Play-By-Play: The Problem

Pretty much all software is intended to solve a problem of some sort. The “problem” simply be “I want to be entertained”, but for “business” software it’s usually more specific. Ideally it’s tied to a profit-generating function. You’ll sometimes hear them referred to as “pain points”, “challenges”, “requirements”, or “opportunities”; they’re all really the same thing.

A few weeks ago I was approached by one of my wife’s friends. She’s opening a bakery here in Calgary and asked if I would be interested in creating some software for her. We met, discussed her needs, and I got a bit of understanding about her industry. From this I derived the major problem that I’ll focus on for the first round of development:

To be successful, low-volume / boutique bakers need to keep good control of their materials costs. However, it is difficult for them to do so, and so many do not.

This manifests itself in two ways:

1. Recipes are written to produce X number of finished units for Y ingredients. The bakery will require X*Z finished units per day, and so much put in Y*Z ingredients. However, if Z is not the values 2, 4, or 10, then calculating X*Z and Y*Z becomes relatively difficult and cumbersome. (The situation is made worse if you’re using non-metric units of measurement.) Instead, many bakers simply round up to the nearest easily-calculable value of Z, resulting in a lot of waste and expense.
2. To determine the per-unit cost of a product, you need to take:
   • The price of one purchasable container of each of your ingredients,
   • Divide that by the number of measured units in the container
   • Multiply that by the amount of ingredient in the recipe
   • Adjust (add) for waste and other miscellaneous uses
   • Repeat the above for every ingredient in the recipe
   • Sum the results for each ingredient together
   • Divide by the number of units in the recipe

   Repeat this process for every product sold (or potentially sold) in the business. Again, all this hoop-jumping means that bakers tend to not do this, and thus do not understand the costs of their products. Without knowing costs, you can’t know margin, profitability, efficiency, value, and opportunity.

Basically, the problem is one of mental math. Humans suck at it. Worse, they are often afraid of it, and so will go to lengths to avoid it. On the other hand, it’s basically the only thing that computers do, and they do it faster, cheaper, and more accurately than a human ever can.

This problem is well-suited to a technical solution: not only does it bridge a gap in human capability, but it’s one that’s extremely easy to build. (I basically just outlined the algorithm above). That means that it can be done without too much up-front investment in time and money. That’s important for determining whether it will make a viable business.