Public Services

Digital transformation in government: Reasons to be cheerful, one, two three..

Published 10 May 2017

When it comes to public sector digital transformation, there are reasons for optimism, despite the enormous economic and structural challenges faced, argues Gary Barnett, GlobalData's head of enterprise advisory

By Gary Barnett

Last month, I attended the Government Computing conference, and shared the keynote with my colleague, and chief analyst for Global Data Public Sector, Jessica Figueras.

Jessica opened by pointing out the challenges faced in the public sector, and then handed over to me to deliver the good news about digital transformation

The presentation prompted a lot of discussion, so I'm sharing some of the key points here; I'll be writing about this topic at more length in the future.

I started by highlighting three lessons we've learned about Digital Transformation; these aren't the only three, nor are they necessarily the first three - although I do believe they all belong at, or close to, the top of the list.

Lesson number 1 : There's no such thing as magic

The first relates to our belief in "magic". I've talked about this before, but it's a persistent issue. When it comes to leadership and change (and this is especially true when technology comes into play) we have an almost childlike belief in the magical properties of the latest new idea or thing.


(If you want to reuse the image you're welcome to - it's licensed under CC BY SA 4.0)

The unicorn pic was inspired by an engagement with a large UK public sector body, and its point is that the latest "thing" may look like a unicorn, but if you really expect a unicorn to rock up, you're going to be disappointed. But, despite its sheepish look, the Donkey is pretty handy little beast; first it has a particular advantage over unicorns, by dint of actually existing, and second donkeys perform useful work (and have been doing so for thousands of years) all over the world.

Whatever the latest magic bullet, we need to avoid giving it magical capabilities; when you're offered a gift-unicorn, the very first thing you should do is grab it by the nose and take a good look.

Lesson number 2: Picking the low hanging fruit teaches you nothing about how to pick the high hanging fruit

It's not uncommon to hear people talk about "going for the low hanging fruit", but there's a two-fold danger here; the first is that doing the easy stuff doesn't actually prepare you for tackling the more difficult stuff, and second, once you've trained your customers to expect a fabulous cadence of delivery it's a bit awkward when you have to go to them and explain that now the low-hanging fruit is all gone, they're going to have to wait a long time for the next improvement.

Digital transformation initiatives that shy away from the hard stuff are simply postponing pain and disappointment. Of course, it's important to create a cadence of delivery; we need to create trust and establish a track-record, but for every "low hanging" challenge you take on, you should be looking at kicking off at least one "high-hanging" challenge at the same time.

Refacing a website is low-hanging, redesigning a process from end-to-end is gnarly, and while the former will be delivered quickly, and is likely to provoke much delight, it's the latter initiative that is most likely to deliver sustainable transformation.

Lesson number 3: There is no "off switch" for Legacy

This lesson was inspired by a casual remark, made at a conference by a Digital Transformation Guru, that we should "just switch the legacy off". As if there was a magical button that we simply press to take away all that crusty gnarliness.

As if that remark wasn't irritating enough, what stuck me was that while half of the audience was, like me, rolling its eyes, a good proportion of the people there had seized their notebooks and were, presumably, writing "must turn legacy off".

The crucial thing about many of these legacy applications, is that they work. They may not be particularly pretty, easy to use, or "agile" in the context of today's pace of change. But legacy is a synonym for "inheritance"; there's a lot of goodness in many of these legacy applications, and while legacy modernisation and transformation is most certainly in the "high hanging fruit" category, it represents a challenge we have to have the courage to address.


(this one is by Cjp24 (Own work) CC BY-SA 3.0, via Wikimedia Commons)

So what should we do about it?

Of course the list of lessons learned is longer than that, but I only had 15 minutes... and I wanted to get to the solution, rather than dwelling too much on the problem.

Action number 1: Embrace Enterprise Architecture

By "Enterprise Architecture" I'm not talking about all of the lovely detailed technical work that the chaps in IT do; Enterprise Architecture is, and the clue is in the name, something different from "IT Architecture" - It's about what the organisation wants to achieve, and sets out how the organisation is going to meet those goals.

This is enterprise architecture:


(this pic was assembled from random clipart - it's public domain)

This image is one I use when running workshops on Enterprise Architecture (for either practitioners or business people) and the trick question is "which one of these is Enterprise Architecture". The correct answer being "its all of them".

Enterprise architecture is the combination of three things; it begins with the overall mission of the organisation; as suggested by the picture of the house. This layer is about what the organisation "is", it sets out the context within which the organisation has to function, and should mean something to the people that work within the organisation, and who interact with it. This cannot be the domain of IT; because IT doesn't define the strategy for the organisation.

The next layer, the floor plan, represents what the organisation does - this encapsulates the business processes of the organisation. Again, it's really not up to IT to define these processes; this is about business change and has to be something that the business participates in, rather than being something that is "done to it".

The last layer - the blueprints is where Enterprise Architecture gets stuck all too often. This is certainly the domain of IT in its role as enabler of change and service delivery. And yes, here's where we get to argue about cloud providers, and debate the merits of different authentication mechanisms. But this layer alone is fundamentally not "Enterprise Architecture".

The eagle eyed reader will have noticed that the three pictures above, all relate to entirely different buildings, and the experienced reader will be feeling a sense of déjà vu (those accents took a while).

The job of the architect is to ensure that these views of the system (or building in the bricks and mortar world) are synchronised - and this doesn't happen by magic, nor is it an automatic by-product of using an expensive modelling tool. This happens through effective leadership, proper engagement with the business, and (perhaps most important of all) a clear mandate from the top of the organisation.

One of the discussions this prompted related to the tendency of IT to "battle on" with the blueprints despite the other two layers being unclear or poorly described. This is one of those groundhog errors that we as a profession seem utterly incapable of breaking away from. Without leadership and direction, digital transformation will always disappoint. If the key stakeholders stop coming to the meetings, it's time to be brave and stop what you're doing. Only unhappiness will follow if you don't

Action number 2: Eat your greens before moving on to dessert

This is a theme I return to constantly, and I've used the graphic below countless times; I fully expect it to see me to my retirement.

Yes, there are lots of exciting and interesting new "things" out there, but those things provide relatively poor sustenance if they're not preceded by a healthy serving of the stuff we know is good for us.

Very few people leap out of bed anxious to get to work to get on with that super-awesome data quality project they've been working on for months.

Very few "digital transformation" prizes have been given for master data management, but these tedious and difficult things form a cohort of stuff we have to do if we're going to raise our digital fitness to the point where we can deliver lasting, sustainable transformation.

(I took this pic, and you're welcome to use it under the CC BY SA 4.0 license)

Action 3: Embrace the right technologies, and use them properly

Yes, despite waxing on about technology not being magical, it does have a role to play. But again; having a silver bullet in your revolver is not going to slay the vampire if you persist in pointing the revolver at your own foot. Equally, if we were stuck on a glacier, and you handed me an ice axe, I can assure you that none of us would be any safer.

I would never have guessed that there were seven things worth knowing about an ice axe.

But I do know that I would probably be wise to acquire some "ice axe wielding" skills before confidently declaring that no-one need worry, and that I have it all in hand.


This is one of the reasons I sometimes despair when I'm involved in technology selection and comparison projects; I can't help thinking that the fact that "Product A" is three millimetres to the right of "Product B" on an analyst's two by two is unlikely to make any difference at all to the outcome of the project.

There is even a voice in my head that sometimes says "it doesn't matter which product is chosen, the project is going to fail for a hundred other reasons".

No... this doesn't "solve" the problem of digital transformation, but I hope it might provoke more discussion

This piece doesn't set out to provide the answer, an bear in mind this presentation was 15 minutes long!

But, if it provokes discussion, debate, even violent disagreement, it'll have been worth all of the image re-sizing.


Post a comment

Comments may be moderated for spam, obscenities or defamation.

We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.