Eight to Late

Sensemaking and Analytics for Organizations

Ironies of enterprise information technology

leave a comment »

Introduction

On one of my random walks through Google Scholar, I stumbled on an interesting paper entitled, Ironies of Automation.  The main message of the paper is nicely summarized in its first few lines:

The classic aim of automation is to replace human manual control, planning and problem solving by automatic devices and computers.  However… even highly automated systems, such as electric power networks, need human beings for supervision, adjustment, maintenance, expansion and improvement. Therefore one can draw the paradoxical conclusion that    automated   systems   still   are man-machine systems, for which both technical and human factors are important. This paper suggests that the increased interest in human factors among engineers reflects the irony that the more advanced a control system is, so the more crucial may be the contribution of the human operator.

These lines were written over thirty years ago, but are ever more apt today – such paradoxes are rife, not only in automation, but in any field in which technology plays an important part. To illustrate my point, I highlight a couple of ironies drawn from a domain that is likely to be familiar to many readers of this blog: the world of enterprise IT. I also present a brief discussion of how these ironies of enterprise IT can be avoided.

Ironies of enterprise IT

In the last few decades information technology has found its way into diverse organisational functions. This trend has been accompanied by an explosive growth in new technologies. As a result of this, corporate IT infrastructures have become  ever more complex and the costs of maintaining them have burgeoned.  Quite naturally, the focus has thus turned to taming both complexity and cost. The favoured approaches to tackling this problem are standardisation and/or outsourcing. However, as I discuss below, both   often lead to ironic outcomes.

An  irony of standardisation

Enterprise IT environments tend to evolve rapidly, reflecting the many demands made on them by the organisational functions they support. This is good because it means that IT is doing what it should be doing: supporting the work of the parent organisation.  On the other hand, this can result in unwieldy environments that are difficult (not to mention, expensive) to maintain. One way to address this is to impose of standards relating to processes  (such as ITIL) and infrastructure (such as SAP or any enterprise application). 

The question is, how well does such standardisation work in practice?

In his book entitled, From Control to Drift, Claudio Ciborra pointed out that IT infrastuctures in organisations tend to drift – i.e. they escape processes, plans and standards, and take on a life of their own. The reason they drift is that they are subject to unpredictable forces within and outside the hosting organisation. The imposition of standards may slow the drift but cannot arrest it entirely. Infrastructures are therefore best seen as ever-evolving constructs consisting of systems, people and processes that interact with each other in often unforeseen ways.  As he put it:

Corporate information infrastructures are puzzles, or better collages, and so are the design and implementation processes that lead to their construction and operation. They are embedded in larger, contextual puzzles and collages. Interdependence, intricacy, and interweaving of people, systems, and processes are the culture bed of infrastructure. Patching, alignment of heterogeneous actors and making do are the most frequent approaches…irrespective of whether management [is] planning or strategy oriented, or inclined to react to contingencies.

The essential message here is that standards and processes overlook the fact that enterprises are complex social systems that are subject to internal and external influences which cannot always be foreseen.   Dealing with these, more often than not, entails the implementation of hacks and workarounds that violate the imposed standards and thus nullify the benefits of standardisation.

In summary, “standardised” IT environments often end up have a plethora of non-standard hacks and workarounds that are necessary, but are generally messy and expensive to maintain.

An irony of outsourcing

One of the main reasons for outsourcing IT is to reduce costs. Yes, I am aware that many decision-makers claim that their primary reason is to reduce complexity rather than cost, but the choices they make often belie their claims.  The irony is that in their eagerness to control costs, they often end up increasing them because they overlook hidden factors.  I explain this in brief below, drawing on my post on the transaction costs of outsourcing.

The basic idea is simple – it is that the upfront fee quoted by the vendor is but a fraction of  the total cost that will be incurred by the customer.  Some of the costs that are generally not included in upfront costs are:

  1. Search /selection costs: these are the costs associated with searching for and shortlisting vendors.
  2. Bargaining costs: these are costs associated with negotiations for a mutually acceptable contract.
  3. Costs of coordinating work: these are costs associated with coordinating external and internal work. This is particularly important in the case of software-as-a-service because the effort required to interface cloud applications with in-house systems is often underestimated.
  4. Costs of enforcement and change: These are costs associated with enforcing the terms of the contract and those associated with change.

The point to note is these costs are rarely if ever mentioned by the vendor, but almost always show up in one form or another. It is therefore important for the customer to try and get a handle on these before entering into any commercial agreements.  The problem is, some of these costs (particularly 3 and 4 above) are hard if not impossible to figure out upfront. For example, if the relationship turns sour the only solution might be to switch vendors. The cost associated with this is often significant and is borne entirely by the customer.  A lack of awareness of such costs associated with outsourcing will invariably result in ironical outcomes.

In summary: attempts to control costs by outsourcing IT can have the contrary effect of increasing them.

Avoiding ironical outcomes

So how does one avoid ironical outcomes?

I have only one piece of advice to offer here: when planning IT architectures or outsourcing initiatives, use an incremental or emergent approach that avoids big designs or commitments upfront. Using an emergent approach not only limits risk, it also provides opportunities for learning.  Most important, it enables one to verify that the envisaged benefits are not just wishful architect or manager-level thinking

Below I outline what such an approach might entail for the two ironies discussed earlier:

  1. For infrastructures/systems:   avoid grandiose system designs that attempt to span the “enterprise” – remember that one size will not fit all of your users. . Consequently, enterprise architectures and governance systems should provide guidelines, rather than detailed prescriptions.  As Anders Jensen-Waud  puts it in this post: they should foster resilience and adaptability rather than conformance,
  2. For outsourcing:  start small, possibly with a small project or system. This will help you get a sense for how outsourcing would work in your environment and help you figure out whether the vendor you have selected is really right for you.  Remember, no two environments are identical so others’ lessons learned may be considerably less useful than you think. Finally, if you’re going to the cloud, be sure to factor in costs and technical challenges associated with interfacing external apps with in-house ones.

Yes, there’s nothing particularly profound here, it is just common sense…but you know what they say about the commonality of common sense.

Conclusion

In this post I have highlighted some ironies of enterprise information systems and have briefly outlined an emergent approach to avoiding them. I believe but cannot prove that ironical outcomes are almost guaranteed if one takes a monolithic, enterprise-style approach or a let’s-outsource-it-all attitude to enterprise information technology. Such a view overlooks the messy little details and differences that trip up big designs and grandiose plans. In the end, the only way to avoid ironical outcomes is to start small, learn from experience and incorporate that learning in an incremental manner in whatever you’re building or doing. Yes, you might end up with something you did not envisage at the start, but you will have learnt much along the way. More important, perhaps, is that you will be able to rest assured that it works.

Written by K

August 26, 2014 at 8:40 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: