Main content

The G-Cloud’s procurement problem - old habits dying hard

Derek du Preez Profile picture for user ddpreez April 26, 2016
Although the framework is a success in many ways, it doesn’t meant that more couldn't be done to drive the G-Cloud further into public sector organisations.

I’m going to begin this by saying that I am a big advocate of the G-Cloud (or the Digital Marketplace, as it is now called). I remember meeting with Chris Chant, one of the early founders of the framework, in a coffee shop just off Trafalgar Square when the G-Cloud was nothing much more than an experiment, but one that I totally understood the importance of - given that we were living in a world of lengthy procurements, crap deals and SIs that continually didn’t deliver.

The G-Cloud gave every supplier an equal opportunity to get access to the whole of the public sector. And equally it gave every buyer a window into a whole world of new suppliers that they might not have known even existed.

It changed the conversation, it has arguably democratised technology buying in government and it made ‘cloud’ a less scary thing for a lot of buyers.

Three short years in and we are now at over £1 billion spend going through the framework.

That being said, the G-Cloud should not be beyond reproach or criticism. As with anything, despite successes, there is always room for improvement.

And this week it was brought to my attention that there could well be a procurement problem surrounding the G-Cloud that is limiting its future successes. A procurement problem that largely sits with the buyers of cloud services in government, where old habits are dying hard and they continue to drive complexity into a process that should be simpler.

The two year clause

This article was prompted by a separate article published this week by my colleague Stuart Lauchlan, in which local government organisations branded the G-Cloud’s two year contract clause the “biggest single barrier to adoption”.

The G-Cloud stipulates that buyers can only sign a two year deal with a supplier. The idea being that technology changes so quickly that it doesn’t make sense to sign a deal for any longer period, plus it puts some of the power into the buyer’s hands with regards to keeping its suppliers innovative.

Traditionally 10 year+ outsourcing deals have meant that buyers get attention from their supplier for the first couple of years and then again at the end of the contract when the supplier is trying to renew - but they go for huge stretches of time dealing with the status-quo.

However, this two-year clause has got people riled up and it’s a discussion that has been rumbling on for a couple of years now. ICT director for Derby City Council, Nick O’Reilly, said:

I don’t want to change my social care system more than once every five years. Every time we change it, it takes a long time and there’s a lot of cost per change. So for some of the large business applications software, two years is just not tenable. There’s no leverage to persuade my business colleagues to do that. That’s where any framework that has a two year limit is going to fail in what it’s trying to achieve.

Is he right? To some degree, yes. However, upon promoting Stuart’s story on my various social feeds I got an incredible amount of feedback from suppliers saying that they are in favour of the two year clause - and it’s the buyers that don’t understand that they don’t need to do a lengthy procurement every 24 months, but rather just reevaluate whether or not the service is still suiting their needs.

PR it or kill it

This comes down to the crux of the problem. It is apparent that the Government

Digital Service and the Crown Commercial Service have done an incredibly poor job of communicating that the simplicity of the G-Cloud framework doesn't require that buyers need to add in their own complexity.

Old habits dying hard. In days gone by, the OJEU procurements were so stringent, lengthy and complex that the simplicity of the G-Cloud buyers’ guide put out by CCS feels too good to be true. It seems that some buyers are adding in this complexity when buying off of the G-Cloud, because in the past if things weren’t that complex, they faced lawsuits.

However, the whole point of the G-Cloud is that CCS and GDS go through the painful OJEU process centrally, saving buyers across government that pain. All the buyers’ guide requires buyers to do is:

  • Prepare your requirements – a list of ‘must-haves’ and ‘wants’ – and get approval to buy what you need.
  • Search for services on the Digital Marketplace using keywords.
  • Review services and apply the filters to create your shortlist.
  • Evaluate your options to find the cheapest or best value for money.
  • Choose your service, award and sign the call-off agreement.
  • Complete the Crown Commercial Service (CCS) benefits form.

According to some sources, this whole process can take just 24 hours if you know what you’re doing. If you had to do this every two years, is that too much of a chore? I’d argue not.

However, if you were running a full OJEU every two years, spending days/weeks interviewing suppliers and then migrating away from your current system, then yes, every two years is a problem.

But that’s not the point. The point of the two years is to give buyers the option. To allow them to reflect on their current situation, put pressure on their supplier to perform and give them a get out clause if needs be. Are those not good things?

The problem, in my view, is that GDS and CCS haven’t done a good enough job of PR-ing and explaining that G-Cloud has simplified the procurement process. It needs to share examples of best practice across government as a whole, or otherwise people are going to continue to use this two year clause as an excuse.

You can’t build it and expect people to come and understand how it works. As one of the commenters on Stuart’s story yesterday noted:

As per the article my interpretation is that the 2 year limit is just there to promote re-evaluation – if the business rationale is there (and can be loosely evidenced) and there are no alternatives that are more viable (considering all the points about off-boarding, and on-boarding a new product) then another 2 year renewal is a perfectly acceptable course of action. Clearly there are procurement overheads to managing this process every 2 years, but perhaps this is the price that needs to be paid to ensure emerging opportunities aren’t missed.

I’m sure clear information is out there with regards to how to procure through G-Cloud. For whatever reason I don’t think it is finding its way (or being found when required) in all circumstances. Maybe CCS could do more to promote the simplicities – an unambiguous ‘demystifying procurement through G-Cloud’ campaign would go a long way here. In turn this would help both supply and buy side to agree and enter into contracts via G-Cloud that are fit for purpose, and continue to further the government’s procurement strategy.


I think it’s worth sharing some of the feedback I got yesterday from suppliers here. It’s worth noting that the only bit of negative feedback I had against the two year contracts was from a large supplier - Hewlett Packard Enterprise. Surprised? I’d argue it works in the large suppliers’ favour to keep buyers thinking lengthy deals are the way to go.

Here are some of the other comments I got. Alex Blandford from digital agency dxw said that his experience has been that local government has overcomplicated the process for itself. He said:

When you do a local gov procurement on G-Cloud, in general, you do the same amount of paperwork as you would do for a full OJEU compliant procurement. You are asked about 14 pages of questions, many already covered in the original framework (information assurance in hosting, public liability insurance etc.) and you are incredibly unlikely to have an interview because they've got the old procurement mindset of blind competition, which doesn't work with what G-Cloud is. For an SME, bidding on this is 2-3 days of paperwork.

So, if localgov wants to do longer contracts, I would suggest that a) they still don't get how quickly digital changes and that you need to refine scope, spec and pricing every two years (at least) to achieve savings b) they don't do the procurement with the minimum possible fuss so it is a complete arse-ache for them and so obviously they don't want to do it regularly.

He agreed that CCS hasn’t done enough to share best practice:

Others soon added that their experience is that the simplicity of G-Cloud isn’t getting through for many buyers:

Richard Godfrey, former IT director at Peterborough City Council, commented on a LinkedIn post of mine:

I regularly speak to councils who do their own procurement exercises so that they have full control over the T&C's! Procurement is one of the biggest barriers to change in the public sector. At Peterborough G Cloud was always our preferred option, although interesting to see if that continues!

Meanwhile, Toby Gavin from Mantis PR, which specialises in public sector technology, commented:

I speak to a ton of local government organisations regularly about technology procurement. When I ask about G-cloud and why they did not procure through G-cloud they often cite the two year clause as a real stumbling block. CCS (and GDS for that matter) need to do a better job at communicating.

There is a good article here from Ian Fishwick over at Innopsis - the trade association for the public sector. He talks about how buyers need to be better sellers:

We regularly see purchasing teams trying to work out how to promote their frameworks when they have no budget to do so. This seems madness to industry. We would never dream of spending a lot of money on product development and then allocating no money to trying to sell that product once we have it. Promoting a framework should be a standard part of the process with budgets allocated up front and stated in the tender documentation. Every framework should have a clear (relatively standardised) marketing plan stated within the tender documentation."

Whilst he is making a slightly different point the main premise of his argument is still applicable in this instance - frameworks (G-cloud) and some of the more intricate points need to be better marketed and communicated. Using blog.gds as the main communication channel is not going to cut it.

My take

g-cloud digital marketplace
GDS and CCS need to do better at this. It’s been over three years and people are still confused about what they need to do. Some sources have told me that some departments that are using G-Cloud have actually increased the size of the their commercial/procurement teams!

That’s not necessarily their fault. You have to drill the message home. And that can only be done with an extensive engagement exercise and campaign, which I’d argue CCS and GDS haven’t properly executed.

Some reading this article will point to the fact that there has been over £1 billion spend through G-Cloud. And that it now has been used by over 1000 buyers in the public sector.

However, it’s also worth noting that over half that spend is from just 10 buyers. And that over three quarters of the spend is on procuring ‘specialist services’ (skills/agile) - rather than pure cloud services.

In other words, there is room for taking G-Cloud further and the complaints from buyers shouldn’t be ignored. That doesn’t mean that we should scrap the two year clause, but rather we need to think about why it isn’t working.

A grey colored placeholder image