Types of sales forecasting
The types of sales forecasting continue to grow and improve as artificial intelligence gets more, well, intelligent. In the past, business and sales leaders had to rely on historical data and personal opinion to forecast the potential sales coming down the pipeline, and the profits those sales would bring in. Today, technology is taking the guesswork out of these reports, and empowering teams to move beyond a preoccupation with accuracy to focus on the real work at hand—growing revenue and generating satisfied customers.
Read more below.
It's time for accurate sales forecasting
The most powerful forecasting tool from the world’s largest network of sales professionals.
- Dramatically improve forecast accuracy
- AI-driven forecasting
- Increase accountability across selling teams
- Eliminate inaccurate CRM data
Automated daily sales forecasting
Updated daily based on your data and external network factors that influence your sale.
- Faster predictions
- Reduce sales cycles and analyze scenarios in minutes
- Real-time, buyer-specific selling recommendations
- Increase deal velocity
Eliminate spreadsheets. Automate your CRM.
Eliminate spreadsheets. Eliminate opinions. Get data-driven results with little to no effort.
- Automated data capture
- AI-driven forecasting
- Network enrichment of contacts
- Networked intelligence - unlock external factors that matter
Types of sales forecasting
New and modern approaches to sales forecasting can improve business outcomes and make life better for sales teams. But what are the types of sales forecast you should consider? Let’s examine the general role forecasting plays at a business level, how the sales forecast fits in, and how the unique characteristics of your business help you choose the type of sales forecast that is right for you.
Types of business forecasting
There are several types of forecasting for a business, including everything from historical data used to project future sales to surveys that gauge actual customers’ intent to purchase. Understanding which type of sales forecasting is right for your team starts with understanding how your sales forecast fits into the broader system of predictions that are informing and supporting your company’s business plan.
What are the different types of forecasting a business could be using to inform decisions?
Here is a list of basic types of forecasting:
General business forecasts
The most general forecasts analyze the market conditions and cultural factors that impact a business, like seasonal shifts, technological advancements, or political changes.
Sales forecasts estimate future sales, predicting the revenue a company can expect over the next week, month, quarter, or year.
These reports relate closely to a sales forecast, projecting what the demand for goods and services will look like in target markets, as well as how much consumers might be willing to pay.
By predicting the future costs a company will incur, from rent to employee salaries and benefits, to the cost of marketing and raw materials to create products, these reports help businesses plan spending
An eagle’s-eye view of accounts payable, accounts receivable, anticipated expenses, market conditions, and more, financial forecasts track the profits that will be earned in a certain timeframe.
For businesses, these types of reports are essential to maintain profitability and predict challenges in advance. But if creating and reviewing all these types of forecasts sounds like a lot of work, that’s because it historically has been. That’s why it’s exciting that there are sales forecasting tools being developed and released that can carry out different elements of these business forecasts in real-time. That means all the talent on your team can focus on the insights and actions that come out of having accurate forecasting at their fingertips.
What are the three kinds of sales forecasting techniques?
The three kinds of sales forecasting techniques are AI-enabled, quantitative, and qualitative. A majority of businesses are still using quantitative and qualitative sales forecasting strategies to make predictions. Based on historical data, these methods can take a lot of research and effort, but they also leave sales teams in a constant game of catch up, hustling to reach a prediction based on imperfect data.
So, what are the three types of forecasting? Let’s examine each and how they serve businesses in the competitive, fast-paced modern economy—or leave them wanting more.
Qualitative sales forecasting is an opinion-based approach to predicting what sales volume can be expected in the next week, month, quarter, or even year. In the best-case scenarios, the opinions and estimates of the sales team themselves are included when making these projections. At the same time, asking a sales person how much they expect to close in a certain time frame can generate an answer that is overly-hopeful, or overly-pessimistic. In other scenarios, a team of executives or decision-makers who aren’t directly involved in the sales process make their estimation of what sales should be, and hand down a “forecast” that is really more like a goal. In both cases, the sales team can feel like they are always running behind, working to make sure the provided forecast stays accurate, no matter the cost.
Quantitative sales forecasting is a data-based approach to predicting future sales and revenue. The most traditional source of data for these forecasts is the history of past sales. Some companies take the approach of simply looking at the same quarter from the year before, then increasing it by the percentage they believe—or maybe just want— their sales to grow in the upcoming quarter… This would be known as a bottom-up approach. By contrast, a top-down quantitative sales forecast would also take into account information about the market and other factors. Anticipated timeline to close and information about price increases are other types of data that might inform a quantitative sales forecast. Though this approach is more rooted in numbers, it can be limited in its accuracy by events or market shifts that are impossible to predict.
The best types of sales forecasting happening in business today allow teams to go beyond historic data. It’s unrealistic to assume the future is going to be exactly like the past. Which means sales forecasts that over-rely on past data start from a flawed presumption. AI-enabled sales forecasting starts with insights from historic data, but can layer in information taking place in the market today to give a more realistic picture of factors that can impact sales revenue. Some basic AI-enabled tools can use machine learning to automate the busy work like CRM management and data entry that can keep sales teams away from the important part of sales—interacting with buyers. But it’s networked data that truly can make a difference for today’s sales teams and their business revenues. When networked intelligence comes into play, AI-enabled tools can analyze market data on a daily basis and learn to make better predictions about what impact outside events will have on the current sales pipeline. And the very best tools analyze buyers, competitors, and social media to give sales teams actionable advice about how to make the most of each opportunity.
Among these three sales forecasting techniques only AI-enabled forecasting has the capability to minimize work for the company while also maximizing results. Qualitative and quantitative sales forecasts require many hours of research, analysis, and critical assessment of assumptions to be reliable. And even when the work is done, accuracy is the highest goal. While an accurate sales forecast has long been the goal for integrated sales teams, it is more helpful—and beneficial for planning—to know the individual activities that populate a forecast. Sales teams have long been trained to hit a number, no matter what. But artificial intelligence offers the opportunity to optimize every action along the way, helping sales people prioritize individual activities and people in their network who can help close deals.This takes the forecast from a goal on the horizon to an actionable insight that helps sales professionals exceed expectations and drive even further as a team.
How to forecast sales using historical data
There’s no question that historical data is important to a sales forecast, even an AI-enabled one. Historical data lets sales teams, or machine learning algorithms, evaluate past deals and actions in a specific time period.
However, the best sales forecasts don’t start and end with a company’s internal historic data. What if you could also leverage publicly-available data like how long the prospects in your pipeline typically take to make a buying decision? Or the average price they pay for goods or services like yours? This innovation of adaptive forecasting goes beyond reviewing a sales team’s past actions and adapts the sales forecast to be more buyer-centric as well as including relevant information from the world at large. As those insights change day-to-day, so will the forecast.
We believe adaptive forecasting in sales is the future. Insights like the ones described above would let a sales professional know when to move on from a lead in the pipeline, because they typically make buying decisions within a certain timeframe. Or, when a lead is on the brink of expiration, machine intelligence might let them know about a new connection in their network who can help move conversations forward. This expands the concept of “historic data” beyond the company’s sales history to include market history, consumer history, and the history of the salesperson’s personal and professional network. Plus, that information is automatically updated every day.
While the ideal modern scenario for forecasting sales using historic data as only one element in the forecasting process, there are more traditional methods as well. In the past here is how many businesses have approached forecasting sales using historical data:
Historical growth rate
This method uses the growth data from a previous period to project the growth in the next period. Let’s say you made $10,000 last quarter, and you know your sales grew 5%. You start by adding 1 to that percentage. (1 + 5%) This equals 1.05. Then, multiply your last quarter’s revenue by that amount to get the sales forecast of $10,500.
The linear extension method requires Excel or another data plotting tool. When monthly sales numbers are plotted on a graph, the X axis represents the progress of time, and the Y axis represents sales revenue. By pinpointing a target date in the future, a straight line can be drawn between the median of historical numbers to show where future points might land.
Calculating a run rate is basically taking the simple average of all past sales and using it to predict the amount of future sales. For example, take two quarters worth of historical data, add up the sales revenue from those quarters and divide it by two. This will show average quarterly revenue. To project future sales, the math from here is a simple matter of multiplication. The same basic principle can be applied to weekly and monthly historical sales data as well to provide more granular forecasting. However, this method is usually only advised for short-term forecasting, or when there isn’t much historic data available.
Time series forecasting methods
Time series forecasting methods are a specific sub-category of historical sales forecasting, used by businesses with a large amount of data to work from. These models still work primarily from historical data, but unlike the static examples we shared above, these methods help identify and explain seasonal fluctuations, changes in trends, and cyclical patterns that could impact business going forward. They accomplish this with algorithms specifically designed to identify trends within a data set and incorporate those trends in future projections.
Though these sales forecasts using historical data allow for more inclusion of the ideas of change, they still rely on the accuracy of historic data for that change to be predicted and made known. If something happens today, these models won’t be able to represent the impact of that change until reality has time to catch up to the projected period and show how close the prediction was.
Here are some commonly-used time series forecasting methods for sales:
Simple moving average
A moving average is calculated similarly to run rate, by adding up the sales totals from previous quarters or years and then dividing the total by the number of periods used. The difference between this and run rate is the amount of data used. The assumption no longer needs to be that the best predictor of the future is the average of everything that happened in the past. This method gives sellers the power to choose certain periods. For instance, sellers can calculate moving averages for every quarter or month individually based on the previous performance during those periods, allowing more perspective on seasonality or cyclical trends than a run rate model.
In this time series forecasting model, data is given more weight in the forecast based on recency. This can be achieved with functions in Excel, or by using an equation. This method is often used to represent data that includes deep trends and seasonality in a way that minimizes the impact of the highs and lows. Simple exponential smoothing is the simplest and most broadly applicable form, while double exponential smoothing is considered more reliable for forecasting recurring trends. Triple exponential smoothing is used to reflect seasonality impacts alongside the trends in double exponential smoothing. These three equations do not replace one another but are rather all used at the same time. As these equations get more and more complex, it is more likely you would use software to carry out the calculations.
Trend forecasting takes time series forecasting one step further by allowing for the exponential growth of a trend. These models can also reflect trend damping, where an increase suddenly stopped due to a change in consumer behavior or a change in the product or services. A third report that can be provided by trend projection is polynomial trend, where a period of growth led to stagnation, and then a decline. With enough historic data, trend projections can help predict continued growth or anticipate where trends might change. Despite its potential accuracy, however, this model still only works from historical data, meaning unexpected events that haven’t happened previously can still massively disrupt the forecast. The more complicated a pattern is, the more uncertain the forecast around it.
Forget the math — get back to selling with Collective[i].
The more reliable models in this list of types of sales forecasts depend on an internal employee using software to generate charts and graphs with often manual algorithms, like creating a formula in Excel. The simpler historical data types of sales forecast don’t require software, but also reflect fewer nuances of your business or past and future trends.
At Collective[i] we think the solution is clear—AI-enabled sales forecasts that not only do the math for you, but automatically adjust based on new information. This provides a forecast that is renewed daily, and informed by the most current information. Publicly-available information is leveraged to give sales teams actionable advice on how to prioritize tasks and get introductions that can move sales through the pipeline, while also protecting sales professionals’ privacy and the privacy of their clients and connections. Lastly, we can even automate CRM updates based on activity of salespeople to minimize the manual busy work even further.
We believe the sales forecast should be a tool your sales team is excited to use, not something they dread having to interact with every week, month, or quarter. With Collective[i]artificial intelligence helps you plot the path to future success. Explore our product today to learn how we can transform operations and culture at your business.