Recommended Free Tools
A sales forecast estimates the sales or revenue a business expects over a defined period. The right method depends on what you sell, how much reliable history you have, the quality of your current pipeline, and how far ahead you need to look. Use the simplest method that fits those conditions, state its assumptions, and compare estimates with actual results.
What a sales forecast is—and what it is not
A sales forecast is an estimate of sales or revenue for a specified period, such as units sold or booked revenue in the coming quarter. It may draw on deal amounts, the likelihood that deals will close, and expected close dates, as well as historical sales and other relevant information. Salesforce’s revenue forecasting guide and sales forecasting guide describe these kinds of inputs.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Sales Forecasting for Pharmaceuticals | $1,021.56 | Buy on Amazon |
| 2 |
|
Sales Forecasting: A Practical Guide | $39.95 | Buy on Amazon |
| 3 |
|
Excel Sales Forecasting For Dummies | $24.11 | Buy on Amazon |
| 4 |
|
Forecasting: Principles and Practice | $57.80 | Buy on Amazon |
| 5 |
|
The Kids' Book of Weather Forecasting (Kids Can!) | $10.31 | Buy on Amazon |
- A forecast estimates an outcome from available information and assumptions.
- A target or quota is a goal. It does not become a forecast just because a team is expected to meet it.
- A scenario explores what might happen under a stated set of assumptions; it need not be the most likely estimate.
Define the measure before calculating it. Bookings, recognized revenue, units sold, and cash received are different measures. If you use one as a proxy for another, say so and explain the timing difference. A forecast is conditional, not a promise, and no single method guarantees a particular level of accuracy.
Choose a method that fits the question and the data
These methods are complementary rather than mutually exclusive. For example, a team can compare an opportunity-based estimate with a historical baseline before deciding whether its pipeline view looks unusually optimistic. Salesforce’s overview of sales forecasting methods and guide to forecasting models describe common approaches and their uses.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →#1 Best Overall
| Method | Useful when | Main inputs | Important limitation |
|---|---|---|---|
| Pipeline or opportunity-stage | Sales are made through identifiable open deals and the near-term pipeline is useful. | Deal value, close date, and an estimated probability or forecast category. | Stale dates and probabilities that are not calibrated to the company’s own outcomes can distort the result. |
| Historical trend or time series | There is a reasonably consistent record of sales over time. | Past sales or bookings by period; trends and recurring seasonal patterns where present. | Past patterns may not persist, and a trend alone does not explain why sales changed. |
| Regression or driver-based | The business has relevant observations on sales and potential explanatory variables. | A sales outcome and variables such as advertising spend or price. | A relationship in historical data does not by itself establish causation; changes and confounding factors matter. |
| Qualitative judgment | History is limited or a product, market, or sales motion is changing. | Expert judgment, market research, or structured expert input. | Judgment remains uncertain even when several people agree. |
How the main forecasting methods work
Pipeline and opportunity-stage forecasting
For a simple weighted pipeline, multiply each eligible deal’s amount by its estimated probability of closing in the forecast period, then add the weighted amounts. For example, a hypothetical $10,000 opportunity with a 40% estimated chance of closing contributes $4,000 to that period’s weighted estimate. This is an expected amount, not a prediction that the deal will close for exactly $4,000.
Probabilities should reflect the organization’s own experience with comparable opportunities and stages where possible. Include only deals whose expected timing fits the period, and keep close dates current. Salesforce identifies pipeline as a useful starting point alongside historical and qualitative inputs in its revenue forecasting guide. For a sales team, compare the pipeline estimate with a historical baseline and inspect large differences rather than automatically choosing the more optimistic number.
Historical trend, seasonality, and moving averages
Historical forecasting starts with recorded sales or bookings by period. Look for the underlying level, direction of change, and recurring seasonal swings before extrapolating. A retailer, for example, can examine several years of sales for recurring seasonal peaks, but should not assume the next season will repeat exactly.
A moving average smooths short-term variation by averaging a chosen number of recent periods. The calculation is:
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Moving average = total sales across the selected periods ÷ number of periods
Suppose a business recorded $80,000, $100,000, and $90,000 in its last three months. The three-month moving average is $90,000 per month: ($80,000 + $100,000 + $90,000) ÷ 3. It is a simple baseline, not a complete forecast: it can lag when sales are changing quickly and does not identify the cause of a change. The University of Kansas Open Textbook Library chapter on small-business management explains the moving-average approach and other forecasting methods.
Rank #3
Regression and explanatory drivers
Regression estimates how a sales outcome relates to one or more variables, such as advertising spend or price. It can help address a question like, “How have monthly revenue and advertising spending moved together in this data?” rather than simply projecting past sales forward. Salesforce discusses this use in its methods guide.
Use relevant observations and treat the result as dependent on the data and assumptions. A correlation does not prove that changing an input will cause sales to change by the model’s estimated amount. Pricing, market conditions, product availability, and other factors may shift at the same time.
Qualitative judgment and Delphi
Expert input and market research can be useful when a business is launching a product, entering a market, or lacks a useful sales history. One structured approach is Delphi: selected experts give input in multiple survey rounds with the aim of reaching consensus. Consensus can organize assumptions, but it does not turn judgment into objective evidence. Record the reasoning, uncertainty, and conditions behind the estimate. Salesforce describes Delphi and other approaches in its forecasting-model guide.
Rank #4
Top-down and bottom-up are starting points, not methods with a universal winner
Top-down and bottom-up describe how an estimate is built or allocated. They can be used alongside pipeline, historical, or driver-based calculations.
| Approach | Starts with | Common use | What to check |
|---|---|---|---|
| Top-down | A high-level market estimate or organizational target, allocated to teams, regions, or products. | Planning for a new market or product and strategic allocation. | Whether the allocation reflects local conditions and realistic capacity. |
| Bottom-up | Rep-level opportunities or local estimates, rolled up to the team or company. | Planning that depends on field inputs, such as hiring or inventory decisions. | Whether inputs are consistent, current, and free of incentives to overstate or understate results. |
Compare the two views and investigate the gap. A top-down estimate may rely on broad assumptions; a bottom-up rollup may inherit weak deal data or uneven local judgments. Salesforce’s sales forecasting guide discusses these different starting points.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.A practical process for building and improving a forecast
- Define the measure and period. Specify whether you are estimating booked revenue, recognized revenue, units, or another outcome, and name the time window.
- Segment where sales behave differently. Separate products, territories, channels, customer types, or sales motions when combining them would conceal meaningfully different patterns.
- Gather actual history and current pipeline. Record where each input came from, when it was updated, and what assumptions it uses.
- Select a transparent baseline. Match the method to the available data and forecast unit. Start with a calculation people can inspect and explain.
- Add judgment or driver variables only with a reason. Document why expert context or a variable belongs in the estimate and what uncertainty remains.
- Compare forecasts with actual results. Review misses by period and segment, identify whether timing, deal values, assumptions, or business conditions changed, and adjust the process where warranted.
- Refresh inputs and method when the business changes. A new market, sales process, product mix, or pricing model can make an old baseline less useful.
These are operating principles, not a guarantee of accuracy. The U.S. Small Business Administration’s guidance on realistic forecasts for new businesses is relevant when historical operating data is sparse.
Best Value
Spreadsheets, CRM systems, and forecast quality
A spreadsheet can be enough for a small operation with a modest number of inputs and a straightforward calculation. It is also useful for showing how a moving average or simple scenario works. As a team grows, a CRM can centralize opportunity details and help share forecast categories, rollups, and management views. Microsoft documents these functions for Dynamics 365 Sales in its sales forecasting overview.
Software organizes information; it cannot make an unreliable probability, stale close date, or biased estimate dependable. Choose tools based on the workflow, the information the team needs to maintain, and how estimates must be reviewed—not on a promise of accuracy.
Quick Recap
Common forecasting mistakes to avoid
- Treating a quota as the expected outcome: goals motivate performance; forecasts estimate what available evidence suggests.
- Mixing measures: do not add cash receipts, bookings, and recognized revenue as though their timing and meaning were identical.
- Using generic stage probabilities: calibrate probabilities to comparable outcomes in the business, and revisit them as the process changes.
- Ignoring timing: a plausible deal amount assigned to the wrong period still produces a misleading period forecast.
- Extrapolating without context: seasonality and trends can inform an estimate, but market or operating changes may make the past a poor guide.
- Over-interpreting a model: a regression relationship is not proof that a driver caused a sales outcome.
- Assuming a tool fixes bad inputs: shared dashboards can make assumptions visible, but input quality remains a management responsibility.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
