Making better decisions is the holy grail of the boardroom. Corporate governance is full of rules and procedures, but very little on actual decision-making. There is however considerable academic discussion of good decision-making. A really interesting example of this was published by Stanford Graduate School of Business in their ‘All Else Equal’ series: ‘The Path to Good Decisions Starts with Data’.
Their basic tenet is pretty unarguable: Good decisions are based on evidence (data), not intuition nor politics. In fact, they claim that the best predictor of failure is hubris, excessive self-confidence, ie too much trust in personal ability and judgement to the detriment of relying on evidence. There clearly is a link – the more self-confidence a manager has, the more they might use their judgement, rather than data, to make decisions. Excessive self-belief is a not uncommon feature of powerful Chief Executives, which can make it difficult for other executives – and perhaps even the board – to challenge their decisions.
The rest of the article is about how you handle errors, ie decisions that subsequently turn out to be poor ones:
Reject bad ideas quickly
They argue that mistakes often show up quite early in a project timescale. We spend a lot of effort and time making a decision, but how much time do we devote to monitoring the early stages of a project for signs that it may be failing? This is certainly true at board level. Projects are presented with a fanfare and a business case, yet once approved silence often follows. If the project is not successful, the board sometimes needs to chase this – often to be told that there were issues and the project discontinued.
Some ideas just won’t die. Most of us will have seen a project that looked like a bad idea, suddenly reappear a few months later, perhaps in a marginally different form. This can be especially true if the sponsor is a senior figure. It may alternatively be that a ‘barmy’ idea from a powerful figure is nodded through by subordinates, who subsequently just do nothing and try to bury it. This however often results in the boss exhuming the zombie idea when they realise it’s disappeared.
In our ‘can do’ culture, saying no and trying to kill projects is usually seen as being negative. More careers have been built on saying yes than saying no. Others will let the initiative go ahead and thereby allow the author to bury themselves with a bad idea. This syndrome is well illustrated by ‘The Apprentice’ TV show, where the candidates, as soon as bad ideas are agreed, bitch separately to the camera that they won’t work, thus preparing themselves for the post-mortem.
Taking this argument further, the article regards any unanimous agreement as being a danger sign. No decision is that obvious that some concerns aren’t necessary. Certainly, in my experience, unanimity is not a predictor of ultimate success.
Use good data, not just static numbers
If you are going to use data, make sure it’s good data. This means not only that the data is not only accurate but comprehensive. In particular, the article highlights the power of trend data over static. Too often board meetings look at historic data, performance last period against budget and last year, but without any feel of how trends are developing. Variances against budget in a period are meaningless unless in the context of time; is this a one-off and are the numbers and variances trending up or down?
Snapshot numbers are also prone to ‘false precision’, with enormous efforts made to get a number ‘right’ for a single period in time. Getting the year-end accounts ‘right’ takes enormous effort, yet that number is true only for one second at midnight on the balance sheet date. Every accountant knows for example that the working capital on the balance sheet date has been worked hard to show a better number. It’s remarkable how cash will tend to flow out shortly after the year-end as this corrects itself.
If you focus on trends, inevitably you start to understand sensitivities and variables better, because every explanation of how numbers change period to period is, by definition, dominated by moving variables and sensitivities. Trend analysis is therefore a great source when it comes to understanding these. Better explanation of variables and sensitivities is key to making better informed decisions.
Admit mistakes
Sometimes, sorry seems to be the hardest word, but this is closely followed by ‘I got it wrong’. We all make mistakes, but the difference between good managers and bad is that the former acknowledge (at least privately) their errors and learn from them. We spend too much time trying to stamp out errors rather than learning from them, leading to people, especially senior ones, not wanting to admit that they made a mistake.
This links into early detection of errors, discussed above. If you don’t admit errors early on, they are more likely to get bigger than rectify themselves. Some companies, like Google, supposedly ‘celebrate’ failure. This sounds a little unlikely in practice, but we should appreciate and even celebrate learning from our mistakes.
The article advocates ‘blameless post-mortems’. We learn far more from understanding the process and control failures that led to an error, than we do from hunting down and punishing the perpetrator of such a mistake. However, it’s much easier and more fun just to blame someone other than yourself. In aviation, for example, safety incidents often involve some pilot error, but they always have a trail (usually about 12) of other factors that contributed to that error, so blaming solely the pilot is a nonsense. A good system is capable of withstanding most human error.
Tips for better decisions
- Challenge whether good data is at the heart of a decision, not the seniority or enthusiasm of the protagonist.
- Monitor especially the early stages of a new initiative and be prepared quickly to change or halt a failing project.
- Kill bad ideas off, and don’t let them back.
- Look at trend data, not just snapshot numbers.
- Focus on variables and sensitivity. They tell you what to look for in any project.
- Admit mistakes and celebrate the learning from them.
- Focus on why things went wrong, not who did it.
There is much more on how boards make decisions, work together, and understand risk in my new book: “Behind Closed Doors. The Boardroom: How to Get In, Get On, and Make a Difference‘.