Some thoughts on SOCITM’s third party suppliers workshop

There is no such thing as a website

Websites are not as most people imagine them to be. In fact you could argue that there is no such thing as a website.

There are digital transactions.

We create a website by bundling them together so that they have a coherence, a consistent look and feel, common language, integrations.

When that happens, it does not, generally delight people. It’s how we expect things to work.

When it doesn’t happen, it’s weird, confusing and feels broken.

In local government we really struggle to deliver this consistency across our digital services.

Mind the gap

Often the break in consistency comes when you move from a digital service provided by the council directly to a service provided by a third party piece of software.

SOCITM, which tries to make local government websites better, organised an event to try to get underneath these issues. And it was largely successful in identifying the problems. It didn’t come up with the killer solution but there were examples of approaches that may help. (They were kind enough to ask me to give a short presentation opining on the importance of APIs and modular design.)

Though in a sense the solution is simple and self-evident: we should buy stuff that works properly with our other digital services. Of course if it were that simple we would already have done it.

Some approaches

Northamptonshire held an in-house summit with suppliers of third-party apps on their site. They involved some business owners (the people in services who hold relationships with the suppliers of specialist software), but on reflection they wished they’d involved all of the owners.

This is one of those ideas that once you’ve heard it, seems so obvious you wonder why you didn’t think of it yourself. But of course you didn’t. However now I’ve been told about it, I’m totally organising this as soon as possible back at base.

Rochdale have driven an extraordinary level of take-up of online benefits applications (nearly 100% of applications) by working with their third party supplier and making sure that the system worked for users.

At Herefordshire we encourage the use of APIs/web services so that we can handle the front end ourselves and guarantee consistency. This has additional spin-offs and if, as a sector, we could standardise around APIs on some of our services it could really transform the market for customer facing tech.

East Riding have strong governance, essentially they have found a way of describing what good looks like and then ensured that services can only buy systems that meet these standards.

We heard from a couple of suppliers of third party systems. They pointed out that some of the problems users experience may be down to poor implementation of their products. They also pointed to the truism that they sell what we buy.

And that’s the killer point. We can not blame suppliers. We buy what we choose to buy.

What does good look like?

Across the sector there is no real consensus on what good looks like. I think that within web teams there is an ever increasing consensus but our colleagues in other areas may not agree or, perhaps more significantly, may not understand why we think that good looks that way.

We need to find ways to bring other professionals, senior managers and elected members into our world. Because we can deliver excellent digital services.

If we agree that it matters.

Render unto Ceasar

I have stepped off the fence

I am a true convert. I have seen the light. I believe in a local GDS.

I wrote a few weeks ago a “one the one hand this, on the other hand that” type post. I was genuinely not sure. Since then I’ve been doing some reading and thinking and discussing and I’ve concluded that local government could better meet a lot of user needs with single platform approaches.

One CMS to rule them all

Let’s think about the most obvious one. One CMS for local government. Essentially what we might think of as a single website for local government. with the way into your council being

When you want information about early years provision in your local authority area wouldn’t it make sense for that to be in the same place as the information about early years provision in a neighbouring authority?

I’m all for local decision making and local distinctiveness. But, really, what do we gain by having hundreds of different websites, different colours, lay-outs, responsiveness, speeds, structures?

What if we had a shared mapping engine so that when you looked up the gritting routes on the east of the county it showed the gritting routes for the next county across, and the next, and the next. Or showed your ACTUAL nearest parks, or schools, or made it really easy to report a pothole wherever you were.

We’d still have the discretion to decide which roads were going to be gritted. But let’s all use the same platform to show that information to the public.

It’s not just the content

We can develop single platforms for services while still having local service delivery.

Let’s take waste and recycling collection. Almost every local authority does this in a slightly different way to all the others. But everyone needs to be able to report a limited range of problems: my bin (or box or bag) wasn’t collected say or find out a limited range of details (when is my bin, box or bag due to be collected).

Wherever I look now I see the same issue. Let’s pick one at random: Local Offers, the detail varies locally but the user need is the same across England. 

OK. So it’s a good idea. How do we do it?

There seem to be 2 basic options. I’m going to call them Apple and open source.

Apple is top down. A crack team of digital experts roll through local government like a whirlwind leaving behind a trail of simpler, more efficient, more effective digital services.

A simple answer would be to widen the remit of the GDS. They already know what they are doing, they do it well. But we could create a new, specially local government outfit.

With the Apple approach you get results. You get a consistent approach, user needs addressed in a common way. Sure some people’s noses get pushed out of joint but you do the right thing and the thing right.

It will be hard to do without some form of carrot (money) or stick (legal requirements).  Local authorities don’t, usually, welcome a bunch of outsiders rocking up and telling them how to do things. Even if they bring a strong track record and wield slimline technology and post-its. It runs the risk of sucking more talent out of local communities and into wherever the local GDS is based (London? Manchester? Bristol). And it runs the risk of creating an infrastructure that once built will look for reasons to keep existing.

Ultimately parliament can tell local authorities to do what they are told. It would be better if we just did it because it’s the right thing to do.

The other approach is to carry on building our own things but to share our technology, our learning and our approaches.  The more we can agree common standards and share our experience, the more effective and efficient we can be. This is the approach LocalGovDigital seems to be developing.

This is messier, not all approaches will be consistent, some councils will fail to comply. It’s more flexible though, it will allow different authorities to proceed at different speeds. It will be quicker to get started (at least for the willing. It is lighter-weight and could, if done right, strengthen the skills in local communities. It is less likely to create a structure that will want to keep renewing itself (though let’s not rule that out).

Follow the leaders

If we can do it I like the messy open source approach. But I’ll buy into the Apple approach if it looks like it’s going to happen.

Either way we need leadership in this space in this sector. We need more digital professionals to join in but we need all professions to join in.

We need directors, chief executives and politicians to be involved in this debate, to see its importance, to understand the issues…

…and to have an opinion.

A local GDS could probably add value but is it the right thing to do?

Should there be a GDS for local government?

I’m instinctively resistant to the discourse around a GDS for local government.

Of course my instant reactions to such things have much to do with what’s going on inside my head and little to do with a balanced assessment of the arguments.

I don’t much like being told what to do. I resent the narrative that I perceive within parts of Whitehall which says local government would be much better if it would just do what it was told by the civil service.

On the other hand sensible people I respect who understand and love local government are arguing for a local GDS of some sort.

The arguments in favour

The GDS is achieving impressive things in central government. is to be admired, the cloudstore is genuinely exciting. A relentless focus on user needs and iterative, agile ways of working is invigorating and depressingly innovative.

And as I recently heard from DCLG

What GDS has managed to do is de-risk technological innovation and save departments lots of money

We really some more of that need that across the public sector.

And having a single government digital service has delivered a lot of benefits for central government. They have REALLY good people working there, they back each other up, they drive each other on, they have a swagger, a confidence that they really know what they are doing. They get people’s backs up but that’s OK because they have a mandate and, crucially, they really are doing the right thing.

Who would not want this in local government?

For a much more detailed (and very interesting) argument in favour do read Ben Welby’s five (that’s right five!) posts on this.

The arguments against

Local government is not the local branch of central government.

It isn’t even, really, a thing at all. It’s 468 different things (according to the LGIU) though they have some similar responsibilities and a surprisingly consistent culture. Crucially we are controlled by politicians with their own mandates derived independently of the politicians the GDS works for.

We can (and in my case, do with gusto) copy and adapt tools published by the GDS. To go further and mandate the way local government designs and delivers services is to take another step away from local decision making. Mike Bracken is a top chap but should he really be able to over-rule the (elected) leader of a local authority?

Well up to a point Lord Copper

Actually central government tells local government how to do things quite a lot. There are inspection regimes, codes of practice, actual legislation, funding streams and bidding rounds which constrain, influence and mandate approaches across all sorts of areas of public service.

Seen against that backdrop it’s actually a bit weird that this area of work, which is so important and so codified at national level, is left to the (widely varying) discretion of local administrations.

What has the GDS ever done for you?

Let’s imagine that a GDS for local government has been created. What sort of model would really help the people who live in my county (Herefordshire). What could a local GDS do that would really add value:

1. Mandate things that need mandating.

There are some things that it might be really helpful if they were required: domain naming conventions, CMS selections, open standards. If these were required of local government we wouldn’t have to re-invent them and we could drive efficiencies.

Specifying a small menu of open source CMS platforms that local government must select from could be really transformative. It would disrupt the business model of some suppliers but we could really develop effective communities around these platforms.

2. Provide APIs for local government to hook into.

For example local authorities administer council tax and housing benefits locally and though there is some local discretion round the edges it’s really a national framework. A GDS for local government could usefully provide a service that local authorities could consume locally. There may be real benefits to be derived in social care (care passport anyone?). It wouldn’t work for all local services. A national platform for leisure services would have little value for example.

3. Properly drive improvement.

How about an inspection regime for local digital service provision? I think there could be real value here. If not that then, at least, a local GDS could work with other inspection bodies to make sure they really understand how digital can and should be transforming the services they inspect. And in either case setting frameworks which could be used for rigorous peer reviews would have value (for the councils that take this stuff seriously)

4. System leadership.

Already I find we can make progress just by saying that

this is the way the civil service does it.

How much more progress if we say

this is the way the local GDS says to do it.

We need more though.

The leaders of our organisations (political and nonpolitical) need to understand what good looks like, they need to aspire to transform services through focusing on user needs, they need to understand how much this should be costing them. They need to be able to ask for help from some people who really know what they are talking about.

And people working across local government need the skills and the tools necessary to really get hold of these new ways of thinking, not just but certainly, in digital and ICT teams.

It’s politics innit?

There is a political judgement to be made here. Ultimately local councils are accountable to councillors and if local people don’t like what the councillors have been doing they can sack them. Local government has got itself into a position where it probably does need some central intervention as a result of decisions for which those councillors are accountable. To decide that those decisions are wrong and to impose something else is, clearly, profoundly political.

And some might argue (with validity I would say) that the solution isn’t to tell local government what to do around (digital) service design. It’s to give local government more to do. We (in England certainly) live in an extremely centralised state. It is the centralised system (directly or indirectly) that brought us to this point.

Is the solution really more centralisation?


The (long term (ish)) future of local government

Breaking vows

I start every govcamp with a vow to be quieter and more reflective. This time, I say I will not pitch a session and I’ll keep quiet in sessions and listen more to others.

This typically lasts about 3 1/2 minutes. And so at #localgovcamp last Saturday I pitched a session on scenario planning for local government.

Now scenario planning is a fairly sophisticated practice undertaken by many serious organisations. It involves looking at the data we have now and developing compelling and coherent scenarios about how the world might change. It’s not saying the world will be this way. It’s saying the world could be this way if these things we see now carried on.

It is a tool to assist with planning, especially planning investments that will pay back over a long period.

Good scenario planning is thoughtful, lengthy and evidence based. Ideal in fact for a 45 minute slot at a govcamp (irony there- or is it sarcasm?)

We gave it a go.

Short or long term

I suggested that we look over a fifty year period. I find that as I get older (I’m 43) fifty years doesn’t seem like such a huge expanse of time. Looking over 50 years frees you from having to consider shorter term cyclical issues (like funding cuts following a financial crisis) and to consider bigger, wider changes. On the other hand technology drives change at such a rate that it might be unhelpful to try to look that far into the future.

We settled on a 10-20 year horizon.


The first trend that the group identified which is obviously at the forefront of local government thinking was the ageing population in the UK. People are getting older, potentially staying healthier older but that then means that carers are ageing too. Potentially, over our 20 year timescale the people who are ageing will have less money to contribute to their own care.

This is likely to increase and change demand on local government, other public services and the wider community.

But it may also widen something at least some of us perceived which is intergenerational inequity as a greater proportion of local resources are focused on older people it inevitably follows that these resources are taken from younger people.


The subject of intergenerational inequity dominated our discussion for quite a while. There is a coherent narrative that says:

as public resources are concentrated on older people (who are more likely to vote) and away from younger people who are less likely to vote this will create a vicious circle whereby local government will tend to more and more focus on the needs and interests of older people at the expense of local people.

Housing was raised in the same context. You can argue that the lack of availability of affordable housing in many parts of the UK is a process whereby older people are becoming more wealthy because they own property funded by young people who do not own property.

We didn’t see it likely that access to affordable decent housing would be radically improved over our twenty year window.

It’s also worth stressing that we were trying to develop coherent narratives, we didn’t make judgements (well most of us didn’t) or try to find solutions. We only had 45 minutes.

Devolution and structures

Will local government be reorganised over our 20year window?

If past performance is any guide then that would be a definite yes. Personally I’ve worked through reorganisation in local government in Wales in the early 90s, Herefordshire in ’97/98 and Shropshire in 08/9. I’m not THAT old.

Whatever the result of the independence referendum in Scotland this year devolution to Wales, Scotland, Northern Ireland and London seems to be a one-way street. There was a consensus in our workshop that England will be affected by this though it’s not clear exactly how. We settled on a concept of “differential devolution” which is inline with devolution so far. So powers and expenditure might be released to different “city regions” as they make separate cases with different governance models. We thought that was easier to see than an English Parliament or regional assemblies.

I tried to argue for a scenario in which there is no local government: where social care is handed to the NHS, highways and regulation handed to national agencies and education and child protection handed to local children’s commissioners. And it was pointed out to me that children’s commissioners would, in fact, be local government (as police and crime commissioners are). So we developed a narrative that suggests that local democratic oversight is likely to be persistent. Though the specific role and structure will change. Local bodies seem increasingly likely to be scrutinising, influencing, maybe commissioning rather than doing.

Network society

This discussion seemed to link very closely with a workshop Catherine Howe had pitched about different forms of democracy. I didn’t attend that workshop so I’ll need to look to others to join this up.

This is all very well but…

I know. Scenario planning might be an diverting way to pass an hour at an unconference but does it have application in the real world?

I would argue that it does. When we plan, when we invest, when we draw up strategies and try to do the stuff we used to call Place Shaping, we are making assumptions about how the world will be. All too often these assumptions are tacit. Scenario planning helps, if nothing else, to make those assumptions explicit. That in and of itself is of benefit. It could even lead us to (whisper it) make better investments and lay more effective plans.

And we don’t have to do that much work. Lots of people do scenario planning. Many of them publish it. So at the very least we could just look at the work others have done and ask how our plans would look in the world they are imagining.

I’m going to try to compile a list of scenario planning resources here (or somewhere else if there’s a better place). Any links, let me know.

Magic fairy dust

A confession

I like the Agile approach to project management,

There I’ve said it.

I’m trying to resist the zealotry of the new(ish) convert but there is a programme I’m involved in where my colleagues have started timing how long I can go without starting a sentence with “In Agile…”

Not long typically.

But I also like the waterfall approach to project management.

PRINCE2 is my preferred poison.

Which is something very few people confess to.

And I also like the IEM approach to project management which is not often described as project management at all. (That’s Integrated Emergency Management, it’s how we co-ordinate efforts to look after people in emergencies and try to return everything to normal quickly).

And I have come to the conclusion that there is no magic fairy dust.

There is no single way in which projects (or change management processes or whatever it is that you want to do) can or should be safely delivered.

There may be the right tool for the right job

1 Where you know what you are doing

Projects that are delivering in a well understood environment are probably best run in a waterfall framework. So the comms work necessary to deliver a change in bin collection services if probably best planned using products and managed using ganntt charts. It benefits from lots of planning up front and then just getting on a doing.

2 Where you don’t know what you are doing

Projects that are delivering completely new products can’t be safely delivered in waterfall. If you have no idea what the new product looks like then planning is largely meaningless. Agile provides a much better framework for this. Transformation or change processes should look much more closely at Agile IMHO

3 Where you want to start delivering straight away

If you are under pressure to start showing some results (or at least progress) immediately then you, probably, don’t want to use waterfall. I can see ways in which quick delivery could be crowbarred into waterfall but frankly it’s not what it’s for. Agile is exactly the right approach for jumping in quickly and then changing your maind.

4 Where you want a big bang

If you’re delivering something that will be a big bang at some point in the future. Like, for example, a football world cup tournament, then Agile is probably not the right approach. This is, in fact, the key use case for waterfall. Except because Agile is timeboxed it can be suitable for delivering something on a defined date, as long as you aren’t to worried about the scope of what you get on that date.

5 Where you want to co-ordinate lots of organisations

Multi-agency projects can be a real challenge. One way to handle this is to create a financial or contractual relationship, so that one organisation buys things from the other organisations. If you CAN do that then waterfall or Agile as appropriate will probably work well.

If you can’t do that then neither will. Agile adherents would probably argue that Agile has more to offer because of the emphasis on teamwork and visibility and I would tend to agree.

But I think Integrated Emergency Management has a lot to tell us here. Essentially IEM emphasises teamwork and communication (like Agile) and rapid delivery (like Agile). But unlike Agile it provides a framework for scaling projects and involving all levels of the organisations whilst keeping a focus on delivering the work at the coalface.

Enough already

This has been rambling on to long enough now. I might return to this IEM theme at a later date.

Or my butterfly mind might take me elsewhere completely.

Social media in interoperability

Exercise Forward Defensive 2012  (41)

Over on the Open Eye Communications blog, Mike Alderson has been raising questions about a new(ish) publication: “Joint Doctrine: the interoperability framework” which has been published by the grandly titled “Joint Emergency Services Interoperability Programme”. Essentially it’s a manual that makes it clear what all the blue light services should be doing at an emergency.

The foreword explains it like this

“This guidance focuses on police, fire and ambulance interoperability in the early stages of the response to a major or complex incident. Its purpose is to provide emergency service commanders with a framework to enable them to respond together as effectively as possible.”

It seems to me (with the disadvantage of never having to be an emergency service commander) to be clear and sensible. It eschews the use of Bronze, Silver and Gold (terms that are applied in variable ways within organisations) to talk about Operational, Tactical and Strategic. This is the form used in Scotland and if we are going to adopt it in the rest of the UK I shall do a happy dance.

It emphasises the need for things like common terminology between emergency services. Progress has been made in this regard in recent years but we’ve really got to keep pushing. A fundamental principle of safety critical communication is clear and common terminology, this is just as important between tactical commanders as it is between an ambulance crew and the control centre.

So good stuff and should be read across category one responders.

But what Mike was pondering was the issue of social media and where this sits in the framework. The document does mention social media (which is a positive development) but it places it very much in the context of a transmit channel.

“An option may include deploying resources, briefing the public (mainstream and social media) or developing a contingency or emergency plan.”

(Section 2.5). This is included in part of the guidance on how agencies should jointly plan.

And a specific task for the tactical commanders is given as

“Provide accurate and timely information to inform and protect communities, working with
the media and utilising social media through a multi-agency approach”

(Annexe D)

I don’t think anyone is going to disagree with these points. Except to say that briefing the public is probably not an option to be considered but a task to be allocated. It reflects, I think, a sense of where senior figures in blue lights are: they get that social media is important and they see it as an important aspect of warning and informing the public. They are not wrong, except that it’s not just a tool for telling people things.

Mike adds an additional layer which is using social media as a source of intelligence. This seems to me to be vital and non-trivial.

Services may gain direct intelligence about the incident from social networks (though processing this data into usable information in real time may be a significant challenge).

The public also gains intelligence from social networks. They do this whether or not the emergency responders choose to brief them. The crowd can often create good quality information pictures but it can get it horribly wrong and it can be influenced by rumour and malicious uses.

Mike asks

“So does the commander (at whichever level) need a social media monitor at their side or someone more engaged, that can not only provide the intelligence, but who can respond, engage, advise and reassure”?

This work needs to be done and needs to be factored into operational, tactical and strategic plans. Does this mean an army of social media gurus being recruited? Probably not.

I’ve seen good examples, especially within the police service, of operational staff fulfilling these roles very effectively. A challenge for the future is how we deal with the large volumes of traffic and how we ensure that all agencies are comfortable with what is being mined from and published into online spaces.

We need to consider how resources are deployed. At present we would split the role described above between an intelligence cell and a media cell. There’s no reason why this could be combined into a comms and analysis cell.

Maybe that could be factored in to a future edition of this framework?

Image is Exercise Forward Defensive 2012 (41) by kenjonbro used under CC BY-NC-SA 2.0