Sales Forecasting

Sales forecasting estimates the revenue likely to close during a future period. A forecast is useful when its assumptions are visible: which opportunities are included, why they are expected to close, and what evidence supports their timing.

This guide explains three practical forecasting methods and the CRM data needed to support them.

On this page

• What is a sales forecast?
• Three forecasting methods
• Data required for forecasting
• Build a forecast step by step
• Review forecast accuracy
• Common forecasting errors
• Frequently asked questions

What is a sales forecast?

A sales forecast estimates expected closed revenue for a defined period. It differs from total pipeline, which includes every open opportunity regardless of probability or timing.

A forecast should answer:

• What amount is expected to close?
• During which period?
• Which opportunities contribute to that amount?
• What assumptions could change the result?

Keep forecast categories separate from pipeline stages. A late-stage deal can still be excluded from the current-month forecast if the buyer's decision is scheduled for next quarter.

Three forecasting methods

Stage-weighted forecast

Multiply each opportunity's value by the historical close probability for its current stage.

$$ \text{Weighted pipeline} = \sum (\text{Opportunity value} \times \text{Stage probability}) $$

If a €10,000 opportunity is in a stage that historically closes 40% of the time, its weighted value is €4,000.

This method is simple and consistent. It becomes misleading when stage definitions are loose or opportunities remain in old stages.

Opportunity forecast categories

Ask the owner to classify each opportunity for the period:

Manager judgment should challenge the evidence behind the category. A commit with no decision date or approved budget needs review.

Historical conversion forecast

Estimate future outcomes from past cohorts, sources, or stages. For example, if qualified opportunities from a channel historically close at a known rate and value, apply those rates to the current comparable group.

Use enough history to reduce the effect of one unusual month. Split materially different motions, such as referrals and cold outbound, instead of averaging them together.

Data required for forecasting

Forecast quality depends on a small number of maintained fields:

• Opportunity value
• Current pipeline stage
• Expected close date
• Owner
• Next action and date
• Product or service
• Source or sales motion
• Final won or lost outcome
• Loss reason

The value should reflect a plausible commercial scope. The close date should reflect a known decision event rather than the final day of the quarter chosen by default.

Use the sales pipeline stages guide to define stages with observable evidence.

Build a forecast step by step

1. Define the forecast period

Choose a month, quarter, or another operating period. Keep the same boundary across reports.

2. Clean the eligible pipeline

Remove duplicate opportunities, close dead deals, update values, and move opportunities to stages supported by current evidence.

3. Check timing

Review each expected close date against the buyer's next decision event. Move dates when the evidence changes instead of allowing overdue dates to remain.

4. Apply the method

Calculate stage-weighted value, owner categories, or both. Showing both can expose disagreement between historical probability and current deal judgment.

5. Record assumptions

For large opportunities, record dependencies such as legal review, budget approval, stakeholder meeting, or procurement completion.

6. Publish a range

A range can be more honest than one precise number. For example, show committed revenue, best case, and total eligible pipeline separately.

Review forecast accuracy

Compare the forecast made at a fixed point with the actual result for that period.

Measure:

$$ \text{Forecast error} = \frac{|\text{Forecast} - \text{Actual}|}{\text{Actual}} \times 100 $$

Also inspect why the result differed:

• Opportunities slipped into a later period
• Values increased or decreased
• Expected deals were lost
• Unforecast opportunities closed
• Stage probabilities did not match observed conversion

Accuracy should improve the process, not punish owners for uncertainty. If close dates slip repeatedly, investigate decision criteria and stage hygiene.

Common forecasting errors

Treating total pipeline as forecast. Early opportunities do not have the same likelihood as signed agreements.

Using stale close dates. An overdue date creates revenue in the wrong period until somebody changes it.

Changing definitions each month. Trends become meaningless when “commit” or “qualified” means something different every review.

Ignoring sales-cycle length. A new opportunity with a typical 90-day cycle rarely belongs in next week's forecast.

Counting recurring value inconsistently. Decide whether opportunity value represents first invoice, annual contract value, total contract value, or another measure.

Frequently asked questions

What is the difference between pipeline and forecast?

Pipeline is the full set of open opportunities. Forecast is the portion expected to close in a defined period after probability, timing, and evidence are considered.

How often should a sales forecast be updated?

Update it whenever material evidence changes and review it on a fixed weekly cadence. Avoid changing values merely to match a target.

What probability should each stage have?

Start with a documented estimate, then replace it with observed win rates for opportunities that reached each stage. Recalculate when the process or market changes.

Should a forecast include renewals?

Include renewals when they are part of the revenue target, but separate them from new business if their conversion rates and process differ.

How can CRM data improve forecasting?

Consistent stages, current values, realistic dates, named owners, and dated next actions make the assumptions visible. Begin with the lead management process if those fields are incomplete.