Government shared services: Cost £1.4bn, savings…erm….

Government shared services are getting beaten up again. Most of the information in the Commons Public Accounts Committee’s report on government shared services was drawn from the NAO report published earlier this year (See previous post.) but it gave MPs the chance to give the Civil Service another kicking. The headlines are that Civil Service shared service centres cost £500 million (55 percent) more to implement than expected and none have shown any clear savings. Two have even managed to increase the cost of support services by a total of £255 million. That takes some doing!

Some of this is due to the all-too-common problems with central government IT projects but other key factors are the failure to mandate the use of the centres and the failure to standardise processes:

Shared service centres were built without a confirmed number of users and the use of these centres by departments has been voluntary. Some centres have failed to attract users and continue to have spare capacity so have therefore not achieved economies of scale.[15] The Cabinet Office acknowledged that the “one thing that drives benefit and efficiency in shared services is scale.”


Many of the customers of shared service centres have insisted on services that have been tailored to their systems rather than the shared service centre providing one process suitable for all customers.[19] This has resulted in overly complex systems and processes which are less economic. The Cabinet Office’s new strategy proposes that shared service centres will provide the same “basic, simple and reusable” process for all customers.[20] Whilst this is welcome, the Cabinet Office must be able to ensure much greater usage of the centres on this basis.

So you end up with departments where some bits are in and some are out. For those areas covered, you have mini HR, finance, IT and procurement dedicated to each customer, ruining any chance of economies of scale.

Colin Cram provides some historical context:

Shared services existed in central government 30 years ago. ‘Joint services’ is a better phrase. They included the Whitehall estate – owned by one organisation, the Property Services Agency, which was also responsible for all construction work and held the budget for it. Other joint services included HR, ICT, training, legal services, procurement including contracting, audit, payroll and recruitment. They were inefficient, delivered poor service and serious corruption existed in the Property Services Agency. The units were broken up and departments were made responsible for running the functions themselves on the grounds that they would have an incentive to run them more efficiently. Unfortunately, departments generally did not have the skills to do so and massive duplication was created. So, we have a Conservative government rolling back the reforms of a previous Conservative government.

Another case of trying to use organisational structure to solve a problem that was nothing to do with organisational structure. Weak managers and corrupt employees don’t become any less weak or corrupt just because you move them elsewhere.

Trouble is, the decentralisation also opened the way for more complexity. Managers in devolved organisations changed things like grading structures, pay review processes and HR policies because….well….because they could. It was less about business imperatives and more about spraying trees to mark territory. Look, we’re autonomous now and to prove it, we’ve got separate grading structures and pay review dates.

Which is all very well until you try to pull it back into the centre again. What had once been similar processes now have considerable variation and the users cling to their preferred way of doing things. Once again, it was hoped that changing the place where things sit on the organisation chart would solve the problem. Once again, it hasn’t.

As Colin Cram says, there is way too much wriggle room here. And civil servants, being very clever people, are good at coming up with justifications for their wriggling.

So, there is plenty of wriggle room for departments, particularly the larger ones. One can bet that a huge amount of wriggling has already taken place, hence the shared services plan already sounding like the lowest common denominator.

There will no doubt be much sophistry in the next few months from civil servants anxious to maintain the status quo or adopt rather ineffective collaborations.

The government needs to maintain its nerve, be genuinely radical and implement joint services very quickly. Funding cuts forced local government to start to address shared services seriously. Further funding cuts to central government departments might help concentrate minds.

Devolution and voluntarism may be where it’s at, daddio, but they haven’t delivered much in the way of savings. Shared services only work if everybody uses them and everybody does things the same way, otherwise you might as well not bother. Unless someone in government forces managers to comply, they will continue to make excuses for opting out. And the cash will keep flowing down the drain.

This entry was posted in Uncategorized. Bookmark the permalink.

14 Responses to Government shared services: Cost £1.4bn, savings…erm….

  1. Pingback: Government shared services: Cost £1.4bn, savings…erm…. - Rick - Member Blogs - HR Blogs - HR Space from Personnel Today and Xpert HR

  2. fatjacques says:

    Economy of scale is a myth in the service sector. Cost is in flow, not scale. Increasing standardisation reduces flexibility, increases failure and waste.

  3. Matthew says:

    Nice summary of what goes in shared services in the UK and how it came about. I do hope we learn the lessons over here in Belgium soon enough and don’t fall into the same trap.
    I just can’t buy in to the solutions that are proposed: suffices to get acquainted with what Vanguard (Method) says and writes about standardising processes, IT-projects, economies of scale. It beggars belief in shared services. Full stop.

  4. Howard Clark says:

    Quick question Rick.

    You write.

    ‘Shared services only work if everybody uses them and everybody does things the same way, otherwise you might as well not bother.’

    But where is the evidence that shared services work even in this context?

  5. PinkPolitika says:

    Isn’t there a rather simple explanation of all this?
    Very few politicians, and especially very few Eton-educated ones, will have had any experience at all, ever, of the ‘services’ which a ‘shared’ facility might be expected to deliver.
    Like the cost of a pinta, it’s something Other People need to know about.
    All common sense anyway, isn’t it?

    • I was of course being ironic about the ‘common sense’… the comments and debate by others below illustrate why.

      But let’s get to the bottom of this: the current administration has no concept of evidence-based policy because it really does not get what joined-up policy would be about.

      Conservatism however is hard-wired to embrace wilful individualism wherever that’s what those in power prefer (which is often, whether it’s one’s home viewed as one’s castle, or any given MP’s penchant for a particular plan off the back of an envelope).

      To understand sensible, approximating-to-seamless policy development you’d have to understand the critical, central underpinning role of the state (not only in enforcement but also as a genuine enabler) in any complex modern society.

      ‘Shared services’ are an example of idea off the back of an envelope. Maybe they could work for some things, but only as others have said with a very firm, supportive, steer.

      Just TELLING people to ‘share services’ works about as well as telling toddlers, without example, to share biscuits and toys.

      But of course that’s not the real agenda. Indeed, George Osborne’s grand plan to raze the state via De-governance ( might well benefit from the chaos of mal-functioning shared services.

  6. Vince Lammas says:

    Hi Rick,

    It seems the Public Accounts Committee and National Audit Office have both accepted, without question, the proposition that larger shared service centres are the right solution for some ill-defined problems. The critics of shared services, John Seddon and Howard Clark amongst them, reject the idea shared services can have any part in the route to efficient and effective service delivery.

    Of course the truth lays somewhere in between. For some, shared services are an ideal solution while others could better match their needs by establishing a local service. Anyone who thinks there is a universally correct approach should go and look at the evidence of what is happening (working with and without shared services).

    There are major challenges in defining boundaries and interactions between shared service teams and their customers. Deciding where “necessary variety” should give way to harmonised processing can be problematic – every client thinks they are special and some of the judgements can be quite finely balanced. This is why shared services prefer to focus on highly repetitive, standardised work – rather than areas with high degrees of variation or subjective decision-making and judgement.

    It is certainly possible for a specialised provider, appropriately managing and investing in its core business, to bring the equipment, organisation, technique and performance-focus which can deliver better quality services than smaller, local, under-invested and overlooked services could hope to offer.

    To understand this in context though, it is important to recognise there are plenty of local services operating business processes and working practices which would benefit from a major overhaul – with structures and techniques that fall far short of the ideal. For some organisations, shared services are a viable alternative to reshaping in-house provision.

    There are, however, many people in the public sector that have a poor experience of shared services; who would describe functions which have been poor at meeting the needs of, and managing the relationships with, customers. These services have been badly designed and managed.

    Where political pressure and single supplier frameworks are used as the main “change levers”, the solutions are likely to be less responsive and effective than where customers are supported to compare alternative and select partners that can match their requirements.

    Whatever the starting position, simply forcing organisations to use ever larger support units is definitely not the route to effective and efficient service design and delivery. “Smarter brains” rather than “stronger arms” need to be applied if shared services are going to provide the right solutions to client needs while reducing costs.

  7. Howard Clark says:

    Vince (Attractor consulting) this is a clear instance of misunderstanding/misinterpreting John Seddon’s argument.

    Re-read what Seddon says about shared services, because he certainly doesn’t …

    ‘reject the idea shared services can have any part in the route to efficient and effective service delivery’

    (as you claim above).

    What Seddon does argue is that Economies of scale are a myth (and he provides detailed arguments & evidence to support this assertion) and that standardization is anathema in services organizations.

    I would suggest that if you have evidence where you believe shared services is working, then you should identify the site and proffer the evidence that supports your assertions. All Seddon is asking is that supporters of scale and standardization should proffer evidence that can be tried and tested to see if it holds any water.

    • Vince Lammas says:

      Hi Howard – we meet (cross swords?) again.

      I understand and agree with much of the critique made of the “management thinking” that underpins many shared services (and other) projects and operations. Not repeating it does not mean I fail to appreciate it

      You might be right that John Seddon himself doesn’t use the words I did … in part because he has reached different conclusions based on different experiences … there’s no harm in that.

      However, the stance taken by some is that the whole of private industry is colluding in a sham. Creating this “impression” through consistent and simple undermining communication does not appear accidental but is simple and effective.

      Nowhere have my comments defended oblivious focus on either scale or standardisation as a way to deliver results. In that context I would proffer the suggestion nobody should waste time looking for evidence these factors are at the heart of successful shared services – that would be like trying to prove the non-existence of an almighty power.

      Please note I am not attempting to defend or support any particular shared service solution or provider. In fact most projects I have been involved in have been improving service delivery and cost within a business rather than through outsourcing or shared services. But that approach does require investment, time and effort that not every organisation can muster.

      As the key to successful service delivery is flow, would you, I wonder, accept there may be elements of service where good flow requires the supplier to do some things the right way efficiently and consistently? I would suggest handling variety in customer demand and simple consistent action on robust processes can both can be important to meeting the needs of a customer in terms they would value. Using Amazon’s shopping delivery service demonstrates this rather well.

      There are many factors which might lead to successful results being delivered more robustly and consistently by an external provider (outsourced or shared service) than by the separate services of multiple potential client organisations – or could mitigate against buying in such services.

      There are also environmental conditions that could support effective competition in the areas of service delivery and price – issues I think will be very important in the long-run.

      My comments above attempt to identify some of those issues and it would be interesting to hear your thoughts in response to these.

      Please note I am not suggesting you don’t understand them but I would say that there appears to be a reluctance on the part of some to engage in a conversation about things other than scale and standardisation. I would be very interested to read some ideas shared freely on what you think external providers or client organisations should pay attention to when planning to deliver or utilise a shared service model.

      • Howard Clark says:

        Why do we ‘cross swords’ Vince?

        I know that you sell shared services. I don’t damn you for that. I am not anti-private sector (we work with hundreds of private businesses large and small).

        I am interested in evidence.

        And I am particularly interested in the absence of evidence.

        So holding up an on-line shopping ‘mall’ as an example (Amazon) is instructive. Presumably it is closer to what you are selling. I would imagine it fits in the self-service arena and online stores.

        But as Grönroos noted some time ago, products aren’t services.

        As before, please provide an example of successful shared services that works.

  8. Vince Lammas says:


    I’m obviously relieved you don’t damn me for the work I do, especially as you don’t know what it is. You assert that I sell shared services, an error on your part as is your guess that I sell some kind of online or self-service solution. It might surprise you to realise neither of these positions you have adopted is based on sound evidence.

    It is true that I know many people working in public services and private industry (a quick look a LinkedIN will reveal this) and I have been involved in projects that aim to re-design services and find ways to improve the quality and cost of service delivery – often, but not exclusively, involving the use of new technology.

    I am interested in the way technology and organisations are evolving and changing the way services can be shaped and delivered for customers. I confess to being a technology enthusiast.

    Organisations have to work in the real world, where a decision to start, or not start, down a particular path has implications. They should always carefully consider the context, strategic and operational, cost and quality implications of the choices they make …. whether that is to form a partnership with an external business, reorganise existing operations or leave things as they are.

    I’m afraid simply returning to your request for evidence of one thing without contemplating the evidence about alternatives looks, to me, equally blinkered as accepting a proposition on blind faith.

    I suspect its best we leave the debate there – though I will continue to highlight and publicise developments, issues, concerns, examples of problems and experiences (positive and negative) with shared services and other solutions.


  9. Howard Clark says:


    How bizarre.

    You would propose going down a track to shared services without the evidence to support the track that you propose?

    ‘Economies of scale’ and ‘standardisation’ are two of the core claims that those supporting shared services make for the reason that they are sharing in the first place.

    But there is little evidence to support these claims.

    Blind faith pushing ahead with shared services without evidence is ‘blinkered’ because it misses the point about services organisations.

    To ask a question with the answer in your head already, seems an unusual approach to problems.

    Reflecting back upon some of the technology solutions over the past decades I can’t help but reflect upon the folly of CRM, SAP, IVR etc.


    • vincelammas says:

      Again, you seem to confuse me with a person who suggests organisations should adopt shared services or a particular business model because somebody else says so. If you look over the articles on my website you should understand that is not the case.

      People need to think about their position and environment, identify options, understand what has gone before, learn lessons and act appropriately. I think organisations need to make informed decisions about a whole range of things and I’m interested in how people can navigate these challenges.

      I cant quite see that “not” doing shared services means an organisation is necessarily benefiting from good service design nor has in place effective working practices (even in terms which you might accept evidence for). So where is the evidence for doing nothing?

      I would also point out that it’s very easy to find technology which has been bought and implemented in a manner which seems to add little value to the customer or the organisation. Apparently the majority of projects fail to deliver the intended benefits.

      Yet I don’t see you saying “software doesn’t work” or “projects don’t succeed”, “where is the evidence?” and I wonder why not. I presume it’s because you accept these failures do not mean that ALL technology or ALL projects should be studiously avoided.

      Rather than keep reading “where is the evidence:, I would be more interested in seeing positive recommendations about the kind of improvement actions which organisations could or should be thinking about when looking as service quality improvements and cost reductions.


  10. Howard Clark says:


    Highlighting the absence of evidence is a highly positive act.

    This is particularly true when the environment is one where lots of shared services activity is somehow translated into evidence of effectiveness, instead just what it is, evidence of activity.

    But I wouldn’t begin with either concepts in mind (technology or shared services).

    Of course, that’s not a money spinner if you sell technology (self-service or otherwise).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s