallace in Plymouth is very open with his learning and he's very happy to share that with other people that are interested in the actual practice of ripple effect mapping and.

Appreciative Inquiry and just some of the data that they've gathered around the work.

So a different work in Plymouth, the work with adults with complex needs.

Personally I hate that expression because to be human is to have complex needs.

But you know, is that policy thing for policy speak for people who experience homelessness, substance use problems, mental health problems, involvement in the criminal justice system, that stuff.

They did a comparison of avoidable deaths for the population of adults with complex needs who'd had encountered their human learning systems work versus adults with complex needs who hadn't.

In the 10 years of this work they've been doing, the population, the adults with complex needs who hadn't been involved in the human learning systems approach, there was something like a thousand avoidable deaths for adults with complex needs in the city over 10 years.

For the population that had been involved in the work, that equivalent figure was 7 over 10 years.

Yeah, it's really, really astonishing and I think this is why the work that Gary and his colleagues are doing in Plymouth and the work that Mark and Ron and co have done in Change of Futures Northumbria is so powerful.

I mean I was very struck with Ronsworth when I talked to him recently.

Just the tenacity that it took to actually track down that historical information about, you know, the impact of the previous ways of working versus the new.

And I think that is part of the story that we've got to tell.

That's very, very hard work to do, isn't it?

But it's necessary.

A bunch of my presentations start with kind of critiquing that target based approach and some people hear that as oh, in a human learning systems approach, data isn't important because in people's heads the only reason to, to collect data is for targets.

Yeah.

So if you don't like targets, you must not like data.

Listen to yourself when you're saying that what, what has that, what has James Buchanan done to you?

In a human systems approach, data is more important, not less.

Because without data you can't understand, you can't learn, you can't.

If you don't, if you're not capturing data about the work that you're doing on a day to day basis, how are you and others going to learn from that?

So what we're doing, saying, and this is the joy of it in some respects, in that people who start to work at a human learning systems way have so much brilliant data about what's going on in the work that they can tell amazing stories about what's happening.

So exactly the stories that Ron and Mark are able to tell about changing futures.

Northumbria, the story about Brian is a fantastic example of that.

The avoidable deaths data from Plymouth and a bunch of the other amazing stuff they have.

If you start working in this way, if you genuinely start with seeking to understand people's lives and what happens when you try stuff with them, then you get brilliant data to tell different stories about what's going on in the world.

Yeah.

And the fact that data isn't easily shared across organizations and systems, that is a reality.

But it's also the work, isn't it, to try and figure that out.

And one of the really nice things is that we can do something about the information systems problem now because there are developers that people like Ron and Mark have been working with and people at my colleagues at Manchester Met University have been working with who have built information and communication systems platforms that enable those collaborative learning processes and enable data to be shared across organizational boundaries in GDPR friendly ways.

So that that data and communication systems problems, we are on the brink of that being a solvable problem.

Yeah.

And I would encourage folks that are interested in this, in this challenge and interested in this movement towards a more human learning system, towards more relational public services, to come to the Relational Public Services Conference in Manchester.

And we'll link the details of that again in the show notes.

Great plug.

If you want to.

Great plug.

It was a good plug, wasn't it?

Yeah.

But I mean, if you want to come across people that are actually doing this stuff and have figured out all of these problems.

Not all of these problems.

That's disingenuous.

Have figured out some of these problems that seem completely intractable.

You'll be surprised that there are more solutions than you think.

Exactly.

And that's the beauty of those kind of spaces, is that no one has all the answers to these problems because every time you solve a problem, a new one appears.

That's just the nature of being in a complex world.

But in those kind of spaces, you get to be around the folks who can help you solve the problems you have right now, because they are all the people who are experimenting with this stuff in different ways.

And so being able to access each other's learning around this is what enables us to avoid having to make the same mistakes that Mark did and Ron did when they started out five years ago.

You can shortcut a bunch of that stuff and you get to make new and interesting mistakes all of your own.

Exactly.

Lots of people that are interested in this work come from a range of different methodological Backgrounds.

So myself and Dennis and colleagues at Basis, we have been experimenting with more agile ways of working for long periods of time.

And agile is just a way of accumulating learning very, very quickly about what's working, what's not, and then helping us to reprioritize what to do next.

Vanguard method, lots of people that are working in this way have come from that background.

Obviously liberated method is very prominent in public services at the moment.

How do you see these ways of thinking, I suppose in the case of Agile or methods, in the case of Vanguard, how do these things link to human learning systems?

So all of those methods, people have used all of the methods that you've just described within human learning systems approach.

Because human learning systems is just a name that we gave to the learning, the collective learning that people have done about what an alternative public management paradigm looks like.

So human learning systems exist at quite a high level of abstraction.

Abstraction because it's all of the learning that people have done about what public facing work looks like differently, about what the task of operational management looks like done differently, what the task of commissioning looks like done differently, blah, blah, blah.

So it's as you say, people have used all sorts of different methods and approaches as part of those things.

And what we're trying to, all that we're trying to do with the human learning system stuff is to encourage and enable people to capture their experiences of doing that in different public service environments and then enable people who want to free themselves from the James Buchanan nightmare to piece all of those things together and say, what does this add up to in terms of this alternative paradigm?

A different, a fun.

What does this mean about the.

Our fundamental beliefs about why people turn up at work to do a good job about how outcomes are created.

So we are trying to create that internal coherence and alignment that made new public management so powerful.

Because the reason, one of the reasons that new public management has been as successful and tenacious as it is is that it has really strong internal coherence.

Yeah.

So the reason that people commission in the way that they do connects to the fundamental beliefs about how people do performance management in the way that they do.

The, the way that people run team meetings, the way that people run PDRs, the way that heads of service, all of it is connected in a internally aligned, coherent paradigm.

Say more about that.

So how does one thing lead to another, to another next?

Well, we commission in the way that we do, so we set targets and we monitor the performance of organizations against those because it's connected to the fundamental belief that if you just grant those organizations money and allow them to do their own thing, they'll use it for their own selfish purposes.

Right.

The directors will go off and have holidays at the Bahamas and blah, blah, blah.

Right.

So the.

The reason that we moved from the situation 35, 40 years ago, when most local authorities would fund the voluntary organizations in their patch via.

By grants and via grants for core funding.

So it wasn't grants.

We will give you this grant if you will work with 15 people in this geographical patch of law.

The reason that I moved from that granting basis to contracting and performance management was the belief took hold that you can't leave those organizations to do what they do because they will waste that money doing what they think is important, not what we think is important.

So it all fundamentally comes down to the idea of control.

Yeah.

So all of that new public management is essentially comes down to the idea that we have to control people in order to make them do the right thing.

And so providing an alternative that has the same level of internal coherence and enables people to find answers to all of the different elements of organizing, funding, and governing public service.

That's what we're trying to do with the kind of human learning system stuff, because it needs to be as comprehensive as the thing that it's trying to replace.

Because it's no good saying, stop doing all of that.

Oh, people ask.

People legitimately ask, okay, what should I do?

What should I do about how to run our scrutiny committee differently?

What should I do?

What if I'm ahead of commissioning?

What should my commissioners do differently?

It's no good saying, you need to stop doing all of that stuff if you haven't got some principles that help people explore how to be different about that stuff.

Because the terror of the blank page is real.

Yeah.

I mean, let's talk about commissioners briefly.

We're coming to the end of this interview, and there's a few things I want to get to before I send you anyway, because I know you've got to get back to the university.

But if you're a commissioner, what's a step that you can take at the top level?

The thing that you can do is to commission organizations to learn collaboratively together, so to literally change what it is you're funding them for so you agree a purpose.

And like, what we're trying to achieve with this resource is we want more people back into work, or we want people to lead thriving lives, or, like, we want older people in this part of the city to lead thriving Lives.

Brilliant.

Okay.

That's a good purpose kind of statement.

And.

Here, Organization X, you can have these resources if you'll learn with the people in this, in that part of the city how to, how best to achieve that.

And you can.

So that's like at its simplest form, you can literally change the nature of what you are contracting for.

You are contracting for organizations who will learn alongside the citizens, residents they're trying to support.

And that's so that when Plymouth did that with the Plymouth alliance, the Adults with Complex needs program.

So 80 million pounds of public money, 10 year contract, not a single key performance indicator anywhere in that contract.

So we know it's possible.

So people want to do this.

They just can.

It's simply a choice that people get to make.

What they did was a slightly more sophisticated version of that.

They said and we will commission organizations collectively to learn collaboratively about how to enable adults with complex needs in the city to lead thriving lives.

So they don't commission individual organizations.

They commissioned an alliance of organizations to work together to learn collaboratively in respect of that purpose.

Interestingly, one of the things that they did was they said any organization can be part of this alliance, but you have to practice complete financial transparency.

You have to open up your books to all of the other organizations in the alliance if you want to be part of this.

Wow.

And a bunch of organizations said yes, brilliant, we're very happy to do that.

And some of them said no.

And so there's an interesting self sorting mechanism there about who.

About levels of financial trust.

Yeah.

Well these kinds of examples of transparency, they play out in Frederic Leloux's work in his book around teal organizations and more self management.

And transparency in the way that you're talking about is one of the things that makes it possible for people to self organize.

You know?

Exactly.

And then sharing the data that you get from the learning from.

If you, if you are reorganizing the work as learning and you say what we're demanding from you is that you share your work with others freely.

Yeah.

Then that's what I mean by kind of changing the nature of the resource bargain.

You're contracting full collaborative learning.

You're not contracting to deliver a service specification.

Yeah.

And this is a process that people are going to have to learn for themselves.

Can work.

There's no ranting that you and I can do that is going to convince people that they will be able to do this in their areas.

They're only going to be able to figure this out by having a go in practice.

Exactly.

And the messiness of doing it in practice is exactly part of the learning.

So I remember Liverpool Combined Authority were one of the first sets of people to look at what Plymouth were doing and thinking we could change how we commissioned our bunch of our homelessness states.

So we worked with a bunch of commissioners there to help them move to a practice around funding for learning.

And the start of that is to build a different type of relationship between commissioners and providers because they had come from very transactional relationships.

So they're the commissioners brought the providers together and said we would like a different type of relationship with you.

We want to commission you on a trust basis where our relationship is about shared learning and the providers turned around and said we don't believe you.

So I remember I was sitting at my desk in the previous office and I got the phone call from the Liverpool people going, we said that we wanted a trust relationship with them and they said they didn't believe us.

What should we do now?

I was like, I have no idea.

That's your relationships.

So what they did, so they were trapped in that kind of catch 22 of it was such a low trust environment that they couldn't, felt like they couldn't switch to a higher trust environment.

So what they did was they changed how they ran their existing contract management meetings because previously their contract management meetings with a bunch of the kind of homeless and providers had been putting performance data on the table and beating the people over, beating the organizations over the head with it.

You said there was going to be 15 of those, there's only 10.

Bang, bang, bang, bang.

Instead they moved to a situation where they put the performance data on the table and said we want to treat this as an opportunity for shared sense making.

Here's the performance data that you've gathered.

What does it mean?

What's going on here?

Why are those.

And we want to hear your experiences of what it's doing, of what it's like to do this work.

Yeah.

And after about two or three of those, the providers were like, oh, I see.

What, yeah, okay, you really mean it about you want a different kind of relationship with us.

And that then enabled them to switch to the commissioning for learning and blah, blah, blah.

So it's actually commissioners doing the work to create trust based learning relationships.

What that looks like and what that requires is going to be different for each set of commissioners and that's the set of experiments that commissioners can do.

Right.

What does it take for me to build a set of learning relationships and treating that as an experiment.

Okay.

We're going to try this type of meeting and gathering some data about what happened when I tried this type of meeting.

What was the effect on the relationship with it?

That's the learning experiments that commissioners get to do.

Yeah.

And there are some fundamental things I think that people can start with.

If you're going to try out these different relationships and arrangements, probably don't do it in front of 100 people all in the same room at the same time where everybody has to engage in face saving behavior.

Think about psychological safety.

How do I start to build that so that people feel like they can stick their neck out and say, oh, I don't think this is right, or I don't think this is working.

Exactly.

So this is the range of stuff that I'm much less good at thinking about.

Like how actual human beings have relationships of trust.

And I design help design concepts and processes.

Right.

I get confused when people don't do the things I said they were going to do.

So it's important.

But this is the human bit in human learning systems.

Right, exactly.

That the language of psychological safety is really important.

And like what are the conditions under which human beings tell each other the truth about stuff?

Yeah, right.

All of that stuff is knowable.

And there are a set of practices exactly as you're saying that enable that to happen.

Those are the things that commissioners and other stewards need to become familiar with because that you need to be able to create the conditions where a diverse set of actors are going to be able to tell each other the truth about what their respective lives are like.

So if you're a, let's say you're a team leader in a community facing service, or you're a public manager, you're a commissioner, you're a leader of a council, whatever it is.

And you think, hang on a minute.

I think we're doing some of this work.

We haven't got all the answers, but we've got maybe some practices, some learning that might be useful to other people.

What's your ask of those people?

Capture what you're doing as experiments.

So the work that you're doing to understand the lives of the other actors in the systems that you created, capture that in a way that that can be shared with others.

What did you learn when you ran, when you tried to bring 20 people together as opposed to 100 people together?

What did you learn when you brought providers together separately from other commissioners?

Literally thinking about that work as experimentation, capturing that and sharing it within your context, but also sharing it out with others who can learn from the stuff that you're doing.

If everyone treats their work as continuous collaborative learning, what would that require of us?

It requires of us to think about what we've done, capture it in some way.

Whether that's writing it down or like doing a audio note to yourself or doing kind of weird things like this, and then find a way to in for that information to be shared with others.

And in the end, this is what the paradigm when, when everyone is doing that kind of practice.

This is when we can tell that paradigm shift is happening.

Because our organizational processes, our information systems, our management systems should be asking us these very questions.

What have you learned?

How have you learned it?

How have you adapted as a result?

If that is the questions we are all.

Those are the questions we are all asking each other all of the time.

Then suddenly that's what we demand from our information systems.

If our information systems make it easy for us to capture.

What have I learned?

How have I learned it?

What have I changed as a result?

Then it makes the sharing of that stuff piece of cake.

So I think we're coming to the end of this and I think that's a really good place to start our wrapping up.

I've got some quick questions for you for those.

Okay.

You dj?

I do.

I'm a hobbyist dj.

Okay.

What is a track that never fails to get people on the dance floor and moving?

My current favorite real banger is a track called Bayama, which is a kind of like a afrobeat house music thing.

Yeah.

Gets a room full of people jumping significantly.

That's amazing.

I'm a huge afrobeat fan.

Book.

What's one book you find yourself either gifting or recommending to people more often than not?

The academic book that I point people at most frequently.

Complexity and the Social Sciences.

The State of the Art by David Byrne and Margaret Callahan.

It's, it's, it's not the easiest of reads, but if you're interested in how complexity applies to all of this work, it's one of the most comprehensive things.

Quotes.

Got a quote that you try and live your life by?

I'm sure I have, but I can't bring one to mind right now.

There we go.

That's a perfect place to end it.

Hey, Toby.

This has been such a joyful conversation for me and I think we've really got into the weeds of some of the things that people are wrestling with every day.

I really think the work that you're doing as a steward for this human learning systems work is, is unbelievably important.

And the amount of times that your name comes up in conversation in the day to day course of me doing my work in the world, trying to engage with these messy challenges is astonishing.

So clearly the message is getting out there.

And I really am grateful for you doing this podcast today.

It's been an absolute pleasure.

Actually.

I've now got.

You've got a quote.

Yeah.

Quote I was introduced to recently.

It's all about who you choose to go dancing with.

That's perfect.

Let's leave it there.

Thanks, Toby.