The great thing about adopting cloud technology is the ease of access and the ability to dip your toe in the water to see if it works for you. It can be a department, small team or even an individual that tests if it is suitable; so no expensive procurement process or pesky capex to be approved. But when it comes to the public sector this approach can rub up against some significant cultural barriers and ways of working.
It is easy for a small technically-savvy company to try one method and then pivot to something else. But when governments pivot the euphuism used is more likely to be a U-turn and highlighted as such by opponents. A service with millions of users cannot undertake changes lightly and still run efficiently whilst meet data governance requirements.
Can the IT department beat Amazon?
One of the things cloud providers bring of course is scale; they can effectively spread their costs around the globe; and with companies keeping unit costs very low the average government department can hardly hope to compete. Why would your local hospital IT department be able to run a data centre better than Google or Amazon? Should they even be trying?
Web-based services are most successful when they have a generic purpose, such as salesforce management, whilst many functions of government provide very specialist services geographically bounded. However these services have commonality amongst them and their size doesn’t preclude them from using the cloud; and even the most specialist services can be underpinned with infrastructure as a service.
So how can the high expectations and scarce resources mix with cloud? It will take time to move from providing infrastructure and platform services to applications that go beyond replacing basic desktop functions for the public sector. It is therefore refreshing to see the government initiative to encourage this change with G-Cloud and the “cloudstore” offering over 7000 services from 800 suppliers. There are not many places you’d expect a government web site to have “Gamification” amongst the categories on offer. Its listed services range from pennies to tens of thousands, and over time it should be expected to grow to offer specialist applications for the public sector. But the most interesting thing is not the technology or even the applications - it’s the way it is being conceived and the way it challenges the traditional approach expected by the public sector.
HMRC will not be downloading cloud apps
The implications of this will take years to play out. Whilst HMRC will not be downloading an app for a new tax system anytime soon health, education, property management and the myriad of public sector functions will start to find more options than the large complex software systems that are typical today. Often these are only supplied from a small number of large suppliers which are locked in for long periods of time, offering little flexibility, and are difficult to include in open data initiatives and to integrate without substantial cost.
So are cloud services the saviour of public sector IT? No, at least not yet. Realistically there will always be a need for government IT at some level, especially for reasons of security and defence. There are other challenges the public sector faces too, such as freedom of information, and high levels of expectation around governance – especially privacy - potentially covering the whole population.
Will the clicked T&Cs be relevant in the future
There are also questions about what happens at the end of using a service; the exit clauses of agreements are not always foremost in the mind when contracts are agreed. When services are started are qualified staff really reading the terms and conditions when clicking on a new service? What will be the switching costs of changing suppliers be, how will large data sets be securely transferred? Will the existing data and backups be securely destroyed? All these questions should be asked.
Data varies in levels of confidentiality, integrity and availability therefore the impact level of the data has to be assessed too - not everything in the public sector is a state secret or very personal such as patient information.
Matching this level of detail to records or any unstructured information is a challenge, but luckily it’s a problem where third-party companies can ably assist. It is necessary as all the data has to sit on a hard disk somewhere and at the end of its life its necessary to know how it will be disposed of appropriately depending on its assigned level.
Putting data in the cloud for private or public sector raises fundamental questions about data governance. It is easy to ignore if an out-of-sight out-of-mind attitude prevails, and possibly with no IT department feeling as responsible for data which they don’t physically hold, serious consideration has to be given to data governance. The technology used may be the same but the environment and cultures are often very different.
Private sector organisations, especially the newer smaller ones, have learnt how to use the cloud effectively which is being followed by greater adoption in larger organisations. It is taking longer to become established in the public sector, not because it is intrinsically harder but because resources and cultures are making it tougher - and that change needs as much attention as any technology.
About the Author
John Culkin is Director of Information Management at Crown Records Management, a company that helps clients to maximise the value of their ‘corporate memory’ through the storage, active management and timely distribution of information assets. Previous to Crown john worked at Ricoh where he started as Senior Technology Consultant in 2006 before moving to Partner Manager in 2008.