Accuracy vs usefulness with back-of-the-envelope calculations

  • by

When estimating there is always a trade-off between the accuracy of the estimate and the time it takes to develop that estimate. Simple ballpark estimates can potentially be made off the top of your head, while more accurate forecasts may take a lot longer to produce. This leads to some tension between the usefulness of the estimate (where more timely estimates are potentially very useful), and the accuracy of the estimate. This article explores some key considerations in estimating approaches. 

The need for usefulness in back-of-the-envelope calculations

Insight into specific areas impacting decisions

The key thing to remember about back-of-the-envelope calculations is that they are designed to be useful – to give you insight into specific areas that can then help with decision making. Gaining some insight into a particular setting is often better than no insight – it can help you prioritize different areas to be focussing on, or decide how much resources to commit to a particular setting. 

Timely analysis

When considering the usefulness of back-of-the-envelope calculations, timeliness is important. Getting information in the moment – potentially during a meeting when you are actively deciding on a particular decision – is often a lot more useful than getting the same information several days later. Information to hand during the decision process can be taking into account, potentially directly influencing the decision that is made. Several days later the information may be much less valuable – the decision may already have been made and the information now redundant. 

The importance of accuracy in back of the envelope calculations

Considering the level of accuracy that is needed

Of course, while the timeliness of back-of-the-envelope calculations is important, it is also important that the information itself is at least to some degree accurate. Basing decisions on incorrect information can quickly lead to poor decisions getting made. Crude, unverified back-of-the-envelope calculations may be an order of magnitude off from the true figure, and basing decisions on such information can quickly lead to missteps being made. 

The size of the decision that is being made can influence the accuracy that is desired in estimations. Large organizational decisions, that are hare hard to reverse naturally require more accurate estimations than more easily adjusted decisions, where potentially more approximate information is sufficient for decision-making. 

Considering ways of increasing accuracy

It can be useful to consider if there are ways of verifying or triangulating information to increase the accuracy of your estimates – particularly if the information underpins critical strategic actions. Potentially you can find information that would support a critical assumption or are able to triangulate the overall estimate with another piece of information. If the verification coincides with what you expected, it can increase your confidence in the estimations – if it is at odds, you can re-visit or look to further verify your estimate before taking any substantial actions. 

Final thoughts - don't let the desire for accuracy reduce usefulness

It is easy to fall into the trap of assuming that you need perfect certainty in your estimates before you should take action on them. Unfortunately, for many estimates, gaining full confidence in the accuracy is never possible. The information may be impossible to verify, and your estimate may be the best you are able to come up with.

This is, unfortunately, the reality of business decision-making – you will have to make decisions under uncertainty. Often timely estimates that can help bring some level of clarity to a particularly uncertain environment can be more useful than ones that are more accurate, but take a long time to obtain.