Forecastio guide - Sales Forecasting Methods.pdf

A Comprehensive Guide to
Sales
Forecasting
Methods
Alex Zlotko
CEO & Co-founder at Forecastio
Aug
2024

Table of Content
Introduction
The definition of sales forecasting 01
The importance of sales forecasting 03
Forecasting methods
Historical Sales Data Forecasting 05
Opportunity Stage Forecasting 07
Length of Sales Cycle Forecasting 09
Regression Analysis 11
Lead-Driven Forecasting 13
Multivariable Analysis Forecasting 15
Qualitative Forecasting 17
AI and Machine Learning Forecasting 19
Consensus Forecasting 21
Scenario Planning 23
Sales Velocity Forecasting 25
Forecasting Based on Deals Probabilities 27
Time-Series Analysis for Forecasting 29
Afterword 33

The definition of sales forecasting
Sales forecasting is the practice of predicting future sales revenue by
analyzing a range of data, including historical sales figures, current
market conditions, and broader economic trends. This predictive
process is essential for businesses to anticipate their revenue
streams over a specific time frame, allowing them to plan strategically
and allocate resources effectively.
Sales 
forecasting
is a...
Sales 
forecasting
is a...
Process
Predicting 
future sales
Predicting 
future sales
Business 
or product
Business 
or product
of for a
Based on
Historical 
data
Historical 
data
Market 
research
Market 
research
Helps with Planning
Planning
Resource 
allocation
Resource 
allocation
Decision 
-marketing
Decision 
-marketing
Responsibility for sales forecasting can vary widely depending on the company's size,
organizational structure, and stage of development. In a startup or small business, the task
typically falls on the , who is closely involved with the day-to-day
operations and has a deep understanding of the sales pipeline. This hands-on approach is
practical when the company is small and nimble, allowing for quick adjustments based on
real-time feedback.
VP or Head of Sales
However, as a company grows and its operations become more complex, the responsibility
for sales forecasting often shifts to a dedicated team, such as
. These teams are tasked with gathering and analyzing data from
various sources, applying sophisticated forecasting models, and ensuring that the sales
predictions align with the company’s broader strategic goals. The establishment of such a
team is usually a sign of a that recognizes the importance of data-
driven decision-making.
Sales Operations or Revenue
Operations (RevOps)
maturing organization
The definition of sales forecasting / 01

Moreover, the role of sales forecasting becomes increasingly critical as the company's annual
revenue grows. Larger organizations require more precise and reliable forecasts to manage
larger budgets, coordinate across multiple departments, and satisfy stakeholder
expectations. Accurate forecasting in these scenarios is not just about predicting sales; it's
about ensuring the company's long-term viability and growth.
In summary, sales forecasting is a crucial function that
evolves with a company’s growth. Initially managed by sales
leaders in smaller organizations, it eventually becomes the
domain of specialized teams as the business scales.
Understanding this progression is key to appreciating the
strategic importance of sales forecasting in driving business
success.
The definition of sales forecasting / 02

The importance of sales forecasting
Have you heard statements like, "Accurately predicting sales is impossible in this fast-
changing market," or "As a startup, we’re too small to worry about forecasting; our priority is
acquiring new accounts"? You're not alone. Many underestimate the crucial role of sales
forecasting in business survival.
To illustrate the importance of accurate forecasting, let's consider the negative
consequences of poor forecasting:
Whether forecasts are prepared by a Sales Leader or a RevOps team,
inaccurate predictions undermine leadership credibility. Poor forecasts highlight
issues with processes, data, teams, and tools, leading to doubts about
leadership effectiveness and lowered team morale. Recognizing the importance
of accurate sales forecasting is essential for maintaining a healthy, growing
business. Let's now delve into various forecasting methods and models used by
B2B sales organizations.
Negative
consequences of
poor forecasting:
Negative
consequences of
poor forecasting:
Leadership
credibility
Leadership
credibility
Resource allocation
Resource allocation
Investments in
development
Investments in
development
Investor confidence
and fundraising
Investor confidence
and fundraising
Inaccurate sales forecasts
often lead to overspending.
Companies may invest too
much in marketing, hiring,
tools, or innovations, driving
up costs while neglecting
efficiency, resulting in lower
profitability and higher cash
burn.
Consistently missing revenue
targets creates
unpredictability. This
uncertainty, coupled with
reduced confidence in
forecasts, diverts focus from
growth initiatives like hiring,
innovation, and
experimentation, toward
managing cash risks.
Poor sales forecasting reflects a
weak understanding of market
dynamics and revenue streams,
shaking investor confidence.
Investors rely on accurate
forecasts to gauge growth
potential and investment risks.
Overly optimistic or unreliable
projections can damage a
company’s credibility and hinder
fundraising efforts.
The importance of sales forecasting / 03

13
Sales  
forecasting
methods

Historical Sales Data Forecasting
Description:
This method uses past sales data to predict future sales.  
It assumes that historical trends will continue into the future.
When applicable:
Stable markets with consistent sales
trends.
When you have access to reliable
historical data.
01
Collect Historical Sales Data
02
Identify Trends and Patterns
03
Apply Statistical Methods
04
Generate Sales Forecast
How It Works:
Collect historical sales data over a specified period.
Identify trends, seasonality, and patterns.
Use statistical methods (like moving averages) to project future sales.
Formula №1
Simple Moving 
Average (SMA): where n is the number of periods.
Historical Sales Data Forecasting / 05

Formula №2
Weighted Moving  
Average (WMA): where weights decrease with each period further  
back in time.
Example calculation:
If monthly sales for the past three months were $1000, $1200, and $1300, then the 3-month
SMA would be:
SMA=(1000 + 1200 + 1300 ) / 3 = 1166.67
Pros:
Simple and easy to use and effective
for markets with steady sales
patterns.
Pros:
Simple and easy to use and effective
for markets with steady sales
patterns.
Cons:
Doesn’t account for changes in
market conditions; less effective in
highly volatile markets.
Cons:
Doesn’t account for changes in
market conditions; less effective in
highly volatile markets.
Historical Sales Data Forecasting / 06

Opportunity Stage Forecasting
Description:
This method forecasts sales based on the stage of each opportunity in the sales pipeline.
Each stage is assigned a probability of closing.
When applicable:
Complex sales processes with
defined stages.
When you need to track the sales
pipeline closely.
Discovery
Prospect identified
0%
Pre-qualification
Initial contact made
10%
Qualification
Opportunity assessed
25%
Proposal
Solution defined
40%
Evaluation
Solution evaluated by prospect
60%
Decision
Presented to decision maker
75%
Negotiation
Prising proposal presented
90%
Close
100%
How It Works:
Break down the sales process into stages (e.g., lead, qualification, proposal, negotiation,
closing).
Assign a probability of closing at each stage based on historical data.
Multiply the value of each deal by the probability of closing.
Formula Forecasted Sales
Opportunity Stage Forecasting / 07

If you have two deals in the pipeline worth $10,000 and $20,000 at 50% and 30%
probabilities of closing respectively:
Forecasted Sales= 
(10,000×0.5) + (20,000×0.3) = 5000 + 6000 =11 000
Pros:
Provides a dynamic view of the
pipeline.
Allows for adjustment as deals
move through stages.
Pros:
Provides a dynamic view of the
pipeline.
Allows for adjustment as deals
move through stages.
Cons:
Can be subjective if probabilities
are not based on data.
May lead to overestimation if
probabilities are too optimistic.
Cons:
Can be subjective if probabilities
are not based on data.
May lead to overestimation if
probabilities are too optimistic.
Opportunity stage forecasting / 08

Length of Sales Cycle Forecasting
Description:
This method forecasts sales based on the average length of the sales cycle and the status of
current opportunities.
When Applicable:
When you have a consistent and
measurable sales cycle.
Useful in industries where sales
cycles are long and well-defined.
Start
Start
Qualification 
30 days
Proposal 
60 days
Proposal 
60 days
Negotiation 
80 days
Negotiation 
80 days
Close
Close
How It Works:
Calculate the average length of the sales cycle.
Identify where each opportunity is within the cycle.
Estimate the likelihood of closing based on time remaining in the cycle.
Formula
Length of Sales Cycle Forecasting / 09

If a deal worth $50,000 is 30 days into a 90-day sales cycle:
Forecasted sales=50,000 × (30 / 90) =16,666.67
Pros:
Aligns forecast with the reality of
the sales process.
Encourages focus on progressing
deals through the cycle.
Pros:
Aligns forecast with the reality of
the sales process.
Encourages focus on progressing
deals through the cycle.
Cons:
Assumes sales cycles are
uniform, which may not always
be the case.
Can be less accurate in markets
with fluctuating sales cycles.
Cons:
Assumes sales cycles are
uniform, which may not always
be the case.
Can be less accurate in markets
with fluctuating sales cycles.
Length of Sales Cycle Forecasting / 10

Regression Analysis
Description:
Regression analysis predicts sales by analyzing the relationship between sales and one or
more independent variables (e.g., marketing spend, economic indicators).
When applicable:
When you have access to multiple
influencing factors.
Ideal for understanding the impact  
of various factors on sales.
How It Works:
Identify independent variables that influence sales.
Use statistical software to calculate the regression equation.
Predict future sales using the equation.
Formula
For a simple linear
regression:
Forecasted Sales
where ‘a’ is the intercept, ‘b’ is the slope, and
‘X’ is the independent variable.
Regression Analysis / 11

Example Calculation:
If the regression equation is:
Sales = 50 + 2(marketing spend) and marketing spend is $10,000:
Forecasted sales = 50 + 2(10,000) = 20,050
Pros:
Can handle multiple influencing
factors.
Provides insights into the drivers
of sales.
Pros:
Can handle multiple influencing
factors.
Provides insights into the drivers
of sales.
Cons:
Requires statistical knowledge
and software.
Data quality and the selection of
variables are crucial.
Cons:
Requires statistical knowledge
and software.
Data quality and the selection of
variables are crucial.
Regression analysis / 12

Lead-Driven Forecasting
Description:
This method forecasts sales based on the number of leads generated, conversion rates,  
and average deal size.
When Applicable:
When lead generation
is a key driver of sales.
Suitable for businesses
with a strong marketing
component.
Lead
generation
Marketing
qualificated
leads (MQL)
30%
convert
50% convert
40% convert
20%
close
Sales
qualificated
leads (SQL)
Opportunities Closed deals
Closed deals
How It Works:
Track the number of leads generated.
Calculate the average conversion rate and deal size.
Multiply leads by the conversion rate and average deal size.
Formula
Lead-Driven Forecasting / 13

If you have 200 leads, a 5% conversion rate, and an average deal size of $1,000:
Forecasted Sales = 200 × 0.05 × 1,000 = 10,000
Pros:
Directly ties marketing efforts to
sales outcomes.
Easy to understand and
implement.
Pros:
Directly ties marketing efforts to
sales outcomes.
Easy to understand and
implement.
Cons:
Assumes consistent lead quality
and conversion rates.
May not account for external
factors affecting sales.
Cons:
Assumes consistent lead quality
and conversion rates.
factors affecting sales.
Lead-Driven Forecasting / 14

Multivariable Analysis Forecasting
Description:
This advanced method combines various factors like historical data, sales pipeline data,
economic indicators, and more to create a comprehensive forecast.
When applicable:
Large organizations with access to
diverse data sources.
When a holistic view of the sales
environment is required.
Historical  
data
Seasonality
Marketing  
spend
Market 
trends
Sales 
forecast
Sales 
forecast
Sales 
pipeline
Economic 
indicators
Competitive 
analysis
How It Works:
Collect data from multiple sources.
Use advanced statistical models (e.g., multiple regression) to analyze the impact of
different variables.
Combine these insights to generate a forecast.
Formula
A general form could be:
Forecasted Sales
where X1,X2,…,Xnare different independent variables.
Multivariable Analysis Forecasting / 15

If a company analyzes marketing spend, economic growth, and sales pipeline:
Forecasted sales = 100 + 3(Marketing spend) + 5(Economic growth)  
+ 2(Pipeline value)
Pros:
Comprehensive and data-driven.
Captures complex relationships
between variables.
Pros:
Comprehensive and data-driven.
Captures complex relationships
between variables.
Cons:
Requires significant data and
expertise.
Can be complex and time-
consuming to implement.
Cons:
Requires significant data and
expertise.
consuming to implement.
Multivariable Analysis Forecasting / 16

Qualitative Forecasting
Description:
This method uses expert opinions, market research, and qualitative data to predict sales.
When Applicable:
New markets or products with 
little historical data.
When quantitative data is 
unreliable or unavailable.
Scenario 
analysis
Customer 
surveys
Market 
research
Sales 
forecast
Sales 
forecast
Sales 
team input
Expert 
opinions
Delphi 
method
How It Works:
Collect data from multiple sources.
Use advanced statistical models (e.g., multiple regression) to analyze the impact of
different variables.
Combine these insights to generate a forecast.
Formula
No specific formulas; it’s based 
on qualitative assessment.
Qualitative forecasting / 17

If experts predict a market growth of 20% and believe the company's market share will
remain stable, you could estimate a similar growth in sales.
Pros:
Useful in uncertain or rapidly
changing markets.
Can incorporate insights not
captured by data.
Pros:
Useful in uncertain or rapidly
changing markets.
Can incorporate insights not
captured by data.
Cons:
Subjective and prone to bias.
Less reliable without supporting
data.
Cons:
Subjective and prone to bias.
Less reliable without supporting
data.
Qualitative forecasting / 18

AI and Machine Learning Forecasting
Description:
AI and machine learning algorithms analyze large datasets to identify patterns and predict
future sales. These methods leverage the power of computational models to process and
learn from vast amounts of data, enabling more accurate and nuanced forecasts.
When applicable:
When you have access to large datasets with complex relationships.
Suitable for businesses looking to harness cutting-edge technology for more
accurate and adaptive forecasting.
Ideal for industries with rapidly changing conditions or when traditional
methods fall short.
Input data
Historical 
sales
Historical 
sales
Market 
trends
Market 
trends
Customer  
data
Customer  
data
Economic 
indicators
Economic 
indicators
AI processing
Neural networks
Neural networks
Random forests
Random forests
Gradient boosting
Gradient boosting
Forecast output
Time
Sales
How It Works:
Historical data, including sales figures, customer behavior, market conditions, and more,
are fed into machine learning models.
The model trains itself by recognizing patterns, trends, and anomalies in the data.
Once trained, the model can predict future sales based on new data inputs and its learned
patterns.
Models can include time series analysis, decision trees, neural networks, and other
advanced algorithms.
AI and Machine Learning Forecasting / 19

Formula
AI and machine learning models do
not rely on traditional formulas but
rather on algorithmic frameworks
such as:
Linear Regression Models
Random Forests
Neural Networks
Support Vector Machines (SVM)
Gradient Boosting Machines (GBM)
These models generate predictions based on the
relationships identified within the data.
Let’s assume an AI model predicts future sales based on various factors like historical sales
data, economic indicators, customer sentiment, and more. If the model has been trained with
significant past data, it might predict that a company’s sales will increase by 15% next quarter
based on positive economic indicators and strong customer sentiment.
Pros:
Highly accurate when large
amounts of data are available.
Capable of identifying complex
relationships and trends that are
not immediately apparent.
Continuously improves as more
data becomes available.
Reduces human bias in
forecasting.
Pros:
Highly accurate when large
amounts of data are available.
Capable of identifying complex
relationships and trends that are
not immediately apparent.
Continuously improves as more
data becomes available.
Reduces human bias in
forecasting.
Cons:
Requires a large amount of data
and computational power.
Complex to set up and maintain,
requiring expertise in data
science and machine learning.
Can be a "black box," making it
difficult to understand how
predictions are made.
Cons:
Requires a large amount of data
and computational power.
Complex to set up and maintain,
requiring expertise in data
science and machine learning.
Can be a "black box," making it
difficult to understand how
predictions are made.
AI and Machine Learning Forecasting / 20

Consensus Forecasting
Description:
Consensus forecasting involves gathering input from multiple departments or stakeholders to
create a forecast that represents a collective agreement.
When Applicable:
In organizations where
collaboration across departments
is crucial.
When no single method or data
source is sufficient to forecast
sales accurately.
Useful in ensuring buy-in from
various teams, aligning them
toward common goals.
Product Sales
Customer 
Service
Operations
Consensus 
Forecast
Consensus 
Forecast
Marketing
Finance
How It Works:
Representatives from different departments (e.g., sales, marketing, finance) provide their
forecasts based on their unique insights and data.
These forecasts are discussed, and a consensus is reached, often using the Delphi
method or other structured processes.
The final forecast reflects a balanced view, considering different perspectives and
expertise.
Consensus Forecasting / 21

Formula
No specific formulas are used; the process is more about
reaching an agreement based on diverse inputs.
If the sales department predicts $1 million in sales, marketing predicts $1.2 million, and
finance predicts $950,000, a consensus might be reached at around $1.05 million.
Pros:
Encourages collaboration and
cross-functional alignment.
Incorporates diverse insights,
potentially leading to more
accurate forecasts.
Builds organizational
commitment to the forecast.
Pros:
Encourages collaboration and
cross-functional alignment.
Incorporates diverse insights,
potentially leading to more
accurate forecasts.
Builds organizational
commitment to the forecast.
Cons:
Time-consuming due to the need
for meetings and discussions.
May lead to compromises that
dilute the accuracy of the
forecast.
Subject to internal politics and
power dynamics.
Cons:
Time-consuming due to the need
for meetings and discussions.
May lead to compromises that
dilute the accuracy of the
forecast.
Subject to internal politics and
power dynamics.
Consensus Forecasting / 22

Scenario Planning
Description:
Scenario planning involves creating multiple forecasts based on different potential future
scenarios, allowing companies to prepare for various possibilities.
When applicable:
When the market is highly volatile or uncertain.
Useful for strategic planning and risk management.
Ideal for industries susceptible to external factors like economic shifts,
regulatory changes, or technological disruptions.
Current State
Current State
Economic Growth Scenario Status Quo Scenario Economic Downturn Scenario
High Growth 
Forecast: +20%
Moderate Growth 
Forecast: +10%
Slight Growth 
Forecast: +5%
Slight Decline 
Forecast: -5%
Significant Decline
Forecast: -15%
How It Works:
Identify key drivers that could impact future sales (e.g., economic downturn, new
regulations, technological advancements).
Develop several scenarios based on different combinations of these drivers.
Forecast sales for each scenario, considering how each driver would influence sales
outcomes.
Scenario Planning / 23

Formula
Scenario planning is not formulaic but rather strategic.
The approach involves qualitative and quantitative
analysis to estimate outcomes under various scenarios.
A company might create three scenarios:
Best-case:
Economic growth and
strong customer
demand lead to a
increase in sales.
20%
Base-case
Moderate growth
leads to a
increase in sales.
10%
Worst-case:
Economic downturn
and decreased
demand lead to a
decrease in sales.
5%
Pros:
Prepares the organization for
various possible futures,
reducing the element of surprise.
Encourages strategic thinking
and flexibility.
Can be a valuable tool for risk
management.
Pros:
Prepares the organization for
various possible futures,
reducing the element of surprise.
Encourages strategic thinking
and flexibility.
Can be a valuable tool for risk
management.
Cons:
Time-consuming and resource-
intensive.
Can be complex to manage
multiple scenarios.
May lead to over-analysis and
indecision.
Cons:
Time-consuming and resource-
intensive.
Can be complex to manage
multiple scenarios.
May lead to over-analysis and
indecision.
Scenario Planning / 24

Sales Velocity Forecasting
Description:
Sales velocity is a metric that measures how quickly deals move through the sales pipeline
and how much revenue is generated over a specific period. Forecasting based on sales
velocity involves using this metric to estimate future sales.
When Applicable:
In fast-moving sales environments
where the speed of deal closure is
critical.
When you need to closely monitor
the efficiency and effectiveness
of the sales process.
Ideal for companies looking to optimize
sales operations and maximize revenue.
Win Rate 
30%
Number of  
Opportunities 
100
Sales
Velocity
Sales
Velocity
Average 
Deal Size 
$10,000
Sales Cycle 
Length 
60 days
How It Works:
Sales velocity is calculated using four key metrics:
Number of Opportunities (O): The total number of deals in your pipeline.
Average Deal Size (D): The average revenue per closed deal.
Win Rate (W): The percentage of deals that are successfully closed.
Sales Cycle Length (L): The average time it takes to close a deal.
Sales Velocity Forecasting / 25

Formula
The formula for  
Sales Velocity is:
This formula gives you the revenue generated per day. To forecast future
sales, multiply the sales velocity by the number of days in the forecasting
period.
If you have 20 opportunities, an average deal size of $10,000, a win rate of 30%, and a sales
cycle length of 60 days:
Sales velocity = (20 × $10,000 × 0.3) / 60 = 60,000 / 60 = $1,000
per day
For a 30-day month:
Forecasted sales = $1,000 × 30 = $30,000
Pros:
Directly ties sales performance
metrics to revenue forecasting.
Helps identify areas for
improving the sales process.
Provides a dynamic view that
adjusts as metrics change.
Pros:
Directly ties sales performance
metrics to revenue forecasting.
Helps identify areas for
improving the sales process.
Provides a dynamic view that
adjusts as metrics change.
Cons:
Requires accurate and consistent
data collection.
Less effective if sales cycles
vary significantly between deals.
factors that could impact the
pipeline.
Cons:
Requires accurate and consistent
data collection.
Less effective if sales cycles
vary significantly between deals.
factors that could impact the
pipeline.
Sales Velocity Forecasting / 26

Forecasting Based on  
Deals Probabilities
Description:
Bottom-up forecasting involves aggregating individual forecasts from sales reps, who
estimate the probability of closing each deal in their pipeline. This method relies on the reps'
knowledge and insights into their deals.
When Applicable:
In organizations with a large sales team where individual reps have deep insights
into their deals.
When you want to involve the sales team in the forecasting process to
improve accuracy and accountability.
Ideal for companies where deal-specific knowledge is critical for accurate
forecasting.
$150k
$100k
$50k
$0k Deal 1 
90%
Deal 2 
70%
Deal 3 
40%
Deal 4 
50%
Deal 5 
30%
Total 
forecast
$60k
$30k
$20k
$20k
$10k
$140k
How It Works:
Each sales rep assesses the deals in their pipeline and assigns a probability of closing for
each deal.
The forecasted revenue for each deal is calculated by multiplying the deal value by the
assigned probability.
The individual forecasts are then aggregated to form the overall sales forecast.
Forecasting Based on Deals Probabilities / 27

Formula The overall forecast is the sum of all individual forecasted revenues:
If a sales rep has three deals in their pipeline:
Deal 1: $50,000 with a 70% probability of closing.
The forecasted revenue would be:
Deal 1: = $50,000 × 0.7 = $35,000
Deal 2: = $30,000 × 0.4 = $12,000
Deal 3: = $20,000 × 0.9 = $18,000

Total forecast = $35,000 + $12,000 + $18,000 = $65,000
Pros:
Utilizes the firsthand knowledge
of sales reps, potentially leading
to more accurate forecasts.
Involves the sales team in the
forecasting process, increasing
buy-in and accountability.
Flexible, allowing adjustments as
deals progress or as more
information becomes available.
Pros:
Utilizes the firsthand knowledge
of sales reps, potentially leading
to more accurate forecasts.
Involves the sales team in the
forecasting process, increasing
buy-in and accountability.
Flexible, allowing adjustments as
deals progress or as more
information becomes available.
Cons:
Heavily dependent on the
accuracy and honesty of sales
reps.
Subject to human bias, such as
optimism or pessimism.
May require significant time and
effort to collect and aggregate
individual forecasts.
Cons:
Heavily dependent on the
accuracy and honesty of sales
reps.
Subject to human bias, such as
optimism or pessimism.
May require significant time and
effort to collect and aggregate
individual forecasts.
Forecasting Based on Deals Probabilities / 28

Time-Series Analysis for Forecasting
Description:
Time-series analysis is a statistical technique used to analyze and forecast data points
collected or recorded at specific time intervals. It involves understanding the underlying
patterns within the data, such as trends, seasonality, and cyclical behaviors, and using these
insights to predict future values.
When Applicable:
When you have continuous, sequential data points over time (e.g., monthly or
quarterly sales).
Ideal for businesses where historical data exhibits trends, cycles, or
seasonal patterns.
Useful for long-term planning and understanding the impact of external
factors on sales over time.
200
150
Historical data
Trend
Forecast with confidence interval
Seasonalpattern
100
0
2020 2021 2022 2023 2024 2025
How It Works:
Time-series analysis typically involves the following steps:
Decomposition: Break down the time series into its components:
Trend: The long-term movement or direction in the data.
Seasonality: Regular, repeating patterns or cycles in the data (e.g., sales spikes during
holidays).
Residual: The remaining component after removing trend and seasonality, which may
include irregular or random fluctuations.
Time-Series Analysis for Forecasting / 29

Modeling: Depending on the data characteristics, different models can be applied:
Moving Averages: Smoothing the data to highlight trends by averaging over a specific
period.
Exponential Smoothing: Similar to moving averages but gives more weight to recent
observations.
ARIMA: As discussed earlier, this is a specific type of time-series model that combines
autoregression, integration, and moving average.
Forecasting: Use the identified patterns and the chosen model to project future data
points.
Formula
Simple Moving
Average:
Formula
Exponential
Smoothing:
Formula
ARIMA:

Suppose you have quarterly sales data over the last three years and you want to forecast
sales for the next four quarters. Using a time-series analysis approach, you might start by:
Decomposing the time series to identify the trend, seasonal, and residual components.
Applying a model such as exponential smoothing or ARIMA to forecast the trend.
Combining the seasonal patterns with the trend forecast to predict the next year’s sales.
For instance, using a simple
exponential smoothing method
with an α of 0.3:
If the last smoothed value St−1 was $100,000 and the most recent actual sales Yt was
$110,000:
St = 0.3 × $110,000 + 0.7 × $100,000 = $103,000
This St can then be used to forecast the next period’s sales.
Pros:
Can handle a variety of data
patterns including trends,
seasonality, and cycles.
Well-suited for long-term
forecasting when historical
patterns are likely to continue.
Time-series models can be
highly accurate with sufficient
and clean data.
Pros:
Can handle a variety of data
patterns including trends,
seasonality, and cycles.
Well-suited for long-term
forecasting when historical
patterns are likely to continue.
Time-series models can be
highly accurate with sufficient
and clean data.
Cons:
Requires a good understanding
of statistical methods and time-
series data characteristics.
Sensitive to outliers and missing
data, which can distort the
model.
consuming to develop and
validate the model.
Cons:
Requires a good understanding
of statistical methods and time-
series data characteristics.
Sensitive to outliers and missing
data, which can distort the
model.
consuming to develop and
validate the model.

Sales forecasting is a crucial element in any B2B organization’s strategy. It’s not just about
predicting numbers; it’s about preparing your business for the future, ensuring efficient
resource allocation, and building trust with investors, stakeholders, and your own team. As
we’ve explored, there are numerous methods available, each with its strengths, weaknesses,
and ideal applications.
From the straightforward approach of historical data analysis to the sophisticated techniques
of AI and machine learning, the choice of forecasting method should align with your
company’s needs, market conditions, and the available data. There is no one-size-fits-all
solution; instead, the most effective forecasting often involves a combination of methods
tailored to the unique challenges and opportunities your business faces.
In today’s fast-paced and ever-changing market, accurate sales forecasting is more critical
than ever. It enables you to stay ahead of the curve, make informed decisions, and ultimately,
drive your business forward with confidence. Whether you’re a startup building your first
sales pipeline or an established company looking to refine your forecasting process, the
methods discussed here offer a comprehensive toolkit for navigating the complexities of
sales prediction.
Remember, the goal of sales forecasting isn’t just to predict
the future—it’s to shape it. By choosing the right methods
and continually refining your approach, you can ensure that
your sales forecasts are not only accurate but also
actionable, paving the way for sustained growth and success.
Afterword / 33

+ = Sales efficiency❤️
Ready to

with Forecastio today!

supercharge your sales efficiency?

Book a demo
Don't just forecast.Transform your sales operations with real-time,
AI-powered insights.
Discover how our highly accurate sales forecasting can revolutionize
your pipeline management.

alex.zlotko@forecastio.ai
Alex Zlotko
CEO & Co-founder at Forecastio

Forecastio guide - Sales Forecasting Methods.pdf

More Related Content

Similar to Forecastio guide - Sales Forecasting Methods.pdf (20)

Recently uploaded (20)

Forecastio guide - Sales Forecasting Methods.pdf