The Government CloudStore – a good thing, but let’s not get carried away

Pinsent Masons are a law firm with a lot of experience in the UK public sector and, in my experience, good and sensible people. But a recent article by one of their lawyers contained some highly debatable arguments as he sought to paint a glowing picture of the  Cabinet Office’s CloudStore (G-Cloud) supplier catalogue / procurement process, which is just over a year old now.

The CloudStore is innovative in terms of the way it has provided a much easier application process for suppliers.  It is basically a framework contract - subdivided into Lots - but with no restrictions on the number of participants, and with very low barriers to entry. That has led to hundreds of suppliers gaining a place, and an opening up of the public sector market to those (often smaller) firms. So far, so good.

But... it still has to be used as a conventional framework under EU regulations. That means, if it is not clear which provider of a particular service offers best value for money, then a mini-competition must be run. So you can't just pick and choose who you want from the list (that would have been innovative, but also illegal).

However, in this article, David Isaac says "Government departments are now mandated to use the G-Cloud framework where they possibly can...” But it’s not mandated in may real sense. The wording I’ve seen used by Cabinet Office is no more than "you should consider..." That’s not a mandate.

Isaac also says "Using G-Cloud more will help deliver economies of scale and makes financial sense during times of austerity.  But it is very clear that there are no economies of scale in the way the CloudStore is set up – and that’s a common problem with frameworks, as everyone from Spend Matters to Ian Watmore to Bill Crothers have identified at one time or another. Because frameworks do not offer the suppliers any committed volume, then they do not obtain the best value. Why should a supplier go public with their very best prices without any guarantee of business? So they don't.

Finally, the CloudStore has seen £6 million of business over the last year. Let’s just get that into perspective. It is significantly less than HMRC pays Cap Gemini every week for IT services.  The same applies in other major Departments and major suppliers.

The G-Cloud and CloudStore is a very worthwhile exercise. It has been well executed and GPS and Cabinet Office should be congratulated on their work. However, to position it as the saviour of public sector IT procurement is totally disproportionate. I ‘d rather we were honest about the issues and the low usage of the CloudStore, so we can look for improvement and avoid complacency in the face of all the actions that still need to be taken to improve public sector IT procurement.

Voices (12)

  1. Mike:

    Peter,

    To rewind to the original point, it would be difficult to argue that G-Cloud will deliver economies of scale because as you say these can’t be optimised through Framework arrangements of this type. However Government is ploughing more than one furrow. The need to develop faster, more flexible mechanisms to contract with an agile community of Cloud suppliers along with opening the Government Market particularly to SMEs remain as absolute priorities. Alongside contracting for austerity, this is a difficult trick to pull off and Cabinet Office is frank about G-Cloud in that it is work in progress. As to a ‘Mandate’ – well yes in the sense that it is a central contract for a commonly bought category but as with other central contracts, switching over is not instantaneous. But as the content and process is increasingly refined and as long as Cabinet Office retains its commitment to it, I predict G-Cloud will see a significant increase in uptake although it may be G-Cloud 5 or 6 before this really materialises. However given that G-Cloud 3 will launch around Q3 this year that won’t be too long coming.

  2. HeadintheClouds:

    I understood that mini competitions were ‘outlawed’ due to subjectivity involved in creating a short list. We understood the correct position was to invite all suppliers on the framework to bid. Quite cumbersone for the buyers to handle and giving suppliers a much smaller chance of actually winning the bid.

    My comment on mini competitions was actually relaying a point made at a Civil Service Live G-Cloud workshop this week where I was eagerly learing how the Cloudstore operated!

    I suspect we are all violently agreeing as I cannot see how you can evaluate a service charged at a day rate, which different suppliers are likely to quote different numbers of days to deliver.

  3. APTA Inc Limited:

    Peter,

    Apologies as i think i replied on the back of Dan2 whereas it should have been on the back of Headintheclouds. You response expands further on the point i am making

  4. marketdojo e-sourcing:

    We are a solution provider on the G-Cloud, having entered upon the second iteration Gii. Since we were already offering our product (e-Sourcing software) at a truly commoditised price level, we actually had to increase our pricing to be on G-Cloud!

    This is because there is a management overhead for us to submit the monthly order form (since call-offs are not handled through G-Cloud itself) as well as to cover the commission fee on any call-off charged to us by GPS.

    However, at least G-Cloud does encourage other vendors to offer a commoditised product to the market which they may not have been doing previously.

  5. HeadintheClouds:

    Peter,

    Interesting comment about running a mini competition – this is not necessary according to the offiical explanation as to how the cloudstore should be used. As you cannot negotiate price or commercial terms, you need to decide whether you are buying on the basis of MEAT or lowest price, then read the catalogue to determine which offering best meets the chosen criteria. If more detail is needed to reach this conclusion, you can ask for clarification in writing or at a meeting. Important thing is to record the rationale for your decision!

    As for savings, I think these will come from buying off the shelf goods and services rather than the bespoke software solutions we use now. However they have to be in the cloud (obviously) to use Cloudstore. This means the real game changer will have to take place in the ICT Dept not Procurement (e.g specifying, integrating and managing a whole raft of different vanilla offerings which are needed to deliver the specific Government service).

    A genuinely intersting development but not one that will save the UK economy on its own.

    1. Life:

      Headintheclouds, I think this is why it looks looser than S-CAT, where mimi comps were at least recommended? As for not negotiating on price, look at the products – most have tariff tables which enable profiling against tariff, at least. I understand that the advice given is to discuss with suppliers once you’ve decided who you’d like to talk to, prior to award…. All in all, whether considered against the theory, or the almost certain practice, isn’t this likely to raise some questions? Forgetting the EU for a moment, is this in any case good practice? Can buyers just choose, meet, and then select a single supplier, and still even under the official guidance be considered compliant? At the very least it seems like quite a departure from normal expectations.

      1. Peter Smith:

        Interesting – you’re right, headinthe clouds, that they say you don”t need a mini-comp. But then you get this –

        “In some cases your short list will not comprise of services that are functionally identical as far as your requirements are concerned, but you still need to make a fair choice among them. To do this you can look at four aspects of the services and use these to measure which service in the short list best meets your needs. You can weight these factors however you like just so long as the same weighting is applied to the whole short list in the same way. The factors are:

        1. Whole life cost

        2. Functional fit

        3. Non functional characteristics

        4. Service management

        To determine which of the services is the best fit you might need to look at Service Definitions in detail and even try the services – but you must do the same with each one on your short list. ”

        Now what does that all sound like? A mini-comp?

        I think there are vulnerabilities to challenge here, but I suspect GPS are taking that risk on the grounds that the likely reward from any challenge is minimal.

        Back to Dan’s point- I wasn’t saying that economies of scale should be a major driver, I was really just saying that I didn’t agree with the lawyer who seemed to suggest that CloudStore WAS achieving economies of scale (which it is not)., I do think it is “a good thing” however, just to be clear.

      2. life:

        Thanks Peter – I agree that the chance of challenge from a peer supplier is likely to be very slight – admit I hadn’t even thought about that! I was thinking more about intervention from the EU if the process itself was found to be ‘non-compliant’. I’m not up on the detail regarding why the S-CAT arrangements were considered unsatisfactory, if this was indeed the case, but my understanding at the time was that the move to what was then Catalist, and the new arrangements regarding having to engage with all capable bidders or a single bidder, was exactly because on S-CAT you could pick and choose suppliers off a long list and form your own mini-competition and this was deemed unsatisfactory. Catalist of course then had a much reduced supplier list for each lot to make the new ‘EU endorsed’ arrangements practical. Obviously the two arrangements are not directly comparable in all ways, but the G Cloud list is longer, and the G Cloud process does not even require the mini competition. Also, from the other side, the vetting to get on the supplier list is very light touch, deliberately to encourage SME participation, but of course anyone can apply. In the context of government procurement channels it does seem to be a unique beast!

  6. Life:

    Actually, is G Cloud legal at all?! Selection arrangements, even if used in the official manner, are looser than S-CAT arrangements, which fell foul of the EU – although is it possible that current politics might stay the EU hand?

    1. Dan2:

      I think it falls under “dynamic purchasing system” part of the regs, so if buying things like hosting space it should be fine assuming the catalogue is set up correctly and buyers award as per the original criteria.

      As for trying to bespoke stuff or pick your preferred supplier through it? Doubt that would fly…

    2. APTA Inc Limited:

      Peter,

      Interesting comment about running a mini competition’ because the GPS ’ FAQ’ buyers guide says, that you do not have to run a mini-competition. I queried this with GPS and the response was that the framework only allows for ‘Direct Award Call Offs’ to be made based on ‘Lowest Price’ or ‘MEAT’ and therefore the question of a mini competition is irrelevant.

      In order to decide if you are awarding on ‘Lowest Price’ or ‘MEAT” the buyer (supposedly) needs to read the catalogue to determine which offering best meets the chosen criteria. If more detail is needed to reach this conclusion, you should ask for clarification in writing or at a meeting and record this as part of your justification. This in my opinion is where the framework could be open to challenge and put the potential buyer at risk (particularly if they have no procurement experience) because the inability to reach a conclusion would clearly imply that the terms laid down in the framework are not precise enough to cover that particular call off.

      Therefore, In the absence of no mini competition process within the G-Cloud framework this could be a case of ‘buyer beware’ given that the Public Contract Regulations 2006 state: “In circumstances where the terms laid down in the framework are not precise enough or complete for the particular call off, a further competition (i.e. a “mini competition”) should be held with all those suppliers within the framework capable of meeting the particular need”

      The danger here is that buyers/departments cherry pick from an extensive list of suppliers rather that using the framework in the spirit and meaning of the procurement regulations.

  7. Dan2:

    Peter,

    While you make some good points about economies of scale – that is not the only lever to bring about cost reduction. In this instance, the lever the cloudstore offers is increased competition.

    This should hopefully disrupt and challenge the existing oligopoly as long as the cloudstore programme can pass out the knowledge of how to successfully buy and run the services offered.

    Anyone who has worked in that field can tell you horror stories about ridiculous pricing offered by the larger companies for basic tasks. All of their corp ohds are huge, because they don’t need to be competitive, because there is little other choice.

    I acknowledge there are lots of other issues with the cloudstore e.g. onboarding your new app or hosting environment into your desktop/helpdesk (which is probably run by an inefficient SI); how data links up between different apps; or one of the large SIs acquiring your previously agile SME; how to agree an SLA over multiple suppliers; or the evangelical statements that ‘this will save money’ without any analysis over what the £13bn Govt IT spend is actually spent on (i.e. if it turns out most of that is on enhancements to existing systems, then it won’t be influenced by the cloudstore) etc etc etc

    However, the ‘economies of scale’ criticism isn’t one of them in my view – this is about competition; which is surely the foundation of the procurement profession.

    Dan

Discuss this:

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