#24 - Warren Smith (Digital Marketplace, Government Digital Service)

The UK's Digital Marketplace


Interview with Warren Smith, Director, Digital Marketplace, part of the Government Technology in the Government Digital Service. He enthusiastically and wilfully oversteps the mark to meet user needs in public procurement, in public sector procurement and contracting. He’s currently transforming the way that the public sector commissions products and services by thinking small about big problems, and thinking big about small successes.


It’s good to have you here, I’ve had you in the list for an interview for some time. Every now and again I like to interview someone from practice, who can bring a completely different view and perspective to what we would call legal, or academic problems, related to public procurement. So it’s great to have someone with a wealth of experience working in practice with public procurement. Speaking of that, could you tell us a little bit about your background, and what is a Digital Marketplace?

Sure. So I have been in procurement for 20 years now, I think. When I was fresh out of university I went straight into a procurement role. Really it’s only in the last five years, I would say, that I’ve been involved in quite transformative projects: working at scale and having a real opportunity to make a significant difference by introducing new ways of thinking about procurement and contracting.

I’ve really been focusing on trying to introduce the concept of user-centred design into procurement and contracting, as well as focusing more on outcomes rather than detailed input requirements, as well as introducing open approaches. 

What does that mean? It’s kind of being much more open about thinking, sharing information and visions, way in advance of actually procuring: so trying to engage with the market earlier. Also open standards and open data have a very important role. For me, making things open is a really fundamental tool for disruption, I think, which we can employ in many different areas. So that’s a little bit about me. 

I am a member of the Chartered Institute of Procurement and Supply, but I tend not to, to shout about that, but it’s a useful background, obviously, in the space that I’m in. But really my interest is in trying to take the focus away from procurement being a driving force, and make it, rather that procurement and contracting are enablers of better public service delivery. That’s a really important point. Often people who I’ve spoken to see procurement as being the end in itself, rather than a means to an end. That’s what’s brought me into the government. 

I had the good fortune of being asked to help GDS towards the end of 2012. I was brought in by the former programme director for the transformation programme, which was focusing on delivering the 25 exemplar services. These were the 25 largest government services by transaction volume, and it was the start of, really, the digital by default agenda within government. 

At this time different policy contexts were being set, Liam Maxwell, the former Chief Technology Officer had been brought into GDS to bring about reforms in the way that the government thinks about, mainly digital and technology. My role was to help rethink procurement in order to support those reforms of digital technology in government. 

So I’ve been here since 2012, and just something to correct you on, Pedro, and this is hot off the press, this week I’ve been appointed as the Digital Marketplace Director. Having been through an interview process, I’m now no longer the interim director, I am the appointed Digital Marketplace Director. 


Thank you very much…

Speaking of the Digital Marketplace, what are its objectives, and what are you trying to achieve with it?

I think you summed it up quite nicely in the introduction, but basically we are all about helping the public sector buy what it needs to deliver great digital services.

This is a key strategic component of the technology group within Government Digital Service, and it’s important that I just go back to that mantra there of helping the public sector buy what it needs to deliver great digital services, because that, again, is the outcome that we are trying to support. 

Everything we are trying to do is about transforming procurement so that it can help that aim of better public services that are digital by default. We are looking at the entire, end-to-end process of procurement and contracting, and focus on the areas where we think we can make the biggest difference, deliberately, taking thin slices out of that end to end process, applying user-centred design to that, and then iterating wildly, so that we can continue to deliver better procurement and contracting experience for buyers and suppliers. 

Those are our real primary user groups: buyers and suppliers. At the moment we are looking at frameworks quite specifically. I know you, yourself, have spoken about that in some of your publications, Pedro, and I think there’s a real interesting opportunity here. Within the digital and technology space in government we are looking to disaggregate large requirement sets into capabilities that meet user needs, and thinking more about how we might diversify and open up supply chains into government. Actually from that what you can then start thinking, is that the work packages and the contracts change into shorter term, lower value and quicker delivery [pieces of work]. So therefore, the sub OJEU threshold might actually be a really interesting, untapped opportunity there. 

We’re developing the Digital Marketplace so that we can award a framework agreement, then provide an end-to-end buying process where buyers and suppliers use the framework, get through the evaluation process, award a contract, get on and do the delivery, using things like the Open Contracting Data Standard to improve the quality of the data that’s actually the published automatically out into wherever, contracts finder or similar.

Can you give us examples of changes and simplifications that you’ve done to the procedures and contracts for the digital services framework?

Well, actually, we started on that with the G-Cloud framework, Pedro. G-Cloud 7, as in the seventh iteration of the G-Cloud framework, was the first time where we ran the entire OJEU  procurement through the Digital Marketplace, so that enabled us to really challenge the  -- I have to be careful of my use of the word “challenge” describing procurement -- but challenge CCS and government legal department to think differently around  what are traditionally the selection and the award questionnaires that you would normally see. 

We looked closely at the invitation to tender templates as well as the framework and call off contract itself. That was where we changed the supplier application process. A supplier simply had to complete what was called a supplier declaration, where they answer a series of questions which typically would have been the selection questions. That became a page, or several pages within the Digital Marketplace where we reused the gov.uk design patterns for form elements to ask the right questions in a different order than normal. 

What we wanted to do was ensure that if a supplier was new to government, or even any supplier, it was easy for them to quickly decide, “Is this an opportunity that I want to go for?”. 

We wanted them to be able to make that decision as early as possible in the application process, for them to be able to decide, “Actually, no, this isn’t for me, I’m going to decline,” rather than going through a load of questions, only to realise that actually, “This isn’t an opportunity that either I am interested in, or isn’t relevant to me.”

It follows the principle of fail early, fail fast. We found that what might seem like a logical order of questions to a procurement person or a legal person, wasn’t a logical order to a supplier, and we’ve found that out through quite simply asking the suppliers, and undertaking user research with the suppliers to understand, well, ‘what are your needs, and how can we design an application process to better meet those needs’? 

That was really a fundamental point in time where we changed a supplier’s experience of applying to be on a government framework. That’s really the first moment where a supplier who maybe has not previously been interested in doing business with government, that’s the first moment where we’re saying, “Hey, come and get involved with  government,” and so we think it’s really important that that first experience is actually a -- dare I say it -- a delightful one…!

That was the first thing. And then the learnings and the insights we gain from that, we’re able to then apply to subsequent framework applications. Digital Outcomes and Specialists (DOS) was the next framework where we really fundamentally redesigned a broken framework. The digital services framework had received quite scathing coverage within the technology press, labelling it not fit for purpose, talking about the relationship between GDS and CCS as being “dysfunctional”, so we had to, and…..Yeah, absolutely. This was escalated to quite a senior level within government, so we had to make a difficult decision which was, we can’t continue with the traditional approach of delivering procurements and contracts in government. 

We needed to stop, we needed to take a different approach. And that was where we agreed with CCS and Government Legal Department that we would form a multidisciplinary team, we would use the GDS design principles as our guidance on how to design and deliver user-centred frameworks and procurements, and using the body of better practice that is the service design manual to basically guide us in all of those things. 

The approach was: working in an agile way; being open; being more engaging with the market to design a completely different framework agreement. And also going live with what we classed as the earliest, what others would say is the minimum viable product, but actually we wanted to call it the earliest usable product. 

This was the buying journey of digital outcome specialist within the Digital Marketplace, so that needed some handholding with CCS to say, “Look, it’s okay to go live with a product or a service that is not complete, because there is no better thing than putting it in front of real users, who are using it for real,” so that we can actually get feedback from those users, by those suppliers, about how is it working. To answer the question: ‘What do we need to change’? 

We could also to use data, as well, analytics from the usage, to inform our iterative service design. That’s never been done before in government, and I’m always proud to be able to say we’re chalking up quite a few government firsts here. 

My next thing, which hopefully we’ll go on to talk about, is just the fact we can iterate and incrementally improve a procurement and contracting experience for buyers and suppliers. This is a massive paradigm shift, I think, in government procurement.

It is, because usually what I’ve seen over the last, let’s say, 10 or 15 years of working in procurement is that once a certain procedure or practice is developed, and usually it’s done organically without any research and looking at what actually works, and without considering the impact it may have on the supplier side, once that is done, it’s done, and it stays there. 

There’s no incentive and there’s no work to revisit what is being done today, to say, “Okay, if you were redesigning it from scratch, how would we do it? And how would you improve it because you now have different technologies, you have different techniques, you can do things differently?” that doesn’t happen in public sector in general. I don’t see this as a criticism more than as a statement of fact. 

So a few years ago, as you know, when I was working with local councils to improve their practice in contracts below the thresholds, that was very obvious, that I looked at procedural, or documentation for the public procurement procedures that you’re following, and you could see it growing up over the years, with new stuff being added without any consideration exactly about what is happening, or what are the reasons for those extra questions or extra elements of fact or whatever to be added into the mix, and

I think it’s very important to get out of that frameset, and instead look at what is being done today, and say, “Okay, do we really need to do this? We really need to improve this.” On that note, bearing in mind that you’ve done a lot of work in simplifying contracts and procedures, have you seen any correlation, or at least relation, between simplifying the contracts and procedures and increasing the participation of SMEs in public procurement?

Absolutely. There is a strong positive correlation. I think, if I state that correctly, that we’ve seen just by simplifying the application process and being more open and describing our intent and what we need, and describing the contractual situation in plain language, I think that we can demonstrate a significant difference between digital services framework, which is the one that was receiving a lot of the criticism, which had something like 150, 160 suppliers on it, across the UK, to Digital Outcomes and Specialists which now has over 1,200 suppliers across the UK, and over 90% of them are SMEs. 

Now, what that is is us creating the opportunity for these suppliers to do business with government, because as we all know, being on a government framework is no guarantee of every actually winning any business, right? But at least we have created the market so that these significant number of suppliers now have that opportunity.

So getting that step one right is really fundamental, because unless you open the door, you’re not going to enable those suppliers to be able to win the business. That’s the first step. 

What we now have to do, and we’re still relatively in the early stages of Digital Outcomes and Specialists, which only went live on the 27th of April, the day before my birthday. We’re approaching three months of usage, but already we’ve seen something like over 160 supplier opportunities published, and we publish those in the public domain, on the Digital Marketplace, because we want suppliers, and not just suppliers, actually, we want suppliers and the public to see, what’s the demand of government? 

What’s the opportunities that, if you’re not a supplier currently on this framework, you’ve actually now got visibility of the demand of government for digital specialists and services associated with the design and delivery of better public services. Having that information in the open is, again back to the open contracting point, is helping to, transform the relationship between the citizen and the state. 

I think we’ll be able to see very soon what’s the actual business that’s being won by suppliers as a result of the changes that we’ve brought about through digital outcomes and specialists. We already have some data about G-Cloud usage. Since its inception, business through G-Cloud has helped the Digital Marketplace go to sales of something like £1.3 billion as it is now. The bulk of that is through G-Cloud, but I think we’ll see it, I’d be very interested, and wait with anticipation, to see what the shift might be from cloud into Digital Outcomes and Specialists, and indeed the growth of cloud as we approach things like G-Cloud 9. 

We’ve just kicked off a discovery to go back and understand how we might need to adapt the design and delivery of that framework based on an understanding of how the changes of the needs within government have happened since 2011/2012, and also how the market has shifted, as well. 

Again, we’ll reuse the approach we took to Digital Outcomes and Specialists to apply a user-centred design approach to deliver the 9th iteration of the G-Cloud framework.

Speaking of what you were saying before, and classified it as a world first, you’re trying to move the government to adopt open data standards, am I correct? So what do you actually want to do in terms of practical changes to the way that the government works, and what do you think is going to be the impact of adopting open data standards in procurement?

What does it mean practically? Well, first it’s great we’ve made this progress. In its open government partnership national action plan for 2016 to 2018, published on the same day as the anti-corruption summit that the UK government hosted, there are very firm and clear commitments around implementing open contracting, as well as a number of other fantastic commitments. 

That gave the backdrop that we can actually then get on and do something. I applaud Crown Commercial Service for making those commitments to implement the OCDS within their operations by October. It’s fantastic, and same with HS2, and beyond that, across governments. 

I saw this as an opportunity to ensure that we’re doing what we can to support those commitments, which is why I’m championing the Open Contracting Data Standard going through the standards hub process right now. Once that’s mandated, that’ll ensure that the OCDS is used across government as an adopted standard across government for open contracting. That’s a practical step to support that. 

Then, as part of that process, we’ll be understanding, “Well, what are the opportunities and the challenges of implementing open contracting, and the OCDS specifically?”. Because we’re effectively designing and delivering an end-to-end buying and procurement and contracting process, in Digital Marketplace we have quite a unique opportunity to embed OCDS and open contracting as a default component of our own platform development.That’s effectively what we’re doing, and we, you know, our ambition is to be the default place for buying digital and technology within government, the Digital Marketplace to be that, and also we can open up our platform for use by others. 

And in terms of another government first, you know we’ve been approached by the Australian Digital Transformation Office to support them with their creation of a Digital Marketplace, so rather than just simply reinventing the wheel, we’ve already shared all of our code, we’ve given them access to our hub repositories, all of our guidance, all of our know-how, and anything that we can share, we have shared. I’m very excited to tell you, we’re sending two of our team over to Australia this weekend, for two weeks, to help the Australian Digital Transformation Office to really accelerate their own development. I’m really excited to see, actually, rather than simply just giving them access to what we’ve done, how might we actually work from a common code base and not just code, but a common asset base for things like the contracts, the design patterns, etc, to really do, government to government collaboration on an area where we have very, very similar challenges and problems around reforming procurement and bringing procurement into the digital age. I think that’s really exciting. We’ve also been talking to the US around similar things, so I see some quite interesting opportunities to collaborate between governments to do great things.

It’s great to see that you’re able to collaborate across borders with other governments, you know, Australia and eventually in the US, as well, but in terms of the rollout in the UK, one of the things I’ve noticed is that there’s a delay or a gap, or even a complete ignorance, to a certain extent, in other levels of government, in comparison with what is being done centrally, in this area, let’s say, of improving procurement practice, in changing procedures, changing contracts, all that and the other. And that, for me, is very, it’s very strange, because coming from a different country where practice is dictated by law, effectively, everyone works in lockstep, so the procedures are very similar in central and local government, or original government, so there’s not too much deviation. So whatever is done centrally, sooner or later rolls out to regional and local governments in Portugal, but to a certain extent that doesn’t seem to be the case here in the UK. So is there anything that can be done to change and improve the rollout of the great stuff that the GDS is doing in other levels of government?

Yes, absolutely, and that’s a really important part of our aim of GDS to transform government together, and work with the wider public sector, not from a position of imposing central mandate or anything like that, but actually being able to reuse what’s considered to be better practices and standards, ways of working, and practices, to help everybody transform themselves into digital organisations, effectively. 

There’s already been some great examples of reuse and collaboration so, for example, the digital service standard, which is, new policy for central government in terms of design and delivery of citizen-facing, and actually internal systems and services, that has been reused and adapted for local government. Now they have their own digital standard for local government, which is, by and large, very similar to the one that was put in place for central government, but, because they recognised that that actually is the way forward, but it wasn’t about us saying, “This is what you have to use,” they recognised that, they took it, they adapted it to reflect the nuances of local government, and they’re now putting that in place.

That’s just one example, and I know that there’s already some interest around looking at things like the technology code of practice, which is again a policy within central government for digital and technology spend controls, which GDS has the delegated authority from Cabinet Office and Treasury to ensure the right behaviours and the right approaches are taken for digital technology spend.

So this technology code of practice is being updated currently, and I think that will be really great to see that reused across the wider public sector, because that’s about helping government to think differently about the way that it approaches its requirements for digital and technology, for service design, for making sure that user needs are at the heart of that, that you’re disaggregating, you’re favouring more competition, from a broader and more diverse range of suppliers, you’re using open standards, you’re making sensible decisions around security, and you’re supporting, effectively, delivery of government as a platform. I think there are some really interesting opportunities. 

There are going to be some interesting conversations along the way. Some of these things are different to some of the entrenched ways of working, and ways of thinking. Helping people to be comfortable with using, for example, public cloud services, you know, more utility-type services, which can be consumed on a pay-as-you-go basis, rather than on-premises infrastructure, takes time. That requires being available to help coach and mentor, and to build the capability and the confidence of people within government and wider public sector who aren’t used to this new way of thinking and working. That’s very important.

One final question. I saw last week a blog post written by you and Jason Waterman from the Crown Commercial Services, about working together to simplify the contract language, and the way that contracts are drafted in public procurement. Can you tell us a little bit more of what you’re trying to achieve, and what are going to be the next steps with that?

Yes, sure. This is, I think, one of the most exciting things that I’m involved with currently, and will be involved with going forward. This is an opportunity to fundamentally transform contracting and procurement within, initially, central government, but I think once we go through the project we’ll see the wider use cases. So Jason Waterman, who’s responsible for policy delivery within Crown Commercial Service, he came to me with a, kind of, a plea for assistance, let’s say. 

Having seen what we’d managed to achieve for digital outcomes and specialists, in terms of a very different way of working, and an output which was a framework agreement that was plain language, it was HTML, I mean, imagine a contract in HTML rather than .docx [laughter] which conforms to the GOV.UK style guidelines, which are very strict quality control to ensure that what gets published on GOV.UK is actually clear, simple, etc.

Very briefly: the process we went about to design and deliver that framework was we had content designers who are actually holding, let’s say, holding the pen, in a digital sense, who are writing the content, but taking input and working alongside procurement and legal subject matter experts to ensure what was created was correct. 

So rather than it being legal or procurement people who were holding the pen, it was people who were trained in designing content. So that gave us a very interesting opportunity to think very differently about the structure, the form of the language, the layout, the usability, the interaction of a government framework agreement. Now, this is not to say that what we’ve delivered for digital outcomes and specialists is the finished product, it’s the nirvana, but what we have done is prove a point that you can apply a very different way of thinking and working, of user-centred design and thinking of the thing as a digital product. 

It’s for the internet, it’s of the internet, it’s applying the digital age, to an area of government that has yet, really, to be transformed and disrupted in this way. To go back to one of your comments earlier, Pedro, is these contracts in government often, accrete over time, based on, case law or whatever. They may have been through a legal challenge and therefore a successful upholding or whatever it meant that we need to do things in a certain way. 

I wanted to introduce the idea that actually we can challenge that in a sense of, “Let’s understand, what are the user needs of a contract? Who are the users of a contract? What are they trying to do?. In terms of the model contract in government, “What currently is working, and what currently isn’t working?” and you’ll see on that blog post, there’s a picture of a contract, or actually it’s a framework agreement, which has been hacked by the user. 

They’ve gone through it and they’ve stuck little sticky notes into the important areas, and they’ve put handwritten labels on that to help them navigate and get quickly to the points within the contract that are important to them. So why don’t we actually design a contract based on an understanding of what the important points are? 

I know we’ve spoken in the past about, one-page summaries, or whatever, but you know, by thinking of this thing as a digital product, we have a unique opportunity to design something that meets the needs of the buyers and the suppliers. Actually thinking of the users being the people who actually need to work together on a day-to-day basis to design and deliver a contract which actually helps buyers and suppliers successfully deliver projects, or whatever it is that they need to do to have sustainable commercial relationships that meet the needs of those two organisations. 

When we start thinking in those terms, there’s a very different set of opportunities which start, coming to the surface. What about if, instead of thinking of the model contract, which is the usual starting point, what about if we have model terms? 

Those model terms themselves will be plain language by default, but also possibly visual, as well as language. They could be visual to help convey what are often complex concepts in, you know, either intellectual property or liability or exit or whatever, that they can actually visualise as well as describe clearly. 

And once we start thinking of model clauses, once we then start trying to understand, what’s the context? What’s the commercial context that this particular contract is there to meet? How would we then assemble those clauses with an understanding of what that context is? So we right-size a contract based on an understanding of that context, rather than starting with all possible clauses with some guidance around, “Well, you don’t, you take this out,” or, “You make sure this is in,” which means you’ve got a really unwieldy starting point. We actually build a contract based on an understanding of what that context is. 

As a digital product, it also means we can put in, and weave into it, the open contracting data standard, so that we simplify the ability to actually disclose consistent, machine-readable data into portals such as Contracts Finder, or the other ones that are available, or even any web service where actually they can just take this information, because it’s publicly available and it’s consistently machine-readable. 

So really, what I wanted to do within that blog post with Jason was just set out what we envisioned is something that’s going to be very, very transformative across all of the common goods and services of government, not just digital and technology. We’re going to be recurrently building the team, and we’ll be kicking off a discovery and then going through the design phases of the service design manual of discovery, alpha, private beta, public beta, then into a live service, but even when it’s live it then won’t be sitting still and stagnating, it will be subject to continuous improvement based on user feedback, user needs and data.

Yes, I think that’s a great project, and something that is going to take some time and energy to get going, and get it to the level it needs to, but I totally agree that is an area that needs to be looked at, and I was listening to you, and effectively immediately came to mind something that has been used for ages in international contracts, or international trade, which is what we call Incoterms, which is effectively a term that contains a clause which is very  clear and very detailed and very stable, so that when you say that whatever you’re buying is subject to Free On Board, which is the name of the Incoterm, you know exactly what it means in any context. So, in effect, to a certain extent, I’ve written about that, on my blog, that one of the things we should be doing in procurement, instead of having only the CPVs for the procurement procedures is having something similar for the contracts and the contract delivery, so some sort of Incoterms for procurement. So I’m really, really excited that you guys are looking into this, and perhaps that is an area that you could explore as a starting point, to see what is out there and what is working already, in a completely different area of work.

Thank you, yes.

We have to finish, it was a great, it’s a shame, I mean, we could have kept on talking for another hour or so, but we need to close it down. So thank you very much for accepting the interview request and giving us almost an hour of your time for a really, really entertaining and engaging talk.