Hot Topic: Service Automation

For how much longer will people deliver IT services? There is a developing science analysing IT services which needs to be considered when making a business plan.

The human function in IT services - consulting, analysis, system integration and so on - is a craft industry. The individuals involved bring together skills, training and experience to adapt methods, techniques and approaches to meet the needs of specific situations. They use tools to make their work more productive, and they may integrate standard components into their work. Many projects look broadly similar, but, examine the content in detail, and each one is unique.

In 1900, automotive production was a craft industry. Over the next couple of decades, Henry Ford and Alfred Sloan were widely credited as the driving forces behind detailed analysis of the car manufacturing process. Their work identified well defined steps that could be selected for automation, and explained how these steps fitted in to the overall process. Car manufacturing became a rigorously systematic and highly automated activity.

Could the same happen to IT services? If so, an investor's view of IT service companies will be transformed. A critical barrier to growth - the ability to recruit the right people - will evaporate. A critical risk - people are expensive to maintain when their time has not been sold - will vanish. Management teams in service companies will invest to develop these capabilities. Of course there will always be a limit, therefore there will be demand for people who can go beyond the limit. The Formula 1 automotive industry is still to some extent a craft. However, as automation capabilities grow, the service provider groups in IT companies will need to adjust their skills and resources.

There have always been 'methodologies' in IT - typically a connected set of small steps. From the buyer's point of view, the methodology is a critical piece of evidence that the provider understands how to do the job required. But, as with car manufacture, most methodologies contain at least one element that looks ready for automation. Service organisations will correctly point out that they continuously adjust their methodologies as they integrate new tools into their processes, and that this use of tools is a tangible example of growth of automation. This is true, but there is an important distinction between the efficiency improvements possible with new tools, and the new business model required when people are no longer needed.

First-to-market with a fully automated offer has a unique opportunity to differentiate their offer, and find new buyers with new budgets.

Consider a hypothetical case in application integration. Even today, there is no longer a need for a coachload of consultants to rebuild code. Instead, there are tools that enable database applications to be nudged step-by-step to the level of interoperability needed for efficient operation in today's environment. These tools are used by IT professionals as part of a service bought by IT professionals on behalf of their users. The state of technology means that the human element in all of this is vital - the mapping of 'equivalent' data items between two applications may be trivial in one case and spectacularly complex in another; and a person's judgement is needed to determine which cases are worth doing.

But imagine if, at some point in the future, both cases - the simple and the complex - were routinely handled by a new piece of software. Yes, of course this is a pipedream. But it would mean that users would totally dominate the buying decision. The IT group in the user organisation would, as always, retain a compatibility veto, but they would no longer be the key source of assessment of the service provider's skills and value.

Perhaps automatic application integration is a hard example to believe. But there are technologies knocking on the door. For example, web services are visible to many users through a desktop offer to "find an application suitable for this file-type". Business process modelling systems can adjust workflow by adjusting a diagram. Infrastructure virtualisation allows configuration of servers, creation of network connections and so on using point and click interfaces.

So product planners in service delivery companies have a real opportunity. Pick the service component which your company can automate and be first in that field. Assign resources to the tools that are used in these areas. Find the people who really understand, get them to define the roadmap for automation, and do it. In parallel, guide marketing initiatives and sales actions to address the new buyers of the automated 'service'.

It is a dream for many service companies to develop products and enjoy the higher margins they offer. In practice this transformation has proved to be a serious management challenge. But for those that succeed, the rewards are generous. So be quick to enjoy a competitive advantage - it won't last long!

A version of this article was first published in the IT section of the Financial Times website, FT-IT Subscription site.

Peter Thorne

back to top

Also in this issue . . . .

Cambashi researches best practice and assists IT suppliers in best practice implementation. For more information on Cambashi services please email info@cambashi.com

To subscribe: send an email with the word "subscribe" in the subject line to : ezine@cambashi.com

© Copyright 2005 Cambashi Ltd

back to top