Several days ago, I wrote a post, The Most Used — Useless Metric In Sales. It’s generated quite a bit of discussion in the various sites where it appeared. In the post, I attacked the weighted probabiliy — based on progress through the sales cycle (you know –Qualified is something like 25%, Discovery, 50%, Proposal 75%, and Closing 100%). A series of discussions on sales forecasting started, so I thought I would toss more wood on the fire.
Accurate sales forecasts are critical—not just for sales, but for the entire organizattion. Resources, funds, schedules, expectations are set, based on the forecast. Forecast accuracy is usually one of the top 5 concerns virtually every CEO, CFO, or Sales Executive has when talking about the sales organization (surpassed only by making the numbers, productivity, etc.).
Try as we might, the sales forecast will never be 100% accurate (unless we forecast after we get the order). At best, the sales forecast can be thought of as an “informed guess.” Some of you, justifiably, are thinking – Dave, you’ve really gone off the deep end on this one, you can’t be advocating that we “guess.”
Let me break “informed guess” down a little. We want to focus the weighting in those words to the “informed” piece. This means using real data and analytics to develop the forecast. Most companies have rich historical data that can provide a baseline for forecasting. Couple this with external data, market information, real time feedback from customers, demographic, behavioral, psychographic, and all sorts of other information, we can develop very rich models that provide greater insight into likely behaviors of customers — consumers and enterprises alike. In the end, people make the forecast, so this data informs us.
On the “guess” side, the challenge is that everyone comes to the table with different assumptions, biases, expectations, and beliefs about the business. On the “guess” side, we need to reduce variability in the “guess” element. We do this doing a number of things. First and foremost is a strong, well defined, and well executed sales process (but you knew I’d get sales process into this). If everyone is doing their own thing, executing the way they want, we have no ability to predict the outcome. Second, we must have a common set of assumptions, rules, processes, and expectations by which we come up with our estimates. If we come to the forecasting table with different assumptions regarding risk, timing, and so forth, it is impossible to develop a forecast that everyone buys into. This variability of approach creates challenges to the accuracy and acceptance of the forecast. Extend this over time, forecast to forecast to forecast, and we can have wide swings in approaches and accuracy or meaning.
So the forecast will always be an “informed guess.” We increase the accuracy both by leveraging data and analytics to be better informed and putting in place strong processes and ground rules to reduce the variability of our guesses. I can, and have, gone much deeper into this, but will stop here and pose the questions: Does this ring true in your experience? is the sales forecast an “informed guess?’ How do you increase sales forecast accuracy?
If you are curious, I’ve written a much more extensive white paper on this: Moving Beyond The Crystal Ball: Improving Sales Forecasting and Increasing Odds To Win. If you’d like a free copy, send me an email at dabrock [at] excellenc [dot] com. I’d be glad to send it to you.
