Sales forecasting, in simple terms, is the process of revenue estimation by predicting how much a sales unit can sell a product or service in the upcoming week, month, quarter, or year. When it comes to one of the most important numbers to a business, the sales forecast sits at the top dictating other variables related to marketing, hiring, prospecting, and even product development.
Accurate sales forecasting is both an art and a science, and getting it right is a tough task even seasoned salespeople struggle with as Gartner reports only 45% of sales leaders have high confidence in forecasting the accuracy of their organization. So what goes into sales forecasting, and why is it tricky? This article explores.
What is sales forecasting?
When businesses produce a sales forecast, they estimate the figure of what they expect the sales revenue to look like. It estimates how much of the product or service they can sell in a period of time, such as a month, quarter, or year. Some of the best sales forecasts can predict revenue with a good degree of accuracy depending on the inputs and the time frame of the prediction.
Sales forecasts are generally made using data from previous performances, and the sales forecasting techniques differ based on the inputs. For example, the forecast might be made based on the intuition of the sales rep or using data and trends fed into artificial intelligence. Sales forecasts made by sales reps are used by managers to estimate the business the team brings. Forecast data from teams are used by the directors to anticipate sales by the department and the VP of sales uses collective data from the departments to project organizational sales.
Sales forecasts answer two important business questions-
- How much you expect to sell?
- And in what time frame?
Every sales opportunity has a projected amount as the potential business to the company and sales teams have to come up with a realistic number that represents the new business and its time frame.
Answering the two questions helps the organization set goals and create sales strategy plans.
The need for sales forecasting
Forecasting creates a reference point for the future that companies can rely on to make important business decisions. Accurate forecasts also establish companies as market leaders and confer credibility in the industry. Here are some of the ways sales forecasting affects the functioning of a company:
1. Strategic decision-making
Sales forecasting can reveal emerging trends and prompt decision-makers to rectify a problem or capitalize on a foreseeable opportunity. For example, a 30% negative deviation from the sales target might uncover poor management or underperforming sales units or even indicate competitors upping their game. On the other hand, a positive deviation would indicate the need to recruit more resources to capitalize on the opportunity. Forecasting sets the baseline, which aids in timely action and course correction.
2. Charting a path forward
Forecasting becomes especially useful during times of underperformance. It enables decision-makers to draw clearer lines and set better goals to reduce damages and even motivate the teams through milestones and action plans.
3. Financial decision-making
Accurate sales forecasting plays a crucial role in different financial modeling exercises. They determine:
- Inventory: Accurate sales forecasts are made through informed assumptions on consumer buying habits or seasonal increases or decreases in demand. These forecasts help in inventory planning and management, which contributes to the efficient utilization of working capital. This efficiency then extends to better raw material planning and hiring.
- Cash Flows: Cash flows are directly affected by sales revenue. Accurate sales forecasting can help model cash movements that allow companies to plan for any shortfalls or windfalls in the future.
What does a good sales forecast look like?
A good sales forecast is highly accurate and easily understandable by different stakeholders. It is also well-balanced against time, effort, and the costs associated with the forecasting technique. Ideally, an accurate forecast model should be built with reliable economic methods. The forecast model should use an algorithm finely tuned to the business and pick relevant data with little manual intervention to make accurate predictions.
However, realistic forecasts are more subjective and time-consuming. Other than the existing numbers, it becomes important to factor in the sales rep’s assessment of future performance. These perceptions can vary significantly from one sales rep to another, depending on their approach and experience. For example, a seasoned rep’s 50% sales estimate might be an understatement compared to a different rep’s 60% estimate, which might be overly optimistic.
Who is responsible for making sales forecasts?
Each organization has its own set of forecast owners depending on the type of business and the hierarchies. Typically the people who make sales forecasts are:
- Sales reps: The people who do the actual selling know their customers and the target market. They are able to set reliable estimates for how much they can sell in a week or month, or quarter based on the market conditions.
- Sales Leaders: The sales leaders pick the numbers from their individual units and make an estimate for their higher-ups. The forecasts can vary based on their seniority - third-line managers, for example, typically consider a wider set of numbers and previous trends in close rates to come up with a forecast, while first-line managers consider opportunities to make their forecasts.
- Product Leaders: Product leaders base sales estimates on what product is available for selling and the time frame for its release.
Sales Forecast Approaches
There are two main approaches to sales forecasting:
1. Top-down approach
In the top-down approach, sales forecasting starts with the bigger picture and works downward to define the milestones needed to reach the target. For example, if the market has 100 million units of a product and the organization’s goal is to penetrate 5% of the market, then the number of target customers would be 5,000,000. With such a large estimation, there’s much scope for rushed and ill-defined judgments that can lead to unrealistic expectations. However, the approach is useful for quickly establishing optimistic organization-wide benchmarks in established companies.
2. Bottom-up approach
The bottom-up approach is a conservative and granular approach that takes into account the resources held by the company. For instance, how many salespeople are there in a unit, and what is a realistic sales estimate for each rep and each unit, or, how many ads displayed on the search engine will lead to a click-through and sales? The bottom-up approach takes a practical look at the efficiency of the business and figures out the variables which can be tweaked to increase sales. This approach relies heavily on existing data to create a more structured and realistic perspective for sales forecasting purposes.
Sales Forecasting methods
Sales forecasting methods can be broadly divided into Qualitative and Quantitative methods. Qualitative methods rely on the subjective judgment of the sales reps and decision-makers, while quantitative method relies on data, numbers, and statistical modeling.
Qualitative sales forecasting
Qualitative sales forecasting often uses 5 major methods. These methods are based on informed opinions about the markets, trends, and prospects. The surveys involved are often time-consuming and expensive. The top 5 methods include:
- The Panel method or Jury of Executive Opinion method: As the name suggests, the approach involves executive groups discussing sales predictions to reach a consensus. One of the main advantages of the method is that experienced members of the jury can bring in plenty of wisdom to the predictions. The same can also be a disadvantage as bad predictions can be made by dominant members of the jury with biased views.
- Delphi Method: The Delphi method is iterative in nature and involves surveying each expert independently. The output is then shown to the experts so their responses can be reconsidered in light of the broader consensus. The approach is an antidote to the group-think that can dominate a jury approach.
- Customer Surveys: In this method, prospects or a sample of the customers are surveyed about their purchase plans in the short and long terms. For larger markets, various survey methods can be employed to determine a generic trend.
- Sales force composite method: This method forecasts sales by pooling the collective numbers of forecasts of individual sales reps. These numbers are then reviewed by the heads and sales managers along with product owners to make distilled forecasts. While the approach has its merits, it also doesn’t take factors such as new trends, regulatory changes, and product innovation into the picture.
- Scenario planning: Scenario planning is an all-encompassing approach that doesn’t come up with a single accepted forecast. Instead, it models different scenarios to let companies prepare for uncertain sales outcomes. This method is used for estimates sales over a long period of time, such as three years or more. Under scenario planning, variables such as recessions, disruptive technologies, changes in prices, and other things that affect sales are brainstormed.
Quantitative sales forecasting
Quantitative sales forecasting uses data and statistical modeling to predict sales over different time scales. Here are two of the most commonly used methods:
- Time Series Method: The approach works under the assumption that sales trends historically repeat over seasons and sales cycles. Hence it uses historical, chronologically ordered data to make sales forecasts. Future sales are calculated by historical sales multiplied by the growth rate. Some of the popular techniques include exponential smoothening, moving averages, ARIMA, and X11.
- Casual Method: In this method, the historical cause and effect between sales and market variables are taken into account for forecasting. With the casual method all possible variables that can affect sales are modeled to make accurate forecasts for the future. The variables include factors such as customer sentiment, third-party surveys, macroeconomic trends, and internal sales results. Popular casual techniques include linear or multiple regression, leading indicators, and econometric approaches.
Sales Forecasting Examples
Here are two sales forecasting examples for both qualitative and quantitative sale forecasting methods:
1. Intuitive method (Qualitative)
The intuitive method is the simplest of the qualitative methods for sales forecasting. The approach relies strongly on the performance and experience of sales reps in closing deals and their track record of matching up to expectations. The method is quite helpful if there’s no historical data to make a forecast for the month or quarter. Instead, the “intuition” or the “gut feeling” of the sales reps based on their initial contact with the prospect is used to determine how much sales can be done. Here’s an example of how it works:
The sales manager asks for an estimate from four sales reps for the quarterly sales. Sales rep 1, who is the top performer, estimates $200,000. Sales rep 2, who is a close performer to the former, makes an estimation of $180,000. Sales rep 3, who has two years of experience, estimates his sales to be around $120,000, while Sales Rep 4 who is a recent college graduate, gives an estimate of $110,000. Summing up the forecasts gives an intuitive forecast of $610,000 in sales for the quarter. However, upon close inspection, it is discovered that Sale Rep 4 has an optimistic exaggerated forecast because of his inexperience. His realistic number is closer to $60,000 in sales. Therefore the revised quarterly sales would be about $560,000.
2. Historical Method (Quantitative)
As discussed earlier, the historical method is an example of the time-series forecasting technique, which uses historical data to make future predictions. To account for the variables such as growth, inflation, fluctuations in demand, and other variables, an estimated growth rate is multiplied by the historical sales to arrive at the future forecast. Here is an example of how the method works: The estimated growth every year is 6.5%, and the sales for last January were $55,500. The forecast for this January would be (55,500 X .065) +55,500 which is $59,107.5.
Designing a sales forecasting plan
Sales forecasting requires a strong foundation of mathematical techniques and detailed knowledge of the typical sales cycle. Coupled with relevant data, the following steps can be used to design a sales forecasting plan:
1. Choosing the forecast method
Data from past sales and forecast models is essential to build new models for the future. A number of simpler sales forecasting techniques, such as opportunity stage forecasting, historical forecasting, length-of-cycle forecasting, etc., can be used to make reliable forecasts. Forecasts with higher precision can be made with more data-intensive models such as multiple regression and exponential smoothening.
2. Determining forecast timing
The type of plan depends on whether the forecast is being made for a defined amount of time such as monthly or quarterly, or whether the sales is tracked for a specific output only. Seasonality factors, such as a release during a specific quarter, can also affect forecasting. Hence that quarter may have to be looked at differently.
3. Breakdown sales cycle
The timeline of each sale has an influence on sales forecasting. By breaking down the sales cycle to the average time spent on each stage, it is possible to determine the average length of the sales cycle.
4. Utilize sales metrics trends
By utilizing the most recent historical data, variables such as the average sales price and renewal rate percentage can be defined for the new forecast. Other variables such as conversion rates, turn rates, average growth trajectory, and the annual recurring revenue (ARR) must also be incorporated into the forecast.
5. Create a forecast template
Generate a template based on the sales cycle, objectives, metrics, and specifics of the sales teams. Smaller companies with limited resources can use a tool like Microsoft Excel when there are fewer products to track. Larger organizations with automated tools that connect to the CRM can utilize the features of automation to make estimates. For instance, lead enrichment software can predict when a prospect is on the verge of conversion. This data can be used to make highly accurate forecasts for different sales cycles.
6. Share the formalized documents with the teams
A formalized documentation is necessary to share the plan with the teams in full transparency. It’s essential that the sales reps understand how the forecasting is made, so they have a solid understanding of their goals and quotas.
How to accurately forecast sales
An accurate sales forecast is a delicate balance of incorporating historical trends, internal changes, market fluctuation, and competitor pressure. Here are 5 steps that lead to accuracy in prediction:
1. Historical trend assessment
To create a basis for a sales forecast, it is essential to build a ‘sales run rate,’ which is the projected sales for the sales period. Historical data from the previous year can be segregated according to the price, product, sales rep, sales period, and other variables to create a sales run rate for the forecast.
2. Incorporating changes
The sales run rate needs to be modified based on several variables, such as pricing, promotion, channels, customers, and product changes. These variables give a more realistic sales run rate for the sales period.
3. Market trends anticipation
Market trends, such as changes to competitor behavior, legislative changes, company mergers, etc., can throw a curve ball during the most unexpected times. It is essential to factor in these changes to create forecast models in the event of a market trend shift.
4. Competition monitoring
It’s imperative to monitor the competition to know their impact on the target market with actions such as price variation, new feature rollout, or new campaigns. New competition should also be monitored for its impact on the market share.
5. Including business plans
Business strategic plans that have an effect on growth, hiring, targeting new markets or kicking off new campaigns can all have an impact on future sales. Therefore it is important to make forecasts while keeping sight of business plans.
Key challenges to sales forecasting
Producing consistently accurate forecasts can be challenging for organizations. Here are some of the major hurdles encountered by most:
- Accuracy: Companies, especially startups who are bootstrapping tend to rely on spreadsheets for forecasting, which can introduce huge accuracy issues to the forecasts. Even companies with CRMs struggle with poor adoption across the company, with employees not entering data on time, data silos, incomplete data, and inaccuracies.
- Subjectivity of forecasts: While forecast quality does depend on good decisions and judgment when predictive analysis takes a backseat to the subjective analysis, it can miss real drivers of accuracy.
- Universality: Sales forecasts that are not useful for stakeholders across the company are ineffective in producing results. Good forecasts always have relevant and understandable data for different teams across the company.
- Inefficiency: Inefficiency can often make it into forecasts when there are multiple owners, varied inputs that remain unreconciled, and too many revisions and different versions.
Sales forecasting during unpredictable times
Unpredictable times are events such as major disasters, a crisis such as the COVID pandemic, or sudden economic turmoil. These events can suddenly flip sales forecasts on their heads. As soon as one of these events hits, it’s important for business leaders to know the following:
- The current state of the sales pipeline
- The best and worst modeled scenarios
- How much has the forecast changed over the week or the month
A real-time view of the sales pipeline is critical in such situations to make an instant business decision that can minimize the damage from a disruptive event. The CRM solution and automation are what would cushion an inevitable blow to the forecasts. Reliable instantaneous data is what would help business leaders to pivot territories and resource deployment that can have a strong bearing on the continuity or dissolution of the business.
Making an accurate sales forecast is both an art and a science that combines the well-developed intuition of an experienced salesperson and reliable data fed into forecasting algorithms. While it is essential that sales teams have the skills to make good forecasts relying on simple spreadsheets and back-of-the-napkin calculation methods, reliable forecasts need software solutions that can give real-time insights and projections based on data.