The road ahead for the ONS: a conversation with Darren Tierney

Jan 27, 2026

Ask episode

Chapters

Transcript

Episode notes

Late last summer, Darren Tierney stepped into the newly created role of Permanent Secretary at a pivotal moment for the organisation. In a new episode of Statistically Speaking our host Miles Fletcher sits down with Darren to talk about what happens next.

TRANSCRIPT

MILES FLETCHER

Hello and welcome to another episode of Statistically Speaking, the official podcast of the Office for National Statistics. I'm Miles Fletcher.

Now, if you've followed the UK news much over the past year, you'll know the ONS has been under more scrutiny than at any point in its history, questions about trust, priorities, its core economic data and whether the organisation has simply been trying to do too much at once, have all been very publicly aired.

Today's episode is about what happens next. Late last summer, Darren Tierney stepped into the newly created role of Permanent Secretary at a moment when some of the UK's most relied upon statistics, especially those based on the Labor Force Survey, were under real pressure. In this conversation, we'll talk about Darren's background in government, running large, complex public services and helping to craft and enact key policies. We'll explore how data has shaped his career so far, and what he found when he arrived at the ONS. We'll also dig into some of the hardest questions facing the organisation, and what absolutely has to be fixed first, what may need to stop, how user expectations are being balanced with that and how full confidence in official statistics can be restored. This isn't just an internal reset. What the ONS does and how well it does it matters to policymakers, businesses and the public alike. So join us as we talk about focus, priorities, and the road ahead.

So then Darren, to begin at the beginning, you're not an economist or statistician by profession yourself. What motivated you to take leadership of what is the central organisation in UK statistics?

DARREN TIERNEY

Sure, well look as a non-statistician you can imagine I was a little bit daunted when this prospect first came along for precisely that reason. I was kind of worried that maybe not being a statistician or an economist might be a real drawback but as I thought more about the job, and more about what was needed for the organisation at the time, and with the split of the roles, I thought actually, I can really help. But what really drew me to the role was that although I'm not a statistician, Iv'e been an avid consumer of ONS products for the 25 years that I've been a policy wonk in Whitehall. And going from a good policy professional to a great one, requires both access to good data and also a real understanding of it and knowing how to use it. And it's that background that kind of helped me and really got me into it, and so I saw it as a real opportunity to come and help an organisation that really couldn't be any more consequential for the UK.

MILES FLETCHER

The Devereux review said that what was needed was someone with really serious operational experience, having run serious civil service operations, and you bring plenty of that. Could you just detail some of that? You've had a very strong rise, you know, through the ranks of the civil service. Just trace that for us, if you would.

Yeah. So I started life in what was the old Prescott empire of DETR, which was environment and transport in the regions as a policy wonk working on aviation. I then went and joined the Ministry of Justice, where I spent three years in the private offices of Harriet Harman, Jack Straw and Ken Clarke, and that kind of real frontline policy making, as it were, is the thing that I did a lot of in the first half of my career.

The second half of my career, which is much more relevant to the ONS, was centred around transformation and change. And again, I spent about 12 years or so in the MOJ, and my senior career there was about transforming systems and running big programs. And I went from there to international trade, just after the Referendum, where like a lot of officials at the time, I was quite keen to run to where the energy or the fire was after Brexit. But I didn't fancy working DExEU, and this new department called Trade seemed quite interesting. So I went there, and again, this was a kind of brand new department, and it grew from a few hundred people to about four and a half thousand. And so that kind of building and running the department as a strategy director for investments and exports, I guess those two experiences in justice and trade are the two that I rely on most in this role.

So you spent time in the corridors of power then, the ministerial heights of the civil service, but also having to try and make the machine run efficiently as well, and to deliver public services and achieve policies in the real world. What role has data played in that? And what sort of statistics were you using? What sort of insights were you getting? Can you share some experiences of how that's shaped your career, and your understanding, particularly of the job in hand here at the ONS.

Sure, I guess the job where it was most relevant was when I was in justice, where I worked in civil and family policy, Legal Aid and supporting on prison reform, and particularly on youth justice reform, and maybe that's a good example to pull out. Justice is one of those areas where people have very strong views about what the right policy prescription is, but there's often quite a lack of evidence that goes behind some of these strong policy positions, if I can put it like that, and one of the things that bedevilled us was trying to turn very strong ministerial desire for action across the human justice system into something that was evidence based, and where we could point to proof that these things would work, and we did that several times in the youth justice system.

Perhaps the one that I look back on most fondly in the youth estate...We were constantly telling ourselves, and indeed the world, that kids in the youth estate were getting thirty hours a week of education, and that was true in the sense that that's what we were buying. We were procuring thirty hours a week from providers, but the kids in the estate were not getting thirty hours a week and we didn't really know until we sent in a team of researchers from the MOJ analytical team to do some surveys, and then we also analysed some of the ROI coming out of that estate. And of course, what we discovered was that although we were buying thirty hours of education, it was always being disrupted by all the other services that these kids needed to access. So every dentist, every doctor, every lawyer that was coming in to interact with those kids, that was happening during what would otherwise be the school day, but we didn't really know that. And for years, we were kind of running this system and telling ourselves and indeed telling the world, that this was what was going on, but having analyzed that system and got ourselves some real data we were able to change that system so that it was education first, and all of those other ancillary services would take place outside of the school day. So sometimes it can be quite basic bits of analysis that can result in the biggest changes.

I guess the lesson there is, we always need to never totally trust the data we've got and find better insights and better ways of looking at things. And I guess that leads us into the challenge here at the ONS. What were your first impressions when you took the plunge in, suffice to say, quite interesting circumstances in the earlier part of 2025. What did you make of your earliest days here at the ONS and formulating what needed to be done?

Your characterization of it is right. Before I even arrived, I was talking to a lot of the stakeholders, including Robert Devereux and others, and in doing that and in reading around the issues, I ended up getting quite a negative view of what was going on, perhaps understandably, because that was what was being reported. So I was expecting an organization that was perhaps much more kind of browbeaten than the one I found. And that's not to sound complacent at all, but the thing that I was surprised at when I arrived was just the desire for progress, for action, for us to kind of turn the page, and for us to get back to being a respected NSI. And that that sort of energy that I found amongst the senior team was great because I was expecting to have to do quite a lot of jigging up of the system to get us into that mode, but actually, when I arrived, people were already there. So that was my first impression. My second impression really was just the quality of the people here, the skills, the experience and the expertise here at the ONS is genuinely inspiring. And I said this to Civil Service World a few weeks ago. We abuse the phrase "world class" in Whitehall, we often claim things are world class when they're not, but we do have some genuine world class expertise in the ONS and I was genuinely inspired when getting to meet some of those people during the first few days and weeks.

It sounds as though people had a good idea of what needed to be done, what was coming out in the media, and of the various inquiries that took place last year, was that the general feeling was that ONS had become spread too thin, and needed to prioritize on those things that ONS is best known for. It does have a worldwide reputation for, crucially, what people most rely upon it to get it right, prioritizing those. Was that what emerged?

It really was. And that sense that getting back to the core business of an NSI (National Statistical Institute) was what people were people were really hungry for us to do. I think there is a danger when saying something like that, of sounding like no one valued that hugely innovative work that happened during the pandemic. And that's definitely not the case. I think the reputation of the ONS developed during the pandemic for innovation, for fleet footedness, and of doing some brilliant work in support of the country at a time when it was most needed, is genuinely well recognized across the system. But I think alongside that, there was a sense that perhaps the organisation took its eye off the ball on the core functions of an NSI. And certainly, in my first few weeks when I went to see the Governor of the Bank and the Chancellor, they were really clear that what they need from us right now, is a real focus back on those core economic statistics and population statistics. So yeah, that was a very strong sense in those first few weeks

But of course, that means stopping doing some things. What is being stopped so far? What do you think is going to have to stop? How is that process proceeding?

So alongside my arrival, and just before, the organization had done some useful internal things to prioritise. So we moved some money around from the IDP programme.

That's the Integrated Data Programe, bringing in data from other parts of government.

And as that programme was coming to an end, we moved some resource from that. We also moved some people from one of our specialist hubs, to get them behind the recovery plans. There was some kind of internal prioritisation as it were, that happened as I was arriving, which was essential really, just to get us on the front foot, to then begin the recruitment of 150 extra people to come and work behind the recovery plans, and we've got around one hundred of those in place, and in Q! This year we will hire the rest. But that wasn't sufficient, so alongside that we made plans for prioritizing our outputs and that resulted in the kind of engagement that we have been doing for the last few weeks. So that process is still underway, and we are hoping in the next few weeks to bring some of those elements to a conclusion. I mean, I think inevitably, as we've done that, there are lots of things that people would like us to keep doing, but I hope they understand, and in deciding what to de-prioritise, again, there's a risk that people think that no one values that work. And of course, that's not true. There are tonnes of people out there who really, really value the output that we might stop, but my message to them is that we really need to focus now, it's quite existential for us to focus on the recovery plans on economic and population statistics, to get those right. There won't be any new money coming our way, so we'll have to prioritize internally. None of that necessarily means that it will be forever, but it is important that we do it, so we're on top of what we need to be on top of for the next period.

It's clear that the pressure is ONS, particularly to get those core economic statistics right, and we'll talk a little bit more about the detail on that. But firstly, on this broad subject of user expectations. Parliament essentially asks the ONS to do two things; that's to provide the statistics to help policy makers, but also to inform the public as well on social and economic issues. Is there a danger in these circumstances, when demands from the bank and the Treasury are to get the first part of that right, is there a risk with the public interest, data which people access to form their own judgment. Is there a risk of that taking second place?

I think there is a risk of that, but we're very alive to it. And one of the good things that we did towards the end of last year is we were finalizing our mission statement. And in kind of doing that consultation on that mission statement internally, one of the phrases that we added at the end of our mission statement was to inform the public, and that was a deliberate decision to remind ourselves about that statutory underpinning that we have, but also to remind ourselves that we can't just focus on one thing. So as much as we need to prioritize, I'm very clear that we have more than one function. We're not just an Economics Institute, we do far more than that, and we will continue to do more than that. And that business to inform the public is absolutely core to what we're about. So your right to raise it as a risk, but I think it's one that we've clocked, and one that we're very clear about.

Okay, let's focus on one aspect of the turnaround plans in progress at the moment, which has attracted more attention than any single issue facing the ONS currently, and that's the very important Labour Force Survey (LFS). That's how the employment figures for the UK essentially have been complied for some decades now. It's one of the biggest, if not the biggest, regular household survey in the country, depended on by policy makers, economic commentators and, of course, politicians as well. It got itself into a bit of a state, didn't it? The response rates, it's not too dramatic to say, collapsed during the wake of the pandemic. How important is getting that right among your priorities, and what is progress to date, and what still needs to be done?

You're right to raise that. I mean, I've been saying to people that there isn't one single thing that we will fix in the ONS where we will declare victory. This will be an ongoing effort and continuous improvement. But if there was one thing that is quite totemic in the recovery it will be the labour force survey and the transformed labour force survey. So there's a considerable amount of effort across the organization in getting behind both of those bits of work. So it is quite totemic. You're right. It's the single biggest thing that the bank and the Treasury would like us to get on top of, and it also does go to credibility, because of the credibility of the response rates. So it is hugely important. I've been really impressed by the quality of the work that's been going on to try and get on top of it. And the team that's there is absolutely laser focused, doing both of those things at the same time. So the first things is fixing the current labour force survey as far as we can, and they've made some real progress with that, so getting response rates back up to pre covid levels. We've now got a field force workforce at similar levels, and we're starting to see some real benefit from that. So the OBR at the budget included a reference to the LFS in their budget report where they said they now have more confidence in the survey and I think that's real testament to the amount of work that has gone into it. But alongside that, I think what we do recognize is that the current LFS isn't sustainable in its current setup, so that's why we are working, again, with laser like focus on the transformed survey, to make it more online, a bit shorter, a bit more focused. And I'm really hopeful that that, over

the next year will prove itself to be a sustainable answer to this issue.

The long-term plan is to go out to people with a survey, which does take a lot less time to fill in. And that was one of the criticisms of the traditional one, that it represented a big time commitment for people and perhaps it wasn't surprising that there was a challenge around getting people to take part. But at the same time, we seem to have a particular problem in the UK, where we are on what evidence is available certainly, an outlier in terms of survey response rates. Have you got a sense of what the issue is there? Was it the ONS getting things wrong? Or is it something more profound about the culture now in the UK, that makes people suspicious about taking part in official surveys?

I think that you're right. We are a bit of an outlier now, because this kind of phenomenon happened everywhere across the globe in the pandemic but we seem to have been worse off than most. I think there is a kind of trust issue across our society in engaging with government agencies on data which is, I think, is more profound than some other societies. So I think we are fighting against that. I think it probably also suffered a bit from being quite a long-standing survey, having grown over time. I think one of the interesting things when you look across the globe though, where countries do this better than we do, at least in terms of response rates, is that most of those countries mandate their social surveys, and certainly their labour force survey equivalents. Now we don't do that here for our social surveys, but I think that's something that is a policy issue for ministers, something that we need to kind of resolve, to decide whether or not that is the long-term answer for a small number of social surveys.

So that's a conversation in progress at the moment, whether people might be required to take part in these surveys in future.

Yeah, it is. It was a recommendation in the Leivesley review that we at least look at it, so we're doing that to try and establish well what are the pros and cons of that sort of approach, so we can inform future policy decisions by ministers on this issue. I don't for a moment underestimate how tricky a policy issue that is, but I think we do need to confront it so that we can put it to bed, or we can progress it, because we are notable across the globe by not doing that.

So, there's a clear sense of urgency in the plan with action, as you've said already, to tackle really pressing issues like the Labour Force Survey. So we're getting more people into the field doing interviews and so forth. And in the meantime, a conversation about prioritizing core statistics. But what comes next in 2026, what should people who might be following the ONS look out for from the organization next?

I think 2026, I've been saying internally, is our year of delivery and action and recovery. So last year, we obviously had the crisis point in the summer. And since the summer, we've been doing a lot of work internally into our overall plans, our mission, how we want to lead the organization, what we want the culture to be, what are our plans for getting on top of the economic and population statistics, preparing for the Census...So there's been lots of planning, lots of strategizing, lots of structural changes, that sort of thing, that I think was absolutely necessary. But the focus for 2026 I think needs to be on execution and delivery. To spend this year delivering against the plans that we've made and that's going to be the focus of the top team for the next twelve months.

The ONS has enjoyed a reputation for innovation, and that really came out during the pandemic. Is there a risk that innovation will cease or are there new things happening on that front to really improve the overall offering?

So again, I think that is a risk, and it's something I've talked about within the organization. That because I've come in with such a sharp focus on the recovery plans, it could just be head head down and focus on those things for the next two years. And then wake up in two years time to realise we are vastly behind the times, on AI or some whizzy new ways of using admin data or whatever it might be, and that's why we've done some things internally to make sure we still have some space and some capacity to do that innovation. I'm really keen that we do that, it can't be at the expense of the recovery plans so that's still our number one priority, but we have carved out some expertise and some capacity to still challenge us on how we should be doing statistics in the future. A big part of that will be how do we learn from others on AI and on tech more generally. So watch this space, it's definitely still something that we're focussed on.

Yeah, the resistance that's out there on some people to take part in those surveys. What about all this data that government has been already it's been a persistent criticism that government collectively hasn't done enough to pool that data and to and to really get get value out of it and to make public services more efficient, policies more effective.

I think that is a legitimate criticism if I'm honest, you know, it's been a conversation live in Whitehall since I can remember - data sharing and the lack of it. Obviously, there's been huge advances in that recently, some really good innovation around data sharing, including in the ONS, but even more generally, but I still sense there's real risk averse culture around data sharing across government departments that I think we need to try and tackle. For me it's as much about culture and ways of working as it is about legislation and risk. And I'm quite keen that we play our part in pushing the rest of the system on that. I think it's going to be one of the benefits of the split of the role, so that we will have a National Statistician full time being the National Statistician who isn't worried about the running of the ONS. And I'm really keen that whoever that person is, in future, if they can help the system really galvanised around data sharing and data linkage, that will be a real benefit of the new approach.

What about the culture of the ONS itself? And as you just said, there are two people at the top of the system now, as it were, an incoming national statistician, a you as Perm Sec. Culture has got to be your department, hasn't it? And the Devereaux review talked about some really quite deep-seated cultural issues. What was the culture that greeted you on arrival? And what mechanisms do you think are going to help to create the right culture and deliver the dramatic changes that you're working towards?

Yes, I think what Robert set out in his report was accurate, and that's certainly what I discovered when I arrived. I think it manifests itself in a range of ways, but perhaps the two biggest for me are probably a lack of trust across the organization. And that's trust as it was between the board and the executive team, trust between the senior civil service and the rest of the organization, lack of trust between our trade union colleagues and management, lack of trust between some of the different teams themselves. And I think that that, having been left unchecked, is quite corrosive. I think it's led to some of the issues that Robert set out in his report. I think it also leads to a lack of collaboration, a lack of join up when that is the predominant culture. So we've spent quite a lot of time, of my personal time, has been focused on breaking through some of that. It's quite a hard thing to do within organizations - to change the culture. I've deliberately not had something called a cultural transformation program, I'm not sure they ever really work. I think what we can do instead is just start doing stuff differently, and that's what we've been doing. Some of that has been a bit managerial. We put in place a leadership statement, for example, something the organization could use to hold senior leaders to account. I 've started sharing performance panels for senior civil servants. We've put in place some slightly sharper performance management. We will start doing regular pulse surveys, just to try and instil some of that managerial approach to fixing some issues. But for me it's also about who do we reward and why. One of the things I've been banging on about since I arrived is the "three C's" - that's clarity, consistency and consequences, and these really speak to the culture. And just briefly for listeners, clarity is what are we for, does everyone know what that is, and does everyone agree with it. And that's why we've created a new mission statement. Consistency is, how do you apply that set of values to the things you do on a daily basis, and you don't let it go when times get tough. And then consequences...there has to be both positive and negative consequences for people in the organization for doing the right thing or not. And I think that that kind of framework has taken hold in the organization, and I'm quite keen that we keep using it keep using it as our bellwether for are we doing the right thing. Do we reward the right behaviours, are we all being held to account for how we want the ONS to be. I think that's how we are going to get the change here.

What would you say to someone who might be thinking of applying to a job at the ONS, they might think well what we do is really important and really interesting, but I've read this stuff in the news which suggests, well, they've got their challenges.

So this is a live issue. So we've been recruiting for some very senior posts in the last few months, and I've been really haertened by the response that we've had. So for example, we are interviewing later on this month for Director General for Technology in the organization, we had over 200 applications for that, and I spoke to quite a few people who were interested in advance. And, of course, the thing is that once you're away from the heat of the crisis that the organisation was in last summer, for people who want to come and work in this organization, if you're a technology leader, for example, coming to work at the ONS, which basically just does data, you know, we collect, analyse and publish data. That's hugely attractive if you're a senior leader in technology or data. And I think I've been able to persuade them that we've turned the corner on what it's like to work here. So I've been really kind of enthused by both the quality and the number of people who have wanted to come work for us for us in some of those senior roles that we've advertised, but similarly, at all levels of organization. So one of our campaigns to hire some junior analysts across the organisation attracted hundreds and hundreds of applications. So I still think that the core business of the ONSremains very, very attractive to lots of people out there. My job is to turn the page on the culture of the organisation and that remains the case.

And as we start 2026, what gives you the most confidence that ONS is now on the right path? And what are the main objectives ahead?

I think there were some emerging plans which were kind of credible, I think we've really strengthened those in the last four or five months. I think the quality of the people will take us a huge way in our transformation and I'm really enthused by that. I always think of Colonel Boyds aphorism of "people, ideas and machines. And in that order." And I think we have excellent people who are generating brilliant ideas, and I think we're getting on top of the machines, as it were, so that we can really transform ourselves. What I really want is for the ONS to be in the news for its outputs and not for the organisation itself, and this year is going to be all about making sure that's true.

Because of course some people might say, well, we've got AI that's going to revolutionize everything. We're not going to need the ONS anymore. Just how wrong, or indeed right, is that?

My sense is that AI will have a huge part to play, but it's not going to take over. So again, that aphorism of people, ideas and machines, in that order. And it's the "in that order", that I think is the most important thing. AI will be tremendously helpful to us, but it's never going to replace expertise. So I'm quite keen to harness it, and use it for our internal efficiencies and productivity of course. We should harness some of the great ideas that are emerging out there in the private sector on stats production and surveys, we should absolutely do that, but I don't think we should be naive in assuming that there's suddenly going to be this great fix, I just don't think that's true. So I'm quite keen to continue to kind of experiment with AI, and we've already been doing that in the organization with some success. We want to keep scaling that up, but I want to temper the enthusiasm with some reality, that at least for the foreseeable future, we're still going to be using surveys, we're still going to be using admin data from across the system, and then on top of that we will use some of our AI expertise to help us

And we're going to have a census in 2031, it survives for another time around...

It definitely does. The government's taken a very firm decision on that, and we're now ramping that up at quite some level, as we've got the senior team now firmly in place for the Census. This year will mark the ramp up phase in preparation for the test in 2027, and it's obviously a hugely important thing for us, and I'm quite keen that we get the maximum benefit from ramping up the census at the same time as focussing on our recovery plans. So yeah, so I'm excited about the census

Darren, thank you so much for that fascinating tour of what you've been doing and what lies ahead. Just finally then, what's your message to everyone who relies on ONS data, or who might be interested in ONS statistics, for 2026?

My overall message is, work with us. This is a really important time for statistics in the UK. We need everyone's help. If you're a citizen, fill in the surveys that you're asked to fill in as that's hugely beneficial for your systems. If you work in the system, bear with us and lean in and help us, because we will need your help as we get through the recovery phase here. But I'm optimistic that we can turn the corner, and that the statistical system in the UK will be in good health by the time we get to the end of 2026.

And on that note, we come to the end of this podcast. I think it's fair to say we've heard a clear sense of direction, a renewed focus on the statistics people most rely on, a commitment to improving quality and resilience and the recognition that trust in official data has to be earned and re-earned over time.

My thanks to Darren Tierney for joining us, and to you for listening. You can follow developments on everything we've been discussing today by subscribing to future episodes of Statistically Speaking on Spotify, Apple podcasts and all the other major podcast platforms. You can also follow us on X, previously known as Twitter, via the at ONS focus feed and do check out the national statistical blog for updates on all things ONS. You can find that at blog.ons.gov.uk, all lower case.

I'm Miles Fletcher, and from myself and our producers, Julia Short and Alisha Arthur, goodbye.