Internal Controls Design graphic Internal Controls Design
consulting and research for internal control with risk management


Preferred ways of working

Clients ask me to get involved in different ways and assignments vary enormously. I'm flexible - and that's an unbreakable rule - but I do have preferred ways of working and you may be interested to know what I'll be most enthusiastic about. The ideas below are the most important, and in the side bar on the right you can see some of the people whose work has influenced me.

1. Design

Most projects involving internal controls and risk are hard work with initial ideas rarely working well without refinement or the need to solve related problems. We should always expect that the initial ideas will need to be adapted, improved, and extended, probably repeatedly.

Designs should be created quickly, guided by an architecture, and rapidly expanded into concrete detail even though they will probably be revised or even thrown away as more is learned and better ideas emerge.

Even on projects that don't seem to call for design work I usually find that design comes into play. For example, a project to investigate thousands of suspected cases of incomplete billing of customers and recover as much money as possible may not seem like a design project. However, the efficient solution involves grouping and prioritising the suspected errors and inventing ways to partially or completely automate the investigation and recovery of groups of bills. This is where a cycle of design and refinement comes in.

2. Efficient approaches

Efficiency is beautiful. Wasteful, unrefined designs that fail to exploit statistical effects, work against the way people think, or otherwise fail to present a near optimal approach are ugly no matter how simple and appealing they may seem.

Also, when we are trying to introduce improved methods or controls it is very helpful if what is proposed is easy and quick to do. Who wants a big fat project when a lean one can give the same or even better results?

3. Evo

This is Tom Gilb's approach to incremental projects. I am not alone in finding this invaluable as a guide to consulting projects that improve systems and controls.

Evo emphasizes early, repeated deliveries of results that are valuable to stakeholders and it stimulates rapid learning. It drastically improves the risk profile of almost any project. I don't insist on Evo of course, but any of its characteristics are welcome.

How many projects have you seen that produce nothing but talk and paper for months? Talk and paper are inescapable in today's organizations but wouldn't you rather get at least some useful improvement after a week or two instead of having to wait months for any kind of improvement.

4. Direct, detailed observation and testing

Reality is so often different from reasonable expectation that over the years I've become increasingly wary of ideas based only on reasonable expectations. If possible I like to go and look at the detail for myself. Usually this suggests efficiencies and quite often it reveals the expectations were wrong.


In summary, I'm flexible, but my preferred way of working is to get into a position where our project is expected to deliver improvements frequently and from very early on, then provide or promote lots of good design ideas that drive for the right effect efficiently, checking carefully that we are in tune with reality, and improving rapidly using what is learned during the work.


Inspirational sources

The earliest and most important influence was my father, an architect and designer who has worked from an office at home since I was a child. Herbert Simon's "The sciences of the artificial" introduced me to the idea of design as a rational endeavour and to heuristics. J Christopher Jones's introduction to his book "Design Methods" contains some insights that are worth the price of the book on their own about how our mental processes are far less organised than the paperwork on which they should be captured. Recently I've come across the work of architectural theorist Christopher Alexander, whose work on pattern languages is the closest to the design process described on my website.

Efficiency has always been connected with sustainability. This is something that also goes back to childhood but more recently Amory Lovins reminded me of its importance with his book "Factor Four".

Tom Gilb has a knack for coming up with highly effective and efficient ways of doing things. Evo is one of his best and is described in various of his books and on his website.

The belief that what we think and say intellectually can be wildly different from the reality in front of us was bolstered by a degree in psychology and by Tom Landauer and Jakob Nielsen's work on usability. Henry Mintzberg's famous study of what managers actually do as opposed to what they say and think they spend their time doing reminds us of the same thing.

  © 2005 Matthew Leitch