How Does It All Stack Up?

This is the final article in our four part series in which FleetWatch IT columnist, Tony Davis, explains the cornerstone to creating an IT solution. This fourth cornerstone deals with the overall structure of an IT system and the underlying technologies that many people focus on. It will be split into two parts. The main elements of a computer system can be pictorially represented by the accompanying diagram. Confusing? Well, we will look at each element next month, put them in layman's terms and explain what you should look for in each. But what about the overall picture?

"Pet Technology" pitfalls

By this I mean selecting a solution because of the technology, not because of the business benefit it affords. This is a serious problem for many non-technical managers when faced with making an expensive IT decision. Most often they are guided by IT personnel who are "in the know". I am constantly amazed at the amount of far reaching, expensive strategic decisions that have been made by IT staff based on their pet technology - the "hard of sight" leading the blind. Ugh! You actually limit your company by their lack of knowledge and preferences.

I must say, it is normally not done because of bigotry but due to a lack of knowledge and exposure. Solution? When deciding this size of strategy, it is always advisable to get some external advice from genuine experts who've had a broad exposure in implementing similar systems. Remember, the technology is there to serve and enhance the business, not for it's own sake or because it's cool.

So what is best?

The first key to answering this is to work in ranges. For example, if I said my budget for a new car was R300 000 to R400 000, could you tell me which cars in that bracket were BAD cars? The reality is that they are all good - you simply have a preference as to which make, model and even colour. But there is a vast difference in comparing one of those cars with one in the R75 000 to R100 000 bracket - it is a different product altogether. Likewise for systems. Pick a solution that will match your business now and in the future.

The second key is to know how it all fits together (see diagram). This is probably one of the hardest for non-technical managers to gauge, as they will probably be reliant on the advice given. "Just buy a bigger server or (even worse) put one in at each depot. We can easily double your network bandwidth". Yeah, and my costs too!

Hidden Costs

When deciding on a system, you should ask how each element of the system affects all the other elements. It may be great to have a graphic intensive screen with cool pictures, colours and revolving images but how much will it cost to drive that? And how much network bandwidth do you need?

These costs mount up very rapidly. Furthermore, the long-term running costs (upgrades to hardware and software, viruses, training, repairs, changing settings, wasted time playing games etc) are in the range of five to ten times the initial capex required. Alternatively, "dumb" terminals for character or web based systems cost half the capex and the running costs are less by a factor of about 20 times. (Wow! I wish someone had told me that up front).

Likewise, your network costs can spiral rapidly. "But I can't do anything about that can I?" Well you can control it if you select the right solution set up front.

Watch out for hidden support costs too. PC's are notorious for swallowing money like a black hole. Multiple servers mean multiple backups, support, upgrades, computer rooms, version control etc. Costs rise exponentially.

With software, beware the "I'll write it myself" syndrome. It always looks quick and inexpensive, but always costs more and slows down your time to market.

Implementation of the system has a major affect on your short and long term costs. Mostly long term. Why? Well, if you do not train your staff sufficiently, (which is the case in 80% of all implementations), they will not use it correctly and your business will not get the benefit of the investment. Would you buy a car and never take it past second gear? Why do that with your IT system?

Finally, ensure your suppliers and their products can continue to support your business and the system you have committed your company to. To change at a later stage will imply a repeat of all the expenses and their respective follow on costs.

Wrap up

Ensure your advisor has the right experience and knowledge. If not, buy some from a reputable consultant. Keep clear of pre-conceived ideas from a technology viewpoint. Technology is the tool only. Understand all the elements of a system and ask the vendors how their product integrates and affects all the others. Then ensure your chosen suppliers can support you, and ask how they will contain those costs. Once you have it all sown up, make sure you follow through to get what you ordered. "Use all six gears!"