The black box of accounting technology

Register now

To me, ease of use has always been the most important facet of technology. But these days, I’m conflicted. Should we know more about the underlying processes of our software?

My first professional jobs, a long time ago, were in maintenance programming. That meant going into other peoples’ software code and making changes to add features or change the way the program worked. Of course, to do that, I first had to figure out what the program was actually doing before I started making changes. That led to me learning techniques of system analysis, and eventually returning to school for an accounting degree to better understand the kinds of business systems I was designing and programming.

Fast forward a few years to when I was editor in chief of Computers in Accounting, the predecessor to Accounting Technology. I wrote in one of my editor’s notes about “the 6 million dollar accountant.” The reference was to a popular TV show, “The Six Million Dollar Man,” about an astronaut who suffers a horrible crash and is given bionic replacement parts.

In that column, I bemoaned the fact that while software was evolving to become easier to use, more powerful, and making practitioners more efficient, it was also evolving into a black box. If you’re not familiar with the term, it refers to a process where there’s no detail between the input to a system and the output. What happens between the two is a “black box.”

Now a lot of things in our life are effectively black box operations. You go to the supermarket and buy a box of cereal, go home and have it for breakfast. You can do this without any knowledge of what goes into making that cereal and getting it into the supermarket. If you flowchart out your actions, that process, is a black box between purchasing the box and eating breakfast.

My lament at that time was, and still is, is that using something such as accounting and auditing software that incorporates complex procedures and statistical processes without understanding what’s going on underneath, leaves you open to some major misunderstandings.

That’s not to say that every accountant needs to be a statistician as well. But if you are using the Simplex algorithm for a client to do mini/max calculations or resource allocations, you probably should have at least an inkling of why the algorithm works, so that the output from the process makes more sense. If you understand what’s happening inside the black box, at least to some extent, the output of the process starts to make more sense, and the decisions you (or the client) make from that output often tend to be better.

Easy does it

Apps are becoming ever easier to use. Even complex operations have been developed down to a few keystrokes or touches on a smartphone or tablet screen, with some very complicated processes and coding invisible to the user. And that leaves me somewhat conflicted. I’m all for ease-of-use being paramount. With the increasing capability of machine learning and artificial intelligence, that’s going to happen at an increasing rate. But there are admittedly a lot of processes where viewing certain parts of the system as a black box is perfectly fine — I don’t have to know about valves, pistons and gears to operate my car.

The bottom line is that the trend to use software applications without a good idea of the underlying processes is not only going to continue, but accelerate. And with processes that go beyond procedures inherently easy to understand, I’m not sure that the increased reliance on performing complex analytical tasks with a black box and little (if any) knowledge of why something works the way it does, is a good idea, especially in the accounting and auditing industry. Hopefully, as technology improves, so will education on why things work the way they do.

In the meantime, you might want to consider doing a bit of continuing education on your own. Understanding what’s going on when you press the “Return” key or tap a screen button can only make you a better professional.

For reprint and licensing requests for this article, click here.