Good sales forecasting helps you grow your business. But, for many years, forecasting has relied on the human element — emotions and hunches can make or break a quarter. Just as big data and artificial intelligence (AI) pervade many aspects of how we work, it’s doing the same for forecasting. However, it will take both the human and technological sides working together to achieve true success. Here are a few ways we can make that happen.
One of the hallmarks of sales is the pipeline. But, too often, opportunities aren’t loaded early enough, or deals are hidden under the radar. Scoring how likely opportunities are won is a key component to automating your sales forecast. With this in mind, you don't need an AI expert to point out that hidden opportunities lead to forecasting errors. It’s to the detriment of a company to over-promise or under-deliver potential. In either case, expectations are not fulfilled and trust is lost. Data and day-to-day operations should be seen throughout the sales organization throughout the entire quarter. Visibility creates predictability and accountability.
Forecasting should be treated for what it really is: a science. Without scientific logic, forecasting often happens on one of the two ends of a spectrum: either overly optimistic or overly pessimistic. Either scenario will affect a company, its investments, and, ultimately, its growth. There is a discipline and rigor to forecasting that can only come when you use data and facts to come to a conclusion. In this case, the conclusion is the result of the end of the quarter or year. The beauty of using AI and a scientific attitude to come to your conclusion is that you can use data-driven rationale to explain how you found your result. A correct prediction is great, but being able to explain how you arrived at it is even better. It is important to create accurate forecasts, but the most accurate forecasting algorithms out there are useless if they don't provide the rationale behind the numbers. This way, if we fail, we learn. In the next cycle, that “miss” will be corrected. By truly embracing data science, forecasting can become the accurate measurement and lever of growth for your business that it has always meant to be.
Remember the days when cloud computing was the big disruption? Now we can’t imagine going without it. Today, the same thing is happening with AI and machine learning. Yet it’s a completely different experience because AI is reshaping how we learn from and leverage data — not just by accessing it, but by really digging into its importance — to create personalized customer experiences. Instead of merely accessing the data, AI and machine learning allow us to really dig into the data’s impact. In terms of forecasting, AI will transform how a company interacts with its own data for and from sales.
AI is brutally honest. Let’s put it this way — it won’t hug you back. AI is without emotion and is unequivocal in its results; it tells you the cold, hard data truth. But it does need you to learn faster; it needs your guidance and input to amplify what data it has been given. You will need some patience for this type of tool as it learns about your business. When I hear that a company thinks AI is useless after just three months, I point out that it’s only learning as fast as data and experiences are being fed to it. Plus it comes with an incredible advantage: It does not forget. Data and coaching will only make it better and better.
For every potential deal, sales reps have a hunch. It’s the hunch of how it’s going to play out, which can be spot on or very far off. The emotions of a sale are still important in forecasting, but they need to be addressed in a new way. Instead of just running on instincts, forecasting must also contain logical variables that can be replicated and explained.
If a sales rep or leader feels strongly about the deal, it’s time to take an objective look with AI as the forecasting tool. The sales rep will be forced to explain what the hunch is driven by. These experiences generate data that allow you to learn from them in the future. When we started on the journey of predictive forecasting, we also developed a way to incorporate how sales execs feel about a given forecast in to the predictions. Both signals are then measured and benchmarked by real values as they occur. Leveraging the human element to augment what AI predicts was key to accelerate learning.
Adopting a new approach to forecasting may seem daunting — or maybe even unnecessary. Too many sales organizations stay rooted in their ways. But there will come a time when they’re forced to make a change. That’s certainly what happened with Salesforce when we missed our internal sales goals one quarter.
We had to troubleshoot our own forecasting and create the technology needed to ensure such a thing never happened again. To be honest, not everyone was on board at first. But, once they started seeing the results of AI, how predictable and accurate it was, they soon realized there was no turning back. Only by learning from mistakes and embracing change can a company seize the opportunity to be better.
“Forecasting should be treated for what it really is: a science.”