Thursday, 21 May 2015

GaaP – 1½ million useless public servants out the door and 35 billion quid off the deficit. What's not to like?

Promising clusters

GaaP. Government as a platform. There's a lot of chatter about it at the moment, it's got something to do with the future of public services in the UK, but what is GaaP? And why is anyone interested?

Where to start ...

... Tim O'Reilly? No.

... Public Servant of the Year ex-Guardian man Mike Bracken CBE CDO CDO, executive director of the Government Digital Service (GDS) and senior responsible owner of GOV.UK Verify (RIP)? No.

... Simon Wardley.

You get your money's worth with Mr Wardley. He's a trouper. DMossEsq has a nascent, muscular relationship with him ...

... and hats off to him, Mr Wardley does one of the best jobs on the circuit debunking management consultants. Treat yourself, now, for 13 minutes and 9 seconds:

Who says music hall is dead?

What Mr Wardley says in the video is:
  • He can predict the future. He knows what's going to happen although, unlike a proper astrologer, he can't say when ...
  • ... which means he can devise strategies for companies. What is a strategy?
  • Strategies are what army generals have. How do they make them?
  • They use maps. Does Mr Wardley have a map?
  • You bet he has. See below. Companies are value chains, value chains consist of components, competition causes components to evolve and Mr Wardley maps the value chain-evolution surface.
HS2 strategic mapping

Which brings us to Mark Thompson's What is government as a platform and how do we achieve it?.

Mr Thompson is another management consultant, like Mr Wardley, and he has worked out precisely how to achieve GaaP. He offers a "simple methodology":
This approach is based on Simon Wardley’s work, extended to enable a clear distinction between “product” and “capability” from which we then derive a simple methodology for achieving government as a platform.
We'll come back to the distinction between products and capabilities later. First, in case you thought we had taken a wrong turn and strayed out of the ballpark, note that it looks to followers of Twitter as though Mr Wardley advises GDS and Mr Thompson locates GDS at the heart of GaaP:
For central government, the logical steward for these shared capabilities is the Government Digital Service (GDS) ...

Digital business change will be achieved to the extent that, stewarded and supported by GDS, everyone can gradually maximise shared capability (platforms), and minimise departmental “product” (bespoke services) ...

Common Capabilities ... Owned centrally in GDS (central plat) ...

Functions to be migrated to the growing set of shared capabilities (stewarded centrally in GDS) ...
It's not clear how much Mr Thompson knows about GDS. He seems to be under the impression that GDS succeeded in deploying its 25 exemplar services:
GDS has already achieved much in driving digital change in government – particularly in securing some quick transformational wins via the25 targeted service exemplars.
GDS are a long way from securing "quick transformational wins" with the exemplars.

But Mr Thompson would be quite at home there when it comes to user needs, whatever they are. Like GDS, and everyone else, he thinks you have to focus on user needs.

He also thinks that user needs exist in a value chain. And that the job of Whitehall's generals is to "determine what components are required to meet those needs".

Although that includes some of the same words, it's not what Mr Wardley said. Don't confuse the two.

And it poses a problem. In Mr Thompson's value chain diagram, how do you distinguish a need from a component required to meet that need?

We've got a lot of circles with black borders, some of them are connected by black lines and some aren't, and one of them has an arrow coming out of the top. Which one is a need and which one a component required to meet a need?

What, you may ask, about the dark ellipse? Forget it. By Step 3, it's disappeared.

Meanwhile, why does the value chain axis range from "invisible" to "visible"?

Step 3 involves adding Mr Wardley's evolution axis to the diagram, thereby making it into a map.

These are the first three steps of Mr Thompson's "simple methodology for achieving government as a platform". At Step 4 we "need to apply this consistent logic to identifying whether we should be developing specific services or common capabilities".

Which brings us back to the products v. capabilities tension identified above. We need "a clear distinction between 'product' and 'capability' ...", Mr Thompson said, if you remember.

We shouldn't. You don't have any trouble distinguishing a product from a capability, do you?

The problem is exacerbated because Mr Thompson repeatedly refers to services as "products". Why? Public services, the domain of this discussion, are definitely services. Not products. What we pay for with our taxes are services. Whitehall is a service organisation. Not a factory. Ditto local government.

To be fair, that is of minor interest to Mr Thompson. It's not the services he cares about, but the exciting "ubiquitous web-based infrastructure to enable commonly shared capabilities":
Achieving more of this sort of service model will require a rigorous distinction between “product” – essentially, bespoke services - and shared capabilities, which are far more important for achieving digitally-enabled public service redesign ...

... everyone can gradually maximise shared capability (platforms), and minimise departmental “product” (bespoke services) ...

The gradual maximisation and consumption of shared capability must therefore be the primary concern and stream of work for all transformation activities across the public sector. Although very important, improving “product” (bespoke services) is a secondary concern ...
With his "clear methodology" explained, Mr Thompson moves on to the "clear operating model and enabling methodology" which is needed to "prime the pump" of the "digitisation engine". The trick is to identify the "public organisations with the highest clusters of potential shared capabilities (SC)", which means developing a "digital profile" for each public organisation.

Digital profiles look at first like Wardley maps but, instead of more or less visible value chain and evolution, this time you have to plot ubiquity against certainty, and draw an ogive on it:

The blue blobs, of course, aren't user needs. User needs are long gone. They're products, i.e. services, or capabilities.

What you're looking for is organisations with "promising clusters" (PCs).

Once you've found a couple of them, you develop a single system which they can share. At which point, the blobs stop being blue. They get greyed out. And here at last is the answer to our first question, what is Gaap. As Mr Thompson tells us, these greyed out blobs "will literally constitute the government as a platform".

It's been a bit of a slog, perhaps, but well worth it. With GaaP:
... we can expect radical disruption of the market, opening up opportunities for innovation and investment by citizens, public, private, and third sectors alike - unleashing unprecedented innovation, efficiency, and savings ...

There are enormous opportunities for rationalisation and simplification across government ...
All we have to do is ...
Identify & agree existing, vertically siloed business models for each public organisation
Convert these into digital profiles, and;
Add the segmented profiles into separately-run backlogs
... and within a generation:
This will encourage the evolution of a service architecture that is based on a service-oriented view, supports an ecosystem of providers around a core set of capabilities, and opens up access to a broad community through standard interfaces ...
... which is what we all want because:
The inescapable DNA of a digitally-enabled public service model is a set of clean, agreed, and common capabilities, distilled and evolved from the currently duplicated and siloed functions, processes, roles and even organisations that exist across government.
He doesn't like silos, does he. Silos are 19th century DNA. Strange that Mr Thompson's own business comprises a holding company (Methods Corporate Ltd) with six separate subsidiary companies, or silos, under it. Look at the duplication that must lead to. Seven audits. Seven annual returns to Companies House. Seven payrolls to operate. Seven "enormous opportunities for rationalisation and simplification". Seven blue blobs all surely pleading to be greyed out.

But that's irrelevant because, turning our attention once again to Whitehall, when it comes to GaaP ...
A rough and ready calculation suggests such an approach could save the UK £35bn each year – but the jury is still out on how best to go about making it happen.
... and £35 billion p.a. is a lot of money. There's the answer to our second question, why is anyone interested in GaaP.

How does Mr Thompson come up with this figure? Click on the link and you be the judge. It's all something to do with a Dutch community nursing agency, Buurtzorg, whose operation is unmistakably analogous to Whitehall offering public services products:
The Dutch model has a ratio of back office to frontline workers of 30 to 7000. If the UK could adjust its ratio accordingly, it would need fewer than 23,000 back office staff to support 5.3 million frontline workers. Reducing the number of back office staff from 1.5 million to just 23,000 would generate possible salary savings of up to £35.5bn.
The logic is impeccable. 1½ million useless public servants out the door and 35 billion quid off the deficit – that's GaaP.


Updated 23 May 2015

Remember Rural Payments

Writing in ElReg yesterday, GDS to handle Govt payments? What could possibly go wrong?, Andrew Orlowski says:
Be afraid. The previous government’s “elite digital team” which so brilliantly borked most of Whitehall’s websites, and that failed to meet its own targets, may be put in charge of handling real money: your money.

New Cabinet Office Minister Matthew Hancock hinted at the opportunity at a gentle public grilling today. Although Hancock didn't mention GDS by name, he told the Institute of Government:
The Government has a multitude of platforms for paying people money. And receiving money. What we don’t have is a platform all across government for receiving and paying money. Departments collect debts, for example.
Armed with the Wardley-Thompson wisdom above, what do you make of the suggestion to create a government-wide payments platform?

How is General Hancock supposed to devise a strategy without seeing where payments is on the value chain-evolution terrain? What is the payments digital profile? Is its certainty high or low? What is its ubiquity score? Where is it in relation to the ogive? How many back office staff can Whitehall get rid of? How much money will be saved? In what way will user needs be better satisfied?

And who should General Hancock put in charge of this battle? "Hancock didn't mention GDS by name", we are told. Hardly surprising. GDS is a front-end shop. They don't hold themselves out as experts in purchase ledger. And for good reason – it's not their bailiwick, GDS do user interfaces.

The general already has a procurement centre of excellence in the Crown Commercial Service:
The Crown Commercial Service (CCS) brings together policy, advice and direct buying; providing commercial services to the public sector and saving money for the taxpayer.
GDS would just embarrass themselves if they accepted this commission. So they won't accept it.

"What could possibly go wrong?", Mr Orlowski asks. Remember Rural Payments, Agile@DEFRA, GDS's Dunkirk.

"Be afraid"? There's nothing for us to be afraid of from that quarter. The whole idea is just a flight of Mr Orlowski's whimsical fancy.

Mr Thompson suggests that turning payments into a platform would unleash "unprecedented innovation, efficiency, and savings". Examples, please. Innovation such as what? What efficiencies? What savings?

General Hancock will no doubt demand a bit more detail before sanctioning the engagement. He won't want to find his troops tied down in a campaign of trench warfare for the next five years only to have nothing to show for it at the end.

Updated 26.5.15

Methods Corporate Ltd and its subsidiaries comprise constitute the Methods Group.

Mark Thompson, the man who plots ubiquity against certainty and deduces that GaaP could reduce the UK deficit by £35 billion (see post above), is the strategy director of the Methods Group.

£35 billion? Every year? Really? The chief executive of the Methods Group, Peter Rowlins, has had to step in to calm down expectations. He has written an article, GaaP: it's not just about the platform, with Mr Thompson,

UK GaaP* is now £35 billion lighter. Not a mention. It's gone. Replaced with the more vanilla "potentially colossal savings".

How does HMG lock in these colossal savings? It's our old friends standardisation and consolidation. And a new one – deverticalisation.

* Apologies for confusing any accountants who've dropped in. We're talking about Government as a Platform, not Generally Accepted Accounting Principles.

Updated 5.6.15

UK government not being transformed:
NAO finds digital not yet delivering Whitehall staff cost savings

... there is "little evidence" departments are making expected savings from digital services

A report by the influential National Audit Office (NAO) into central government staffing costs has warned that despite the extensive work going into digital services across Whitehall, those efforts have yet to deliver significant staff cost savings.

Although the government expects digital services to reduce staff costs by processing transactions efficiently and introducing more customer self-service, with departments developing and implementing digital exemplars, so far, the NAO says, "we have seen little evidence that departments are making the expected savings." ...

The NAO has previously pointed out that for digital services, particularly in relation to efforts in moving towards Government as a Platform , there must also be a convincing economic case as well as a vision behind the government's digital strategy. Although at a macro level, the potential for cost saving is high, it was "missing learning and case studies from individual service transformations."

1 comment:

Anonymous said...

How right about GDS they only do web forms but reality is “digital” and “user needs” are about business operations a skill set missing at GDS. I notice Wardley has ERPM as off the shelf in his rather complex view. Well consultants have made fortunes installing in the business world there hugely complex expensive systems that users hate so definitely should not be on any government agenda! Government is all about income and expenditure requiring reporting/control on the people supporting processes. Are you listening HMG NEVER buy ERP!

Then Mark Thomson slightly less complex diagrams mentions "need to apply this consistent logic to identifying whether we should be developing specific services or common capabilities" He is getting close but sadly like GDS fails to realise in reality there is only a need for some 13 task types to deliver any digital service – all pre-coded so just configure as required. GDS have failed at great cost to the taxpayer to investigate…misfeasance in my view but that is for later…..Oh dear Thomson then says “It's not the services he cares about, but the exciting "ubiquitous web-based infrastructure to enable commonly shared capabilities": the blind leading the blind here!

Interesting the Dutch model mentioned they are much smarter than UK as they listen to and take seriously their innovative companies! All UK Government initiatives over the past 5 years on seeking innovation failed to deliver….why? I have my views…but I wait until ”later”…..

You make a good point about what is the digital payment profile? Well actually is it based upon the outcome of business operations in effect that summary report of relevant data that authenticates a “payment” There may well be a case for a centralised payment BUT first get the front end processes digitised – not difficult ask the people who actually do the work and use next generation software that removes geek coding! Thomson has the cart before the horse and the future of GDS would look doomed but no doubt self interest will prevail…or will Matthew Hancock be brave enough seek advice from business savvy people that understand IT capabilities (very few of us!) and to ask the right questions….?

Post a Comment