Sales Management: How can I improve my revenue forecasting?

One of the perennial challenges for any sales manager is to provide accurate revenue forecasts. So much depends upon these numbers that this creates enormous stress on a virtually continuous basis. Clearly there is no “silver bullet” to remove this stress, but there are techniques that can minimize the fluctuations and which are within the control of the sales manager.

Some may say that forecasting revenues is anything but within the sphere of control of a sales manager. There are some many variables at play that impact the size, the timing or even the very existence of sales revenues. Yes…and no.

The Sales Manager interacts at a pivotable stage in the forecasting process, and this is where the opportunity to control lies. Generating revenue and forecasting expected revenue are not the same thing and indeed they have quite different purposes.

Revenues reported by the individual sales team members will carry all of the inherent market uncertainties mentioned above, along with the unconscious bias of the people themselves. Some will be aggressive, others conservative, some naïve and others pragmatic. The underlying character of the person involved will inevitably influence the revenues they report.

On the flip side, the revenues that you report to your management do not have to correspond directly to this raw data. Your internal forecast must above all be seen to reflect reality overall, and be reliable enough for key business decisions to be taken. So why is this an opportunity?

Your challenge is to “manage” the uncertainties around the input revenues and provide a single forecast to your management. Here is a way to achieve this.

Step 1: In the initial phase, ask your team to provide you not only with their anticipated revenue for the month, but also the worst case and the best case. I suggest using something like P85 and P15..ie a 85% probability of making a lower value and a 15% probability of making the higher number. Their estimate should lay between these two, but does not have to be exactly in the middle. Some people will take the easy route of simply adding/subtracting a percentage around their estimate, but that’s ok if it happens. For this initial period, say 6 months, you will compare the actual revenues against the worst, proposed and best estimates, and reality shown in red. Some people will be consistently lower, others higher etc as reflected by their own character.

As Sales Manager you can now adjust each individual forecast based upon these trends in order to remove some of the bias inaccuracies. If a team member regularly sandbags, you can make an adjustment accordingly based upon actual facts. Already your forecasts should start to become less variable.

Step 2: You also have your own unconscious bias, so by conducting the same exercise on your own adjusted forecast will further reduce fluctuations.

Ultimately you want to achieve 2 things. Tighten the spread between the best and worst cases, and reduce the variation across your team.

Step 3: As you can see from the graphs above, the estimated forecast does not actually serve any tangible purpose when using this approach, so if you wish, you can eliminate it completely and ask only for the best and worst scenarios. In itself this will bring focus on the spread, and encourage people to minimize it. At the same time you now have the data needed to engage in one on one coaching for the team members who are consistently positively or negatively skewed, or simply provide such wide spreads that it may indicate a lack of knowledge more than a reporting issue. You not only now have some data to support your observations but can work with that person to understand the reason behind it.

Is this a silver bullet? Of course not. Is it within your control as sales Manager?..absolutely. Will it help understand some of the variables and compensate when you make your forecast? Yes it will.