“What’s my budget for this?” is a question I’ve been asked, or have asked myself, over the years. It usually arises from a marketing effort, or the purchase of hardware, or deciding how much to spend on a new business initiative. My answer has always been, “depends on what we get in return.” Rather than set budgets, I’ve always preferred to have a good model. With the solid understanding of your business born from having an accurate model, you’re free to make the best possible cost/benefit decision for any given opportunity or program.

The typical process of setting an annual budget seems both anachronistic in our fast-moving world, as well as susceptible to the problems of big design up front. How do I know in October what opportunities the next year will bring? Is the time I would spend putting together a budget even worthwhile? Should I formalize a budget at the point of maximum ignorance? Will we always seek to maximize our return for any spending if there’s a pre-determined budget? Shouldn’t we focus just as much on the value we’re getting from spending, and not just the spending?

I believe the underlying point of a budget is understanding, not pre-determining how to spend money. So if you don’t budget, aren’t you taking a big risk? Isn’t it a bit like running blindfolded through a forest? How will you make financial decisions responsibly?

An accurate model of your business is a powerful tool. It lets you know what you need to pay attention to, and what doesn’t matter so much. It helps you easily play “what if” games with issues that do matter, like utilization, compensation, rates or hiring support staff. This ability in turn helps you set goals and make big decisions responsibly.

My economic model for Atomic Object is a spreadsheet organized as an income statement (aka profit and loss statement, or “p&l”). It’s quite detailed. Each employee is represented individually, for instance. It calculates revenue from things we measure and control (rates and utilization). It predicts the values of metrics we track.

We validate the model periodically. When I was first developing the model I would compare the prediction of the model to actual results every quarter. That helped me find inaccuracies in my modeling. It also resulted in careful alignment of the model with our book keeping, as I wanted to make it as easy as possible each quarter to do the validation. Ten years on I trust the model and don’t feel the need to validate it every quarter.

With careful tracking of money and time, and an accurate, validated model of the business, you can confidently make financial decisions while attempting to optimize the return on each investment.

Forget the budget and work on a model.