8 Proven Sales Forecasting Methods for Greater Accuracy

Making accurate sales revenue forecasting models is almost as important as meeting the revenue target. There are many different methods to use, but which one will give you a clear picture of what’s coming?

A new study by CSO Insights shows that 60% of deals forecasted do not actually close. This is unsurprising because 25% of sales managers are unhappy with their forecasting accuracy. Choosing the right predictive sales model can help you predict future revenue more accurately.

The following post will discuss three sales forecasting models that have proven to be effective for us at HubSpot. I?ve found that the best results come from a combination of all three.

I will give a brief description of each sales forecasting models we use, but I recommend you experiment with them and tweak them to fit your company before implementing any changes.


Need Help Automating Your Sales Prospecting Process?

LeadFuze gives you all the data you need to find ideal leads, including full contact information.

Go through a variety of filters to zero in on the leads you want to reach. This is crazy specific, but you could find all the people that match the following: 

  • A company in the Financial Services or Banking industry
  • Who have more than 10 employees
  • That spend money on Adwords
  • Who use Hubspot
  • Who currently have job openings for marketing help
  • With the role of HR Manager
  • That has only been in this role for less than 1 year
Just to give you an idea. 😀

What Are The Three Kinds of Sales Forecasting Models?

Forecasting is based on the idea that there are three types of input data. The forecasting technique depends on what type of input data it’s using in order to predict future sales. So, how to forecast sales using historical data? Forecasters use these techniques:

  • Qualitative techniques
  • The technique of projecting the data in a series to estimate future events is called time-series analysis.
  • Causal sales forecasting models example

The qualitative technique relies on data that is more specific and refined than the time series analysis, which focuses on patterns. The causal model requires highly detailed information about relationships between system elements.

The three sales forecasting techniques imply that there is no single technique that can be applied to every situation.

When a new product is introduced, the time series analysis that relies on historical data may not be as helpful for forecasting future sales.

What are the general functions of three types of sales forecasting techniques?

#1. Qualitative Techniques

Qualitative techniques are used when the data is scarce. For example, qualitative forecasting methods can be useful for new products because there isn’t a lot of information to work with.

When estimating a project, the estimator uses qualitative data and their own judgment to estimate costs. The goal is for all information about an estimation to be accounted for in order make it more accurate.

If you are unsure of your product’s market acceptance or penetration rates, qualitative techniques may be useful. You can also use them when developing new technology products where multiple inventions will need to come together.

Qualitative techniques include:

Panel consensus

Panel consensus is a sales forecasting methodology technique that can be used by commercial organizations to predict the future of their product. This method relies on experts from various fields coming together and providing input, rather than just one person’s opinion. Communication between these experts is encouraged in order for everyone to get more viewpoints.

Delphi method

The Delphi method is a forecasting model that predicts future sales. It does this by surveying experts and then asking the group to come up with their best estimates of what will happen.

Salesforce composite technique

In the sales force composite technique, company representatives are asked to make their forecasts. The assumption is that these reps have direct contact with customers and other members of the distribution channel because they know more about what people want than anyone else.

Buyer?s expectations

To estimate demand for a product, survey potential buyers and try to predict the amount of sales that will occur in the future. This is done by extrapolating information about their intentions with regards to buying this specific type of product.

Market research

Market research is a procedure that involves testing hypotheses about the real market and looking at what people actually want. It doesn’t just involve guessing.

#2. Time Series Analysis

The Time Series analysis sales forecast technique is used when you have a lot of data available about the product or if there are clear trends and relationships with it. This can be done in order to make predictions for future demand.

The forecaster uses the past data of a product?s performance to get an idea for what its current state is and how it has changed. The acceleration or deceleration in rates is used as the basis for forecasting.

A time-series analysis is a set of chronological points in raw data. It helps to explain the correlation between two variables, especially when one variable has more variability than another.

  • Trends in the data
  • Performance patterns that recur every two or three years
  • The data in the following charts is not necessarily a reflection of seasonal changes, but there are still regular patterns that can be observed.
  • The amount of data in the world is growing at an exponential rate.

#3. Causal Models

The model for causal forecasting is developed when you have enough historical data about a product and analysis carried out. The factors to be forecasted should show up in the analysis, as well as other economic forces and social-economic factors.

If you want to predict sales, use the causal model. It can include market survey information and other factors that are relevant in predicting future results. The technique also incorporates the result of a time series analysis.

The theory of flow system dynamics takes into account how people interact with each other and the environment around them. It uses predictions such as promotions, strikes, to help analyze what might happen next.

What Are The Sales Forecasting Methods?

#1. The ?Lead Value? Sales Forecasting Method

The forecast model analyzes historical sales data from each of your lead sources. This information is then used to predict the value that a particular source will bring in during future periods.

The beginning of a buyer?s journey can tell us if they will end up purchasing. It is similar to how the outcome of romantic comedies are predictable based on their early scenes.

By understanding the value of each lead, you can better predict how many leads will turn into revenue.

The following metrics are needed for this model:

  • In the past month, how many leads did you generate?
  • Improve the conversion rate of your sales leads by determining which source is best for them.
  • Average sales price by source

The Calculations:

Average sales price per lead

To find the average sales price by lead source, you can just take all of your customer data and divide it into categories based on where they came from.

The average close for a website lead is $1,000 while the average close from demo requests are at $1,500.

The information in your CRM is only as good as the reporting functionality that it offers. If you want to find out how much each sale was worth, then export this data into an excel file and quickly get the average sales price from there.

Average Lead Value

The average close rate for a lead source is the number of people who buy divided by the total amount of leads from that type. Multiply this by an average sales price to get your lead value per sale.

Average Lead Value = Average Sales Price * Conversion Rate from lead to customer

For example, if I know that the average customer who comes to us from paid advertising spends $2,000 and they convert at a rate of 10%, then each lead is worth $200.

$2,000 x 10% = $200/lead

Total Number of Leads

If you know your goal and the average value of a lead, then divide that by how many leads it takes to reach your revenue goals.

Leads Needed = Desired Revenue / Average Lead Value

Continuing from the example above, let?s assume our sales team needs to hit $100,000 in revenue next month. To do this we need 500 leads on average per lead value of $200 which means that each individual will have to generate an average of 10 leads a day for 30 days straight.

100,000 / 200 = 500

Marketing should be consulted to learn what upcoming initiatives they have planned and where lead flow is expected to come from. Lead values vary depending on the channel used.

Once you?ve made a spreadsheet to keep track of your salary and commission structure, the numbers will look like this:

sales forecasting models

Considerations:

In the beginning, I made a common mistake of just focusing on pay and commission rates. However, as time went by it became clear that other factors such as customer service also have an effect.

To be successful, you need to factor in the time it takes for each lead source. If you are using this type of forecast, take into account how long a customer will wait before purchasing.

It’s not just about what you’re paying your salespeople. The conversion rates of other business initiatives such as improving the process, changing prices or discounts can change lead values and will affect results.

Marketing teams will be changing their plans based on new data and trends. It?s important to stay in line with them, or you might end up missing your expected lead volume.

If you have a lead source that is not easily identifiable, then it can be included in the bucket of ?other? and put into your forecast.

#2. The ?Opportunity Creation? Sales Forecasting Method

Concept is a predictive model that takes into account various factors, such as location and past purchase history. It then identifies which opportunities are more likely to close.

In a romantic comedy, it is easy to predict what will happen next because the character?s appearances and personalities are so predictable.

Predicting the probability of a deal closing is difficult. By looking at demographic and behavioral data, we can better estimate how likely it will close or what that closed opportunity might be worth.

The Calculations:

When it comes to business, I want to know what’s worked in the past. This way, when we’re looking for new customers who will close deals with us, we can find ones that have similar characteristics.

I want to share an example of how we’ve implemented this model at HubSpot.

We have found that the best way to evaluate an opportunity is by looking at how many employees and annual revenue a prospect has. These two factors are great predictors of our success.

It turns out that a paycheck is not the only thing that matters. There are many other factors which determine whether or not someone will close an opportunity, such as their role in the decision-making process and behavioral patterns.

When it comes to customer service, there are many factors that go into a successful company. One of the most important aspects is knowing your ideal customers and what they want from you.

Lead scoring is a second layer of analysis that determines which leads are worth pursuing. Marketing and Sales teams work together to set up lead scores for each type of customer, usually based on how much time it will take or money they might spend.

HubSpot assigns leads a score from 1-100, with 100 being the best fit. These scores are then grouped into four buckets labeled ?A? for good lead prospects and D for bad ones.

Once you have your scoring system in place, the opportunities that are worth investing time and resources into can be identified.

Expected Value of Opportunity = Average Sale Price * Average Close Rate

If you want to be able to forecast your expected value per opportunity, then first calculate the close rates for each of your lead buckets. For this method to work, there must also be an average sales price that is known.

sales forecasting models

The article is primarily about how pay and commission structure are not enough to motivate employees.

I am now realizing that there are many different factors of motivation, and it is not just about the paycheck.

I like how this model helps my reps focus on the most important opportunities because it shows them where they can do more.

Considerations:

In order for this model to work, you need clear criteria on what opportunities your sales reps can create. Even with that in place, it is still up to the rep whether or not they will follow procedure and stay consistent. You?ll have to keep an eye out.

Hiring salespeople is also not enough. You have to create a system for evaluating performance and provide incentives so that they can achieve their goals.

To help ensure that the data is accurate, you should test your new opportunity scoring system with one salesperson for a set amount of time before rolling it out to the whole team.

#3. The ?Opportunity Stage? Sales Forecasting Method

One of the most popular forecasting methods in sales is called “Concept.” It predicts an opportunity’s probability to close based on where it currently stands in your sales process.

You need to know your average sales cycle before you can start forecasting. If the company has a clear process, then they will be able to accurately forecast and make better predictions.

It?s important to know that when it comes to deals, not all stages are created equal. Here is an example of the deal process and probability:

  • Appointment Scheduled (20%)
  • Qualified to Buy (40%)
  • Presentation Delivered (60%)
  • Contract Sent (90%)
  • Closed Won (100% Won)
  • Closed Lost (0% Lost)

The Calculation:

When creating a forecast, you multiply the amount of each opportunity by that opportunity?s probability of closing.

Expected Revenue = Deal Amount * Probability to Close

For this forecasting technique to work, you need a well-defined sales process with detailed steps for each stage of the deal. You then assign probabilities that show how likely it is that your company will be able to close on any given sale.

Below is a template that can be used to map out your sales process. You can download an editable version here.

Below is a list of ways to increase sales motivation:

sales forecasting models

In order to accurately forecast, you’ll need a CRM system that allows for assigning win probability as the sales cycle progresses.

After 6 months, you should take a look at the rates to see if they are high enough. If not, adjust them so that your team can be more productive and have higher conversion rates.

Considerations:

The data that you have been relying on for your forecasts may not be as accurate as it seems. In order to make sure that this is the case, make sure you update and revise old opportunities regularly.

If you are considering this model, it is crucial to look at historical data and calculate the probability of success based on past opportunities.

Make sure you have a list of requirements that need to be met before the deal can move on. If not, there is no way for an accurate reading and analysis.

#4. Length of Sale Cycle Forecasting

A company can use data on the time it takes a prospect to convert into a paying customer as well as other factors in order to estimate how long they will be selling for.

When a sales cycle lasts for six months and your reps have been engaging with the lead for three of those, there is a 50% chance that they will close. This forecasting method is objective because it doesn’t depend on how emotional or engaged someone might be.

It is also advantageous because it can be used for a variety of sales sources, so it’s perfect for companies that want to know exactly when they acquired their customers.

Calculation:

The length of the sale cycle is determined by when you start counting. To calculate it, use this equation:

# of days from the first contact + customer conversion = # of days of combined sales

If you look at the combined number of sales and average deal size, this will give you an idea of how long your company’s sales cycle is.

Consideration:

Leads can be triggered by a customer downloading an ebook from your website and then requesting to see the product in person months later. These types of leads create second-hand opportunities for sales reps.

#5. Intuitive Forecasting

To find out if a sale will happen, you should ask the salesperson because they are in tune with what is happening on their end.

So, you can ask your salesperson if they are confident in closing a deal and when. Your reps know how things go with prospects better than anyone else, so the intuitive forecasting method relies on trusting their opinion.

The downside of this approach is that it can be subjective. If the sales reps are optimistic, they may provide high estimates and there’s no way to assess them.

Calculation:

With intuitive forecasting, sales reps will try to estimate how much they’ll be able to bring in for a certain period. For example, I plan on bringing $X dollars during the next X days.

Consideration:

A salesperson will analyze the projected profit and break down how much it could be worth. The sales rep should take into account all factors before giving an estimate.

#6. Test-Market Analysis Forecasting

Concept: Test-market analysis is a forecasting method that allows you to roll your product or service out to only the people who need it. You can move the product in one location and see how well it performs.

In order to better forecast future sales, a company can use their new product rollout data and find out how the public responds. This is an ideal strategy for big companies who want to roll out a new product but are unsure of its success.

Calculation:

The idea behind this method is to divide the market into two regions, a test and an advertising region. The first one would be where you bring the product without any ads.

Then you have the control market where advertising has already been done. The difference between sales in this market and test markets is analyzed to predict future success of a product.

Consideration:

The Test-Market method is best for launching new products or exploring new markets. It requires a considerable investment, but has proven to be worth it.

#7. Historical Forecasting

This forecasting method takes the past sales and assumes that they will be higher for this period.

The old forecasting method is outdated because it does not take into account the changing market. If your competitors run a promotional campaign, you may notice that sales decline.

Calculation:

If you know the MRR for June is $30,000 and your year-on-year growth rate is 10%, then you can expect that July’s revenue will be at least $30,300.

Considerations:

The market is always changing, so you should take this into account when using the technique.

#8. Multivariable Analysis Forecasting

When I was looking for the most accurate forecasting method, multivariable analysis forecast turned out to be an excellent choice.

The system is not just based on the size of commissions, but also factors in other techniques like opportunity stage forecasting and sales cycle length. It can even be tailored to individual rep performance.

Calculation:

The multivariable analysis is complicated, and you need clean data. If your sales reps are not tracking the progress of deals or activities, then their results will be inaccurate.

For example, if one salesperson has a $20k deal in the pipeline and another is working on an 8k sale with 65% chance of closing, your forecasted amount would be 5200.

The total forecast for this account is $5,000 +$5200 = 10,200.

Considerations:

This forecasting method may be impractical for small businesses because it involves complex math. If your sales reps don’t track all the deal progress, then you’ll get inaccurate results.

Sales Forecasting Using a CRM System

One of the most common ways to forecast sales is with a table. If you are just starting out, this method can be helpful because it?s easier and faster than other methods that involve more detailed analysis. However, if your company has been around for awhile or you have an established customer base then customize the reporting section in your CRM.

I?d love to hear from you about your experience with these models and how they have helped. If you have any other methods that work, please let me know in the comments below.

Also published on Medium.

  • Sign in to comment. Comment Comment Show: Oldest Newest Most Upvotes 0 @Elle Speicher ( 0 POINTS ) 4 years, 3 months ago On the Lead Driven Model, what is the Leads Expected Column? What purpose does it serve if you know your total number of leads and your closed won? Thanks 0 @Shaun Ling ( 0 POINTS ) 3 years, 11 months ago which crm software you recommend?
  • 0 @Elle Speicher ( 0 POINTS ) 4 years, 3 months ago On the Lead Driven Model, what is the Leads Expected Column? What purpose does it serve if you know your total number of leads and your closed won? Thanks

What is the Leads Expected Column on Lead Driven Model? What purpose does it serve if you know your total number of leads and closed won cases?

  • 0 @Shaun Ling ( 0 POINTS ) 3 years, 11 months ago which crm software you recommend?

what is the best customer relationship management software?

sales forecasting models


Need Help Automating Your Sales Prospecting Process?

LeadFuze gives you all the data you need to find ideal leads, including full contact information.

Go through a variety of filters to zero in on the leads you want to reach. This is crazy specific, but you could find all the people that match the following: 

  • A company in the Financial Services or Banking industry
  • Who have more than 10 employees
  • That spend money on Adwords
  • Who use Hubspot
  • Who currently have job openings for marketing help
  • With the role of HR Manager
  • That has only been in this role for less than 1 year
Just to give you an idea. 😀
Editors Note:

Want to help contribute to future articles? Have data-backed and tactical advice to share? I’d love to hear from you!

We have over 60,000 monthly readers that would love to see it! Contact us and let's discuss your ideas!

Justin McGill
About Author: Justin McGill
This post was written by Content at Scale, a solution that uses AI + a team of optimization specialists to publish hundreds of high quality, SEO optimized content straight to your blog. It’s the first and only solution that allows you to truly scale content marketing.