A good CFO can make a huge difference to a company. They do more than keep track of the books. They can shape conversations about the the future of the company, provide options and alternatives, and most importantly they make it easy to make good decisions. A bad CFO can become a anchor dragging a company down.

In our post yesterday, we looked at the need for CEOs to add a skills for allocating capital, making big greeenfield investment decisions. The natural person to fill this role is the CFO. For start-ups, having a full-time CFO is probably overkill. In early days, having someone part-time is probably sufficient, but once the company raises significant capital, it is important to have someone who brings the right skills.

A common mistake we see companies making is promoting the accountant. Of course someone needs to monitor the company’s books, but that is only a small subset of a CFO’s role. Start-ups are not alone in this. Many large companies make the same mistake, conflating the CFO with the accounting function. This often leads to problems, but in a subtle way, so it can take years for the problems to surface. For start-ups, having a great accountant is crucial, but often the best long-term role for that person is the Controller.

This is not an insult to accountants. Accounting, like engineering, is a role which requires the ability to go deep into every issue. As we noted yesterday, the role of the CFO requires a broader perspective.

We knew a largish media company with many brands. Many years ago, they had a CFO who was hired to help the company cut costs, his compensation was tied to raw dollar amounts he cut. He was very good at that, cutting expenses everywhere, even in profitable businesses. Unfortunately, by focusing on costs rather than profits, he wrecked many brands, leading the company to underperform and ultimately into a very modest sale. Similarly, a big problem at the US auto companies for many years was there focus on accounting models to make investment decisions. This lead them to have factories which could very efficiently build cars that no one wanted. The list goes on.

By definition, CFOs are not experts in the engineering side of their businesses. The CFO’s role is to monitor and predict spending, but this requires a thorough understanding of what the engineering and sales people are doing. They have to understand the business and this is not easy. A big issue we find with finance teams is that they do not always grasp what it is exactly the company does. We know a wireless company where outside the CFO, no one on the finance team really understood the difference between 3G and 4G, let alone 5G.

A big part of the problem, especially at start-ups, is communications. Predicting the timing of a software deliverable is hard, even for skilled engineers. So often we find CFOs frustrated by the unpredictability of other parts of their companies. Here the fault rests with everyone, but CFOs have to guard against becoming overly critical. That leads to the stereotype of the CFO as the cranky old man telling everyone to get of his lawn, the Dr. No of the title of this piece.

By contrast, we recently read an account of the early days of Amazon. Their first CFO Joy Covey. This account told of a junior member of the finance team whose job was to update hundreds of pages of graphs and charts. (There was a lot of Excel involved which drew us to this account.) Covey reviewed all of these every month. Through this she was able to build a deep understanding of all the moving parts of the business. She knew that a dollar of spending in one area would drive three dollars in revenue or two dollars in cost reduction elsewhere. We know another CFO who reads a lot of espionage novels. He once described his work as similar to building an intelligence network within in his company. Nothing underhanded, no assassinations and very few vodka martinis, just a matter of befriending people throughout the organization to find its pulse. Another CFO we worked with put on a facade of the grumpy, critical CFO so that when he showed interest and understanding of other parts of the company, he built up the confidence of the team. “Look, even our CFO is onboard.” There are many ways to accomplish this.

Armed with this knowledge, a good CFO can help shape future conversations. The company may need more growth, or lower costs, or big investments. Our point is that companies should not default to letting the CFO make decisions. Instead, their role is to build the framework for making those decisions. Does the company need to invest in new areas? Does the company need to hire a new team? Make an acquisition? Well, these are the trade-offs. Here is how we should think about timing. This is how we can measure success.