Thought for the day: MFI and the case for flexibility

As the furniture chain admits to problems with an implementation, Colin Beveridge says being flexible with your business model...

Colin Beveridge  

As the furniture chain admits to problems with an implementation, Colin Beveridge says being flexible with your business model will enable you to make changes more effectively




The MFI Furniture Group is the latest business to suffer from the true cost of new technology as their share price tumbled dramatically following news of computer system difficulties.

MFI blames problems with its new supply chain system implementation for a substantial loss in its UK retail business and the parent group has warned investors that group profits for the first six months of the year are down by almost 50% when compared with the same period last year.

That’s one hell of a hit for any business to sustain, and it will give more ammunition to those who challenge the value of sinking millions into major IT programmes. They will point their fingers at MFI’s experience and say there is further clear evidence that IT is a waste of money.

And the splash-back from the MFI losses could affect all of us, particularly those who are seeking approval for similar spending initiatives.

But that might not be a bad thing because the MFI supply chain system problems appear to be symptomatic of too many failing programmes.

We haven’t seen the full numbers yet but let’s consider the evidence in the public domain so far and see if there are any lessons to be learned.

For a start the MFI supply chain programme has the classic pre-requisite element for major IT disappointment, an imposing programme name: Atlas.

Nevertheless, in my experience, big project names are often an early warning alert for big underachievement and this seems to be borne out in this instance too.

The scale of the apparent underachievement in the MFI supply chain system scenario is fairly substantial. As reported by Computer Weekly in September 2003, the Atlas programme was scheduled to cost up to £25m and deliver annual benefits of around £35m in reduced operational costs, described by the chief information officer at MFI as vast cost savings.

I suspect that these ambitious projections will now have been seriously revised, in line with the announced £27m fall in half-yearly group profits.

Furthermore, I presume that the Atlas costs had already been properly factored into the 2004 financial position so the full bottom line impact of the Atlas programme on the MFI group finances, both in terms of revenue and share price, must have blown the Atlas business case to absolute smithereens.

No doubt there will be some very interesting discussions in the months to come as the MFI management try to unpick the root of their problems and understand where it went wrong.

As a relatively uninformed observer I would humbly suggest that MFI’s Atlas programme was probably fatally flawed long before conception, for two reasons.

First, the programme was heralded as a solution to earlier problems of falling sales at MFI - and I have never been convinced that any new computer system can resolve a fundamental sales problem. In fact, where such a problem exists, it could be foolhardy in the extreme to compound a bad revenue position by spending millions on a new system without properly addressing the real problem of poor sales.

Second, the stated technology rationale for the Atlas programme was to achieve vast cost savings through improved systems integration. A worthy ambition, and often stated, but rarely delivered in practice, which is why systems interoperability is rapidly replacing systems integration in the strategic planning arsenal.

Monolithic ERP systems are difficult to implement, as MFI has discovered to its cost, and impose serious constraints on future commercial agility.

If we really want to deliver agile business systems I believe that we should abandon the fully integrated, modular “Lego-brick” mentality of the past 20 years and embrace the much more flexible, albeit less aesthetically pleasing, “stickle brick” approach.

This would make it much easier for us to bring together interoperable system components that can be quickly assembled and, more importantly perhaps, disassembled as our business needs change.

After all, isn’t that precisely the business product model successfully adopted by self-assembly furniture retailers?

Colin Beveridge is an independent consultant and leading commentator on technology management issues. He can be contacted at [email protected]

MFI to save £35m a year from SAP-based overhaul >>

Read more on IT project management