Technical debt in financial models
Excel has very little inherent structure. Any structure your model has is going to have to come from you.
The fact that Excel is just a big table that you can do anything with is one of the reasons for its success and ubiquity. It's also why financial models can be so awful.
Financial models need structure.
And because Excel has none, that structure has to come from the Modeller.
You have to impose it on the blank canvas of Excel.
That means you have to be consistent and disciplined. And that's hard. As modellers, we often work under a lot of time pressure. It's very tempting to cut corners. To hack in solutions with a promise to ourselves that we will "come back and fix it later".
With twenty years of financial modeller experience, I can promise you this:
That day, when you come back and fix the hacks in your model, will never come.
And so you'll leave the shortcuts, and the hacks, and the minor inconsistencies.
And all the time, you're building up "technical debt" in your model.
Technical debt is a term from software engineering. It applies well to modelling.
Technical debt refers to the implied cost of additional rework caused by choosing an easy but limited solution now instead of using a better approach that would take longer.
Like financial debt, technical debt has a cost. The "interest" on technical debt takes the form of:
- reduced "understandability" for other people;
- increased risk of errors, especially when others try to change the model;
- decreased flexibility later in the build process.
You will only "pay off" the technical debt in your model when you spend the time removing the hacks and the inconsistencies.
This is time that is very hard to find.
A better approach is to build the model without them in the first place.
Comments
Sign in or become a Financial Modelling Handbook member to join the conversation.
Just enter your email below to get a log in link.