Otherwise, the data that you get from your time and technology investment may not be what you need to make the right decisions or achieve a real difference in results.
Here are five things that matter most in sales forecasting:
- Don’t bother with CRM if you don’t have a sales process. Without an effective sales process in place, how can you trust your CRM technology to provide relevant insights into where deals are stalled or progressing in your pipeline? How can you begin to measure verifiable outcomes and assess the performance (or coaching needs) of your sales force? How will you recognize leading indicators of customer engagement and gain greater confidence in forecasts? There’s an old saying: If you don’t know where you’re going, any road will take you there. Without a sales process, the metrics you pull from your CRM will often be just numbers.
- Forecast with metrics that matter. Many sales forecasts are built on probability analysis using weighted metrics. The scenario might go something like this: My historical win rate for opportunities in Stage Two of managing my sales pipeline is 50%, so by doing the math, I can forecast revenues of $X in the current quarter. But, win rates are only one layer of the answer. You need to ask more questions and fine-tune your metrics to get a better probability breakdown. How accurate are your sales reps at quantifying their sales opportunities? How well can they predict the timing of revenue once the deal is won? If you only look at win rates, you only get a myopic perspective.
- Dig deeper. A probability-based sales forecast assigns values to different verifiable outcomes within the sales pipeline. Let’s say those values are 25% for an opportunity in the “analyze and develop” stage, 25% for “position and follow up,” and “50%” for “negotiate and close.” The problem is that these analytical slices can give you a more conservative view of your forecast than necessary. That’s because you typically don’t win a percentage of a deal: it’s either win or lose. If you just use a probability approach, you will only consider part of the deal’s value in the overall calculation. What I do at Richardson Sales Performance is use different weighted-metric probability analyses as my pressure test for the overall forecast and to give the executive team better support for a scenario range. Then, I take a second look using a model that is based on risk and scenarios, focusing on a roll-up that is based on the likelihood that my sales reps and managers provide on the deals they have in the pipeline.
- Know your sales reps. One thing that gives me more confidence when using a risk-based approach in forecasting is knowing my sales reps. I also know that having a 90% win rate doesn’t mean someone is a super sales rep; rather, that sales rep might only be entering deals into the pipeline that he/she knows can be won. Conversely, someone who puts everything into the pipeline might have a low win rate, even though that sales rep performs well. I can look at the Richardson Sales Performance pipeline and know immediately how reliable I think it’s going to be based on which sales rep entered the deal, when it was entered, and where it came from. This kind of sixth sense comes from my knowing and working well with the entire Richardson Sales Performance team, including regional vice presidents, managers, and sales reps. I can do this because we are a small organization. In larger organizations, you have to rely on your sales managers and sales executives for this thin-slicing. In addition, tools like those offered by Marseli and others allow companies to automatically score opportunities based on criteria, such as how often a close date moves, to get a quick-strike thin-slice that is more data-driven.
- Recognize the art and science of sales forecasting. There is no one right way to leverage your CRM for sales forecasting. You have to take into account various metrics and to trust those in the sales organization who can thin-slice the data. You need to understand the analytics and to develop an intuitive sense about what is credible and what needs further proof. Most of all, you should accept that forecasting is both an art and a science.