Monday, June 27, 2011

Thirty Years Of Jobbers - Chapter 1.9


In the few cases where customers tried to keep a parallel set of books, in an attempt to test the computer, the practice always ended up in disaster. Attempts at parallel bookkeeping never worked because no one but David ever ran their manual books the way a computer does. It’s too time consuming. A computer keeps track of every tiny detail in the bookkeeping process. Manual books are massaged through the bookkeeper’s mind and a great deal of their work involves summary data; for instance, total gallons sold times the tax rate, instead of tax charged on each individual sale.

Another example: Most manual bookkeeping systems threw the fuel taxes into their cost of sales, when it is clearly a prepaid asset. Any accountant will tell you, you never put anything into expense or income unless you plan on leaving it there; otherwise your interim income statements are ambiguous.

However, fuel taxes were so complicated doing it David's way, for a jobber that was moving several million gallons during a single month it was just too complicated for a manual bookkeeping system to handle it that way. Consequently, the manual books never matched the computer books, and employees would spend hundreds of unproductive hours trying to match their manual books to the computer. The smart ones would see the problem straight away and abandon the process, but I've had a few who had to learn the hard way. Mostly it was because the boss wanted it done that way. In reality, if the boss had had to do it, he never would have attempted it to begin with.

As I said, the overworked, assistant bookkeeper in the current story, who had been assigned the job of getting the computer up and running in her superior’s absence, threatened to quit if I didn't leave. So, I left in disappointment and returned to McGehee, thinking ‘I will never get paid for that one’.

A week later, I got a call from her boss. She said, “Okay, what do we do next?” I was extremely lucky to have had that lady in the mix. Later she told me she knew it would result in a better environment for them if they could ever get it to work. She just didn't want to give up her vacation to do it. So I went back and finished the job.

This owner was a professional convenience store operator. He knew the convenience store business inside and out and I learned a great deal while pleasing him. One thing he asked for was a "Tahiti Report." He described it as the kind of information he would need to run his stores from Tahiti. It consisted primarily of the fuel and non-fuel weekly summaries for the last thirteen weeks, extrapolated out into the future.

Along the way, I learned a great deal from the ways my customers used my computer software to solve problems. It’s an opportunity most programmers rarely have. One day a customer called me all excited because he had discovered by cutting back by one, the number of packages of taco sauce the employees summarily handed out with tacos, it increased his deli profits by 2 percentage points. (I must admit I never really understood how my system told him that.)  

It was at my next installation where I really got an education on computer hardware problems. We reverted back to that huge 20 million character hard disk unit instead of the smaller one I used at my second installation. Not that it was better. It was just that jobber number two had local support and their hard disk had already been repaired several times.

No comments:

Post a Comment