Data-Driven Forecasting: The Future of Sales Forecasting Using AI & Historical Data

What is Data-Driven Sales Forecasting?

Data-Driven (or Evidence-Based) Forecasting is a modern-day sales forecasting approach that utilizes sales data to build a predictable and accurate sales forecast for an organisation. Rather than using seller intuition and gut feeling, an evidence-based forecast uses current and historical data, prior performance, trends, and AI to predict an outcome.

There is a problem with sales forecasting. A study by Upland Software found that, “Salespeople spend about 2.5 hours each week on sales forecasting and for most companies, the forecasts are less than 75% accurate. When success or failure is usually measured in margins far less than 25% - these forecasts are truly worthless.”

Especially now with a 95% forecast accuracy quickly becoming the benchmark, teams are struggling to find a way to rapidly achieve their goals. Data-driven forecasting is the way forward.

Why is this approach better to achieve more accurate sales forecasting?
Eliminates Gut-Feelings

This type sales of forecasting eliminates the seller’s emotions and favors hard evidence. No longer will a seller use their gut instinct on whether or not a deal will close. Evidence-Based forecasting is so accurate because it essentially removes the opportunity for human error from your sales forecast.

Rather than sellers assuming a deal will close simply because of positive feedback within a meeting, evidence-based forecasting requires that there be a demonstrable progression of meetings, interactions, and positive steps taken towards a closed-won deal.

Accurate Forecasts are Based on External Factors

It is also important to note that the best evidence-based forecasts rely on external actions taken by buyers, rather than the steps taken by sellers. Salespeople can easily expedite deals by sending pricing sheets, performing demos, or discussing implementation.

Although these steps may appear to demonstrate a healthy deal. Buyers may not be interested in the information provided. Instead, teams using evidence-based forecasts should use buyer-driven steps, like requests for a demo, pricing asks, and so on.

Levels of Data-Driven Forecasting

There are different levels of data-driven forecasting. The first of which can be accomplished by nearly any organization, but requires manual work. As the levels advance, the forecast becomes more accurate, but tools and technology are needed to automate processes needed for deeper analysis.

Basic Sales Forecasting

The first level requires a sales process or methodology like MEDDICC, which serves as the guide by which forecasts are made. As a sales rep completes each step of the MEDDICC methodology, the deal is considered to have a stronger and stronger chance of success.

To forecast properly in this stage, every deal must be reviewed by a manager to help keep the reps honest. However, this can be complex for teams with a large number of salespeople or a high volume of deals, as managers will have to dedicate time to reviewing every deal, every week.

Data-Driven Historical Forecasting

In the next level, teams must have a significant amount of data to leverage within their forecasts. They should be able to apply historical performance from a rep level and use that to gauge outcomes in conjunction with current quarter performance.

With historical data, it is important to consider salesperson tenure (more experienced sellers), rep headcount, activity levels, and external market factors.

Data-Driven Forecasting with AI

The final level, which has only just recently become available is the addition of AI for sales interaction analysis. RO&I platforms like BoostUp now have an AI that is capable of not only analyzing the interactions but also the conversations that happened within them.

The platforms are then able to determine risk on a deal-by-deal basis. They examine what interactions have or have not taken place, what conversations were had, who was involved, and what was said to predict the outcome of a particular deal.

Why is Data Driven Forecasting Becoming so Popular?

Evidence-based forecasting is becoming more popular as the sales forecast itself becomes increasingly important throughout every organization.

The Importance of The Sales Forecast

Sales is no longer the only team that cares about the forecast. Now, marketing, customer success, customer support, solutions engineering, and more all make decisions based on the sales forecast. Therefore, the revenue team must get it right.

Further, the proliferation of revenue operations (RevOps) has more and more organizations thinking about how they can improve their businesses. Many now believe that sales shouldn’t be involved in forecasting at all, and rather that it should be the duty of RevOps to gather the necessary data and predict an outcome.

Today, accuracy pays. Put simply, data-driven forecasting is the most accurate sales and revenue forecasting method there is. As Brandon Purcell, Principal Analyst at Forrester says: 

“Opinion-based forecasts have low predictability and accuracy, are prone to bias and manipulation, and yield limited value to the B2B organizations that adopt them. Fortunately, artificial intelligence is infusing and enhancing B2B sales forecasting."

Benefits Across Sales and Operations

Evidence-based forecasting provides a true picture of the sales team’s progress against their goals, identifies weak points, and wins.

Operations Teams

For operations teams, they can pinpoint exactly where improvements need to be made, how resources can be best allocated, what adjustments will improve progress against goals, and what can be done to increase sales performance.

Sales Managers

For sales leaders and managers, evidence-based sales and revenue forecasting can provide them with the insights necessary to improve individual sales reps’ effectiveness through sales coaching. They can identify reps who are underperforming against their goals, who have win rates that are historically lower or are not converting leads, and specifically address their needs to improve overall sales performance.

Evidence-based forecasts also afford far more granularity than any other kind of forecast. It allows for individual deal inspection, and combined with AI risk assessment, can alert teams to potential pipeline issues before they even begin.

Sales Teams

Sales teams can save deals from slipping, address specific needs within each opportunity, and coach reps on the areas where they actually need it to increase sales performance without the cost of additional headcount. In fact, 97% of teams who implemented a best-in-class forecasting process achieved their sales targets.

Eliminates Common Issues From the Sales Forecasting Process

Evidence-based forecasting solves many of the problems that cause inaccuracies in other forecasting methods. Some organizations generate more revenue in some quarters than others, headcount ebbs and flows, new managers or leaders impact team performance, different sales tools impact productivity.

When there is a change in the market, products, competition, or elsewhere, historical forecasts and weighted forecasts become unreliable. Evidence-based forecasting relies on the current state of your pipeline to produce the most robust and reliable forecast possible.

How is Data-Driven Forecasting Different from other Forecasting Models?

As previously mentioned, data-driven forecasting is different from other sales forecasting methods in that it is 100% based on data, rather than seller intuition. There is little to no “gut feeling” in evidence-based forecasting and that leaves no room for error.

When Evidence-based forecasting is used in conjunction with a revenue intelligence or forecast intelligence platform, it also provides real-time insights into pipeline progression. Instead of snapshot data, which is updated weekly (at best), teams with revenue intelligence platforms can see exactly where they stand against their forecast, at any time.

This real-time information is highly important in today’s rapidly changing and highly competitive business environment, as teams strive to make their forecasted numbers month after month.

How Does Data-Driven Forecasting Work?

Weekly Submissions by Sales Rep

The data-driven sales and revenue forecasting process begins like any other sales forecasting method. Sellers first submit their forecasts on a weekly basis, preferably at the beginning of the week. These submissions must include more than just a number, however. Reps should submit the specific deals they expect to close, including their committed, best-case, pipeline, and closed business.

Within each of these deals, salespeople should include the deal amounts, as well as the stage of the sales process they are in, with proof. In the case of an organization using MEDDICC, they should provide evidence of which components they have secured.

Due to the complexity of this information, standard spreadsheets or even CRMs are not recommended. Revenue intelligence platforms make this data far easier to organize and analyze.

Sales Manager Inspection

Next, managers inspect each and every one of the submitted deals to determine if the amounts and predicted outcomes are accurate. Managers also have complete leeway to overwrite or change any submission if they determine that the forecast is not realistic.

These overrides must be completely unbiased and driven only by the data presented. Regardless of how great a conversation might have been or how excited a prospect sounded, if there is no demonstrable progression, then the deal may not count. It is also important to remember that the progression must be driven by external factors, rather than seller actions.

Risk Assessment

If a team uses a revenue intelligence platform with AI, this is also an important time to reconcile what the AI sees and what the seller believes. AIs like BoostUp’s examine the language used in each email and within each meeting, as well as what actions took place and between whom to determine how much risk is within each deal.

This risk score is based on sentiment, engagement, and the seniority/roles of all parties involved. If there is a gap between what the seller believes will close and what the AI predicts, a deal might need further involvement from a manager. AI is also highly useful in determining the next best step, and can assist sales reps in increasing their close rates.

Key Considerations with Evidence-Based Forecasting

There are two main considerations that every team must keep in mind when implementing or running an evidence-based forecasting program: Training, and accountability.

It is crucial that all members of the revenue team use the evidence-based forecasting method. There should be a total understanding of the goals of the method, and why it is imperative that no judgment be involved.

Further, the sales process being used in conjunction with the revenue forecasting model should also be reinforced. Ensure that all salespeople, managers, and leaders are familiarized with the method so that it can be used as a standard practice with no deviation.

In terms of accountability, a rigorous sales process is crucial for accuracy. As with any forecasting process, teams must create a forecasting cadence and stick to it. Each component of the forecasting process must be executed on the same day in the same manner, regardless of what else is happening. Each individual role must be accountable for their duties.

What Revenue Operations & Intelligence (RO&I) Tools Can Help to Achieve Data-Driven Forecasting?

For maximum efficiency, productivity, and effectiveness, teams will need a revenue intelligence tool for evidence-based forecasting. AI is needed to analyze as much data as possible, and to take full advantage of all the benefits of this forecasting method. Additionally, the structured submission process, streamlined workflows, and enhanced flexibility afforded by a platform like BoostUp significantly increase the impact of evidence-based forecasting.

AI-Driven Insights

Revenue Intelligence platforms like BoostUp are built for evidence-based forecasting. With BoostUp’s automated sales activity collection and AI-driven insights, teams get the most accurate, effective, and efficient deal-by-deal forecasting available.

The AI calculates an engagement-based risk score based not only on what has happened but also on what has not happened. As a result, sales teams get early detection of issues and the steps needed to remedy them to close as many deals as possible.

Deal Inspection

It also provides granular inspections, with the ability to drill all the way down from the pipeline into a deal, and the individual activities that occurred within it to understand exactly what influences outcomes.

In a study of sales teams using BoostUp, it was found that AI-driven deal-by-deal insights saved an average of 20 hours each week, resulted in three more won deals each quarter with no additional headcount or hiring, and teams consistently landed within five percent of their forecast.

Teams were empowered with their full deal insights and the ability to inspect each deal to the keyword level. With BoostUp, as managers review deals, they can understand which competitors were mentioned, what requirements are, and what factors make or break a deal.

To learn more about how BoostUp works with evidence-based or deal-by-deal forecasting, download our Guide to Deal-by-Deal Forecasting.