Wrote this article a while back, it was published in Computer Headline magazine, thought it was worth a repost here.

Enterprise System Adoption and Scale Free Networks

It is now well established that the return on investment from an enterprise deployment is in no small way linked to the level of end-user adoption of the system itself. Change is never easy, and many implementations have been stalled because users did not adopt the new technology and processes necessary to make them work. New and increasingly complex applications promise great returns with their richness of function and interface. Unfortunately this can also leave the end users perceiving that, compared to the existing legacy systems, the new systems have more screens, more data to enter and generally make their job harder than it was before. This inevitably leads to resistance on the part of the users.

The traditional approaches to combating this resistance involve communications programmes that precede and ultimately integrate with the end-user training programme. A lot of money is often spent on conferences, user forums, newsletters and the like to ensure that the user understands and accepts the need for change and hopefully the benefits that the new system will provide. Do these approaches work? The desired ROI requires much more than the occasional convert or, at best, widespread apathy. If companies are going to achieve the excitement and support that they want from their end users, it is clear that some new thinking is required.

Now for the science. Recent scientific discoveries have shown that various complex systems have an underlying structure that is controlled by shared organising principles. These so-called Scale-Free Networks are made up of an uneven distribution of connections. Nodes of these networks do not have a random a pattern of connections; instead, some nodes act as hubs with many connections. It is the way that these hubs work that dramatically change the way such networks operate.

This research has shown that Scale-Free networks can be found in a variety of common situations. For example, the mechanics of the spread of a disease and the development of the Internet itself have been shown to obey the laws of Scale-Free networks. Further research has shown that similar rules apply to social networks and communities. It is clear that the end-user population of an enterprise system is a similar sort of social network and perhaps by understanding its characteristics, we can best influence the adoption of the system.

The keys to a Scale-Free network are the hubs. In the case of an epidemic, these would be the carriers – the few people who come into contact with many, many more, spreading the disease as they go. In relation to the Internet, the hubs are the search engines e.g. Google. We can easily see how the treatment of such hubs can have a dramatic effect on the whole network. Cure the disease carrier and the epidemic dies; bring down Google and much of the Internet becomes inaccessible.

In the case of end-user adoption in a large user community, the hubs are the key influencers in the organisation. These are the people that, by the natural force of their will or character, set the agenda for those around them. In any large group of people, there are always significant numbers waiting to find out what their opinion should be on a subject – and that opinion is, in many cases, set by a colleague or peer. These people are the hubs, the carriers and the ‘disease’ they can carry is opinion. The crucial point is that until this opinion is set, it is not yet determined if it will have a positive or negative effect on the network as a whole.

The science of Scale-Free Networks clearly shows that these individuals, the hubs, hold the key to the rapid diffusion of opinion or information. If they are detractors, then overall adoption will significantly decrease, if they are advocates then the reverse is true.

If you believe in the theory then this sets an intriguing and difficult organisational challenge. Traditional communication in an enterprise will usually follow a top-down, hierarchical flow manager to subordinate down the chain. Alternatively it could be a “tell one, tell all” strategy of mass communication. The theory says that neither of these will have the same positive effect as identifying those few individuals in the organisation to whom everyone else listens. These are not usually the managers, who can easily be eyed with suspicion, but the influencers.

The corollary of this is, of course, that most of the money you are spending on the communicating with everyone is being wasted on the majority who will only ever be influenced by the diffusion of opinion from the hub minority. Look around your office today and you probably know who they are, focus your attention on them; win them round, show them why the new system is of benefit to all and the rest will follow. They have to – it is science after all.

Further Reading: http://en.wikipedia.org/wiki/Scale-free_network