Friday, July 28, 2006

The Weakest Link

A chain is only as strong as its weakest link and everybody knows that except management levels in the organization that are too high up to be mucking around in the weeds. As it turns out very often the weakest link is buried deep in the weeds. A familiar problem pattern I have encountered at almost everywhere I have consulted involves an Excel macro on steroids trying to stand in for a missing workflow automation, business process management or in some extreme cases business intelligence tool.

For a operation running out of the CEO's basement that would be easy to justify - getting any fancier is too expensive to be an option. When a company with a few billion dollars in revenue and in possession of state of the art technology allows an Excel macro to feed into business critical applications, it does not make as much sense. Particularly when instead of going after the root cause a dozen Big 5 consultants are brought on board to business process reengineer.

I have tried to understand the Excel macro phenomenon that I have seen repeated almost everywhere I have worked. It usually starts with a bunch of business users who are unabashedly and incurably non-technical. Whereas IT types are routinely hectored to acquire business savvy so they are able to understand end-user pain and solve for it intelligently, business types are not encouraged to reciprocate by understanding the toolset that IT would employ to solve their problem.

This is a serious limitation because the lines between business and technology are blurred with sophisticated tools and solutions that strive to put the end user in the driver's seat. The goal is to eliminate dependency on IT in a business as usual mode of operation. When end users refuse to educate themselves about options that technology offers, they set themselves up for failure. Instead of providing direction, they expect IT to broker between their needs and available solutions.

This engagement model is fraught with trouble all the way and everyone suffers. IT only half understands business-ese, and with that limited understanding they translate it to technology. Once translated business users can no longer validate that the proposed solution does what they require it to do. When at last the solution is delivered, they discover that it meets close to none of their expectations and have no use for it.

Enter the Excel macro. A low-fi, side of the desk, seat-of-the-pants hack that acts as band aid for what has proven completely impossible to translate from business-ese to technology. Typically it is created by one of the frustrated business analyst types who gets business-ese somewhat and is good enough with logical constructs to put together a Visual Basic script. Despite familiarity with both business-ese and technology, the analyst fails to create a bridge between the two worlds due to a variety of reasons - lack senior management support being the most likely cause.

The macro starts out being a little job aid that makes the analyst's life a little less hellish. Soon, he discovers potential for improvement and soups it up some. Over the years, fortified by knowledge gleaned from Visual Basic for Dummies, the macro is doing a devilish amount of processing and the consumers of data that it spews out are blissfully unaware of how their data came into being. Often this is the most authoritative system of record and the source that will feed a host of internal and external systems - astounding as it may seem, it could well be what drives the business engine.

So there is this entirely bizarre situation where terabytes of data get crunched, sliced and diced to make critical decisions, content management integrates with workflow automation and CRM tools spew a ton of metrics - and yet at the core of this fantastic operation is a humble Excel macro running on the desktop of a business analyst. The powers that be may or may not know of its existence - they most certainly don't recognize it to be the weakest link in the chain that it really is.

2 comments:

SFGary said...

I have heard stories of people using Excel to create databases and accounting systems. There are excel jockeys who feel uncomfortable in any other app.

In my case I have come across PP gurus. They think just by putting up a few slides their assertions and conclusions are rock solid. Its when you question the fundamentals you find that they are standing on quicksand. Its quite scary.

Like you said there's plenty of blame to go around from management to external consultants to IT. The best line I heard from an IT guy in one of my previous jobs was when I asked why everything was obscure or slow when I asked them to do something. and the pat answer was "hey, its job security." The second best one was from the VP of that group when some of us questioned the wisdom of writing your own database application, "we have the staff to do it." Inevitably after spending a lot of $$ the users dumped it.

Heartcrossings said...

SFG - A homegrown database system ?? That would have cost $$$$$$$ and tanked. I hear you about the PP decks seen a lot of those myself. I had blogged a while ago in the job insecurity issue ...

http://heartcrossings.blogspot.com/2006/04/spanner-in-works.html