11 Types of Forecasting Models

Forecasting models

Thinking about creating a new product?

Or maybe you’re ready to enter into new markets or open up branches in new areas? However, you don't know the size of your market, or how the market will be evolving in the following years.

To identify how your business would work in different future situations, you should use a forecasting model.

Forecasting models take into account the following:

However, it’s important to understand how to implement the different types of forecasting models, but also to figure out which model is the most appropriate for a certain situation or problem.

Apart from forecasting models, in this article, we'll cover the top two financial forecasting methods — quantitative and qualitative. We’ll also mention some forecasting tools and how they can facilitate your future projections.

What is forecasting?

According to the PMBOK guide, “a forecast is an estimate or prediction of conditions and events in the project’s future, based on information and knowledge available at the time of the forecast.”

We may use forecasts in various situations. For example, in finance, companies use financial forecasting to project employee’s wages or set the annual budget. On the other hand, in stock trading and investing, forecasting is used to predict the future market price and performance.

Moreover, in a business setting, forecasting can help business analysts study the impact of certain changes in the working environment (such as adjusting business hours). Another type of forecasting is weather forecasting, which predicts future atmospheric changes for a certain area and time, or changes on the Earth surface, based on meteorological observations.

In this blog post, we’ll stick with the financial forecasting.

What is financial forecasting?

Financial forecasting refers to projections made about the future performance of a company in order to estimate how current trends and business metrics will affect the financial position of the company. This can be done by exploring historical information regarding business performance (sales, revenue, or expense figures) as well as current business trends, and other important business variables.

Now, why do we need financial forecasting in the first place?

Financial forecasting focuses on overcoming business challenges regarding strategy planning processes, including:

Companies use forecasting to determine if their expectations align with the possible outcomes. In other words — forecasting helps them make predictions about how current business and market trends (revenue, costs, consumers, demographics, etc.) will affect the company’s performance or operations.

Why does your company need forecasting?

Forecasting is an important part of business planning and operations because it helps businesses estimate their financial situation. With forecasting, companies can analyze current and past data in order to make predictions about future trends and changes.

Since business decisions are based on current market conditions and forecasts about future events, stakeholders will be more comfortable in making informed decisions and developing better business strategies. For example, forecasts may help you decide whether to fund a specific project, increase the staffing, or estimate the annual budget.

Top benefits of forecasting

Let’s see how forecasting can help your business succeed:

 To sum up, forecasting is an absolute necessity for any business because it helps you:

There are various tools that help businesses get better insight into how operations and processes currently work, and find out what needs to be changed or improved. We will mention a few forecasting tools below. 

Forecasting method vs. forecasting model

Sometimes, the terms forecasting method and forecasting model are used interchangeably, however we want to point out that these terms are completely different.

In fact, a forecasting method uses mathematical calculations (created for a specific purpose) that does not elaborate on what actually happens in the data, but rather it is solely used to produce forecasts, with or without a forecasting model.

On the other hand, a forecasting model breaks up the data into a structure and allows you to examine the process further.

Given the rising competition and volatile customer loyalty, it has become challenging for businesses to work out any reasoning behind certain events. For that reason, forecasting models can be used to change and control different business variables and give a clearer picture of the company’s future. Many organizations employ forecasting models to predict various business metrics including sales, profits, consumer behavior, supply and demand, and then set yearly goals.

For example, forecasting models can help you understand whether your marketing strategies are effective or whether your sales are struggling. In addition, they help businesses allocate their resources properly and plan the upcoming period of time regarding the aforementioned business metrics.

What is forecasting in project management?

What are the two main forecasting methods?

Main forecasting methods

Businesses can choose between different types of forecasting methods, including:

Quantitative methods of forecasting 

In order to make realistic and accurate forecasts, quantitative methods include mathematical processes such as:

In addition, mathematical techniques such as linear programming, dynamic programming, and inventory control can also help decision-makers guide their business strategies.

Quantitative methods also include statistical processes such as:

So, rather than basing the results on opinion and intuition — quantitative methods implement readily available data to interpret results. These methods are usually used to make short-term predictions by analyzing older, raw data.

Finally, quantitative methods can be further divided into:

Qualitative methods of forecasting

Unlike quantitative methods, qualitative methods are subjective in nature and rely largely on:

These types of forecasting methods do not implement any mathematical calculations, and are mostly used when the historical data is too narrow or not expected to be followed in the future.

Qualitative methods are also used when the available data cannot be projected into numerical analysis, or when the trends and habits are constantly changing.

Qualitative forecasting methods can be further classified into 5 forecasting models:

  1. Market survey,
  2. Sales force opinion,
  3. Delphi method,
  4. Visionary forecasting, and
  5. Panel consensus.

We will explain these types of forecasting models in further details below.

Types of quantitative forecasting models

The entire range of forecasting models is enormous and is growing rapidly every day.

As we mentioned earlier, quantitative forecasting methods are based on mathematical or numerical values, and they are objective in nature. Thus, the types of forecasting models under this category rely on mathematical computations and calculations, using data from past company operations.

Let’s take a closer look at the different types of quantitative forecasting models. Bear in mind that these models are basic (there are also some more advanced models), however they can all help you forecast how future trends will change in the forthcoming years.

Time-series forecasting models

Time-series is a popular forecasting model which explores past company behavior to forecast future company behavior (consumer behavior, sales behavior, etc.). This type of forecasting model uses historical data in terms of hours, weeks, months, and years to come at a point in the future based on these past values.

Time-series uses information gathered over several years to analyze sales velocity based on the business needs. Based on those figures, you can create future forecasts using mathematical formulas. There are several models of completing time-series forecasting which will help you formulate future estimations.

The subtypes below are all examples of time-series forecasting models:

Let’s find out more about each one of them.

#1 Straight-line method

The straight-line method is a time-series forecasting model that provides estimates about future revenues by taking into consideration past data and trends.

For this type of model, it’s important to find the growth rate of sales, which will be implemented in the calculations.

For example, the annual growth rate of a company can be fixed (6%) over the past 5 years. Consequently, the company anticipates that the growth will continue at 6% over the next couple of years. Using this information, the company can make accurate forecasts about its future decisions for the following years. This will help businesses predict how growth might affect the available data. 

So, if your company has a continuous growth rate, the straight line forecast can help you get an idea of the ongoing growth at the same rate. Aside from revenue predictions, this model can also be used to predict additional business needs in order to make quick financial decisions.

#2  Moving average model

The moving average model is similar to the straight-line forecasting, except that it’s often used to predict short-term trends (such as daily, monthly, quarterly, or half-yearly intervals). Companies use the moving average model when they need to forecast sales, revenue, profit, or other important business metrics.

With regards to calculating future revenue, for example, the model focuses on observing the past and current revenues (i.e., the average number of the revenues in a given time period) to predict future outcomes. This type of forecasting model is useful when calculating the performance of a specific metric within a certain time limit.

For example, if you want to forecast the sales for the upcoming month, you may take the averages of the previous quarter. This will help you identify the demand during peak selling periods.

Here’s how you can do the calculations to get the moving average:

A1 + A2 + A3 … / N

A = Average for a specific period

N = Total number of periods

Let’s say you want to calculate the moving average of sales figures for a period of 4 years (2019-2022) taking 2 years at a time (a two-year moving average). You would need to find averages for the following data subsets: 2019-2020, 2020-2021, and 2021-2022.

Year Sales ($M)
2019 4
2020 7
2021 6
2022 5

So, how would you calculate the moving average of sales with the above data set over a period of 4 years?

In order to find the average sales value for all 4 years, you will need to include all the 4 total numbers of sales, and then divide them by 4.

(4M + 7M + 6M + 5M) / 4 = 5.5M

Then, we get a moving average of $5.5 M.

#3 Exponential smoothing model

Similar to the moving average, exponential smoothing is another time-series forecasting model which can be used to predict new values by using a set of weighted averages based on past observations.

Exponential smoothing helps to predict the future by using past company data. The weights start declining exponentially with past observations in order to predict the upcoming period. To put it simply, if the observation is a more recent one — the associated weight is higher. This means that more weight is given to recent values instead of past values.

New forecasts are predicted by including the past forecast and the percentage of value (the difference between the current and the past forecast). The idea behind this model is to attribute importance to more recent values in the series — when observations become older, past values get exponentially smaller.

How to create a project timeline

#4 Trend projection model   

The trend projection model works best in situations where you could work out the future influence of certain variables (dependent or independent) based on its past behavior. The model examines past events in order to identify patterns and trends that could recur frequently.

Trend projection can be used to forecast future activity since it considers that all factors involved in past trends will continue in the future as well. The model requires long and reliable time-series data which is arranged in chronological order for evaluation.

By identifying the patterns of trends, the company will be able to get a vision of the future. Consequently, once the trend has been identified, it will be able to predict future demand.

Associative (causal) forecasting models 

According to associative or causal forecasting models, the forecasted variable is interconnected to other variables in the business system. Therefore, the forecast projections rely on these associations. 

Associative models are an advanced way to forecast your sales because they implement specific mathematical calculations to identify the connection between different variables that can affect your business activity.

The subtypes below are all examples of causal models:

#1 Simple linear regression model

The simple linear regression is a type of associative forecasting model that provides a more detailed context to your forecast by examining how the independent variable is correlated to the dependent variable.

The dependent variable is the predicted value (e.g. sales), and the independent variable (e.g. profit) guides the expected value of the dependent variable.

Simple linear regression can be visualized on a graph by portraying one metric on the X axis, and the other one on the Y axis.

Here’s the formula for calculating simple linear regression:

Y = BX + A
Y = dependent variable (predicted value)
B = the slope of the regression line (measure of its steepness i.e the ratio of the rise to the run, or rise divided by the run)
X = independent variable
A = Y-intercept (the point on the Y-axis by which the slope of the line sweeps)

Apart from identifying the relationship between sales and profits, it can also demonstrate the rate of increase and how that rate varies in order to help you find ways to maximize profit.

Calculating the simple linear regression is a tedious process, so you might want to use statistical programs to help you analyze the data.

#2 Multiple linear regression model

As its name suggests, the multiple linear regression model follows the same approach — i.e. makes the same assumptions as the simple linear regression — except that it applies it to a number of different business variables.

So, when business performance is influenced by more than one variable, this model allows you to explore the relationship between two or more independent variables and one dependent variable. This will help you get a clear picture of the situation and a more accurate forecast.

For example, multiple linear regression can be used to determine daily cigarette consumption, which can be predicted by independent variables such as smoking duration, starting age of smoking, type of smoker, etc. Or when the margin for a certain product is affected by variables such as labor cost, materials, machine efficiency, etc.

However, it can be difficult to perform a multiple regression by hand as these models are complex, especially when there are too many variables involved, so you’ll likely need statistical software.

Types of qualitative forecasting models

Another method of forecasting is qualitative forecasting.

Qualitative forecasting methods are different from quantitative methods because they’re subjective and intuitive in nature. They are based on the opinion and judgment of consumers and experts.

In addition, qualitative methods use factors such as demand trends and seasonality to create more accurate forecasts.

They can only be used if past data isn’t readily available.

All the forecasting models below belong to the category of qualitative forecasting methods.

#1 Delphi method

The Delphi method is a type of forecasting model that involves a small group of relevant experts who express their judgment and opinion on a given problem or situation. The expert opinions are then combined with market orientation to come up with results and develop an accurate forecast.

The Delphi method is performed in such a way that each expert is questioned individually to gather their insights. This helps to prevent bias and ensures that the company’s forecast is based solely on their own expert opinion.

Furthermore, other employees or outsourced parties collect, summarize, and analyze experts’ responses. They may pose additional questions to the participants who can then reconsider their original responses in order to come up to a meeting point or final consensus that would be beneficial to the company.

Highlights of Delphi method

Some of the highlights of this model include:

When to use Delphi?

The Delphi method can be used to:

Other situations where the Delphi method makes sense is when you:

#2 Market research model

Market research is a qualitative forecasting model that evaluates the performance of a business’s products and services by interviewing potential customers about them. Their reactions and responses are recorded, and then they’re analyzed in order to come up with a sales forecast.

This model can be performed by staff members or third-party agencies (specialized in market research) by:

Some examples of market research strategies include:

However, the strategies might be adapted based on the current market conditions and challenges.

These techniques are used to gather valuable insights from consumers so that the company understands which products or services to continue launching and which ones need to be revised.

Highlights of market research 

The market research model can help companies:

When to use market research?

The market research model can be used:

Conducting market research is also beneficial when you need to:

#3 Panel consensus model

Panel consensus (also called expert opinion) is a qualitative forecasting approach where experts or employees from all levels of an organization (from low-level to top-level) discuss a product or service. The members act like a focus group, expressing their thoughts and recommendations in order to develop a forecast.

Anyone can speak up during the discussion, however, sometimes lower-level employees may feel intimidated to express their opinion due to their lack of market knowledge. This is one of the drawbacks of this model.

On the other hand, the forecasting process involves a high number of participants, therefore, the outcome would be more balanced and reliable compared to an individual person’s opinion. The meeting will end once a consensus has been reached.

Highlights of panel consensus

Some of the highlights of the panel consensus model include the following:

When to use panel consensus?

The panel consensus model can be used:

Other examples where panel consensus may be appropriate is when companies need:

#4 Visionary forecast model

The visionary forecasting model is based on personal opinions, judgements, and insights of a relevant and experienced individual. The projections are backed up by data, information, and facts in order to predict future scenarios. When available, historical analogies can also be used to hypothesize potential future forecasts.

In other words, the ‘visionary’ prophesies a set of future events by examining past events and developments. Therefore, this model is subjective and non-scientific in nature, and is solely based on an individual's guesswork and imagination.

The only downside of visionary forecasting is that there might be a confirmation bias because visionaries may only look for evidence that supports their own beliefs and disregard any contradicting evidence.

Highlights of visionary forecast

The visionary forecasting model is:

When to use visionary forecast?

The visionary forecast model can be used to:

Additionally, this model can be used in the absence of historical data. Some steps in the business planning process do not require the use of historical data (analyzing the current financial situation, studying the company’s competition, making future scenarios, or using existing industry trends), so this model could come in handy.

#5 Sales force composite model

Another reliable qualitative forecasting model is sales force composite where the input of sales staff is used to estimate future sales. When estimating future demand, the company may decide to collect information from the salesperson that would help in determining customer’s needs and predicting the sales in a certain region and given time period.

According to the sales force composite model, the sales agent better understands the needs of the customers since they interact with them on a regular basis. This information will help in adjusting business operations in order to meet the client’s needs and maximize sales. 

The sales person is questioned about the experience and satisfaction of the customers with the company. This model is easy to conduct, since it only requires an appointment with salesforce experts. Each person gives their own opinion about what they expect to sell in their specific region. 

Highlights of sales force composite 

Some of the highlights of sales polling are the following:

When to use sales force composite?

The sales force composite model can be useful when you need to:

How to choose the right forecasting techniques?

Many forecasting techniques (or methods) have been developed over the years so it becomes challenging for managers to select a proper technique for a particular situation. That’s why it’s essential to understand the possibilities of each method and how they can help when forecasting a specific problem or situation.

There are some key considerations that need to be taken into account when choosing a forecasting method and model.

The following factors influence which forecasting method and model will be used:

However, before selecting the method and the subsequent model, you will also need to consider the purpose of the forecast and what variables should be included. These considerations will help the forecaster choose the right method and model that will generate accurate results for the business.

Project Management: 31 best techniques, practices, and tools

What are the best forecasting tools?

As we previously mentioned, accurate forecasts are a crucial part of every organization, because they help them estimate and prepare for the future. However, you'll need certain tools that will help you make projections and plan ahead of time.

Furthermore, forecasting tools can automate such processes, which in turn, improves efficiency. Your team can stay focused on completing projects and addressing your customers’ needs while you get reliable and real-time data for forecasting.

When it comes to forecasting tools, you have plenty of options to choose from. But you need to find the right one for your business. Luckily, we’ve already done the research for you.

So, here are the most common types of forecasting tools you may use:

There are a variety of tools available, so it’s important to find the one that aligns with the scale of your organization and available budget.

How to calculate and rate employee performance

How to create forecasts in Clockify

Clockify can help you get a better understanding of how your project is performing based on the time you track in the app.

This time-tracking software can help you analyze your project performance each month and make more accurate forecasts by keeping track of all project stages. In other words, this means that tracking project performance can help you identify unforeseen changes and find ways to approach them.

Forecasting feature in Clockify

Additionally, Clockify allows you to get insight into how much time each task requires and avoid time overruns and delays. This can help you plan your upcoming projects.

In Clockify, you can track time for each task to avoid overtime and lengthy hold-ups.

Clockify also allows your team members to clearly see their tasks and get a better idea of what they are expected to accomplish in a given time period. Tracking project performance will also help you estimate and allocate the annual budget.

Forecasting feature in Clockify

Final words: The best forecasting model is the one that aligns with your business goals 

There are dozens of forecasting models, therefore, it’s important to know how to choose the right one for your business.

There isn’t a single model that will work for every business, industry, or situation, since each model has its own strengths and weaknesses. Moreover, not all models would necessarily serve the business’ purpose, so you will need to choose the one that aligns with your specific needs and goals.

We gave our best to carefully examine 11 forecasting models for you because there is no one approach that fits all of your business problems. You can choose the one that nearly solves your problem, and then experiment, or adjust the others over time.

Additionally, you may combine the forecasting methodologies that you think are more accurate for your unique situation, and in that way, eliminate the shortcomings of one model by substituting it with another.

After all, it’s up to you to decide what type of forecasting model is best suited for your business.