Enterprise IT and Maintaining the Innovation Edge

Excerpt from Outlook: partly cloudy with sunny spells to follow written by myself and Dan Scarfe (publised soon)

It is pretty hard to envisage an enterprise today, that is not wholly reliant on their IT to function, and this has been true for some considerable time. This is irrespective it seems of the size or age of the enterprise concerned. Until the recent economic downturn, IT investments have seen continual and sustained growth for many successive years; responding to new technology advancements (that themselves have the promise of saving money) and new business opportunities that technologies create. However, it is interesting to note that while reliance on IT has been on the increase, this has been in parallel to the continual need to control enterprise IT costs. It makes sense that as IT investments increase, so do its costs (although they may have a net money saving effect), but in many cases it appears that costs increase significantly faster, suffocating the ability to innovate and invest further, as IT spends more and more on managing and maintaining existing assets. The problem for many is clear; that the demand to respond to business change outstrips the ability for IT to manage the increase in complexity that results. The outcome is systems that are difficult and fragile to extend, that are hard to decouple from one another and that create a complex set of interdependencies in which one seemingly innocent change can have very unexpected consequences elsewhere. In response, many have turned to the promise of Service Oriented Architecture (SOA) only to find that it has failed to deliver a significant return on investment. This is not to say that SOA is intrinsically wrong, far from it, but by failing to consider SOA as an Enterprise wide investment it has proved limited in its ability to decouple the assets of the enterprise as a whole.

Macro and large-scale IT and Business process outsourcing, once seen as a way to rid the enterprise of the cost of IT, has also come under the microscope. With many of the original deals coming up for renewal, it is apparent that the outsourcing model has not reaped the returns that were promised, creating situations in which change incurs rather than saves costs. Many are now, if able to do so, actively re-insourcing but in so doing the smart ones are being selective over what they in-source and what remains outsourced. In addition, these enterprises are looking to partner more broadly, and the notion of having multiple suppliers (or multi-sourcing) is becoming more widespread. This is coupled with the emerging trend of differentiating IT assets between those that are core and those that are not. In so doing, the enterprise is considering what IT assets it needs to invest in and which it does not. By looking at these assets as being functions or capabilities, it is possible to provide a map of the enterprise and more easily understand the implications and value of developing an Enterprise wide SOA strategy, where some elements could remain in-house, on-premises and others sourced by someone else.

As a consequence, large ERP systems also come into view. Why would one invest heavily in bespoking a business capability, that itself is not intrinsic to making the enterprise grow and succeed? By accepting that these are not central, it becomes easier to consider adopting simpler industry-based best practice solutions for delivering these capabilities, reversing the trend of mass-customisation to fit highly personalised business processes. This makes it far easier to select from a broader set of suppliers at a much reduced price. Furthermore, one can start to unbundle these over-engineered and over-customised systems that lock the enterprise into a single supplier, increasing choice and agility over the longer term.

This move supports a much wider trend across enterprises, often referred to as the “consumerisation of IT”, where innovations in software in the consumer space are adopted in the work place. This is especially widespread in web technologies and the mobile space, doubling the disruptive influence on large enterprise-grade solutions and their providers.

Although, many enterprises are struggling under the weight of their existing IT investments, there are many emerging examples of enterprises that have made appropriate strategic architectural decisions and are now leading the way for others to follow. Maybe the most striking emergent trend in IT, and more broadly in the enterprise business model itself, is the concept of “innovation outsourcing”. First pioneered in manufacturing, enterprises are outsourcing R&D to cut costs, but more importantly to exploit new markets faster. There are many obvious concerns and risks with this approach around IP retention, but a more subtle variant is the creation of innovation or incubation centres within the safe harbour of the enterprise itself. These support the creation of new self-contained entities in response to new market opportunities that the enterprise seeks to exploit. Internally, they are allowed to operate with reasonable freedom, acting much the same way as a start-up but having seed funding, support and guidance from the parent entity. Longer term, in some cases these capabilities might be allowed to exploit other opportunities external to the parent in seeking new customers, eventually being set entirely free as the solution moves from an innovation to a market standard. This move has appeal on both sides; the innovators being free to continue to innovate and the parent benefiting first from early market access and secondly from continued innovation as the market opportunity matures.

This move to a market driven supply of business capabilities is perhaps the most exciting transition, starting to solve some of the major limitations faced with a traditional project or silo approach to SOA. Fully liberating new capabilities that are self-determining and self-sustaining is a major step towards realisation of a fully Service Oriented Enterprise and, with the emergence of a global cloud platform, this brings SOA even closer to reality. The result of new cloud capabilities is a rethinking of the enterprise itself and raises key questions. What, specifically, has to be done in-house anymore? How can you differentiate your business from another?