Sales forecast: pipeline coverage, probability, and weighted amount

A sales forecast is the revenue you expect to sign in a period, read from the open pipeline. It is not the sum of every quote you sent. It is a filtered reading: stage, probability, close date, and the deals you are willing to remove.

An honest forecast is rebuilt every week. A forecast filled in for the month-end meeting describes the team’s memory more than the pipe.

Three numbers, not one

1. Open pipeline. Sum of amounts on opportunities that are neither won nor lost, with a date inside the period.
2. Weighted amount. For each deal: amount × probability. A 20,000 opportunity at 25 percent counts as 5,000.
3. Coverage. Open pipeline ÷ target. If the quarter target is 100,000 and the open pipe is 280,000, coverage is 2.8×.

Coverage only means something next to the historical win rate. At 30 percent of deals won, 3× target is a common order of magnitude. At 15 percent you need more. At 50 percent, 3× is already comfortable. Win-rate math is in conversion and velocity.

Probability by stage, not by feel

Set a starting probability per column, then let the seller adjust it with a reason:

Stage	Starting probability (example)
Qualification	10%
Proposal	25%
Negotiation	50%
Verbal / purchase order in progress	80%

These figures are a starting point, not a standard. Replace them with your real stage conversion as soon as you have history. A probability with no stage is an opinion.

Commit, upside, slippage

• Commit: deals whose next step is a dated signature, with the right stakeholder. You can still miss them, but you defend them one by one in the pipeline review.
• Upside: live deals with an uncertain date. They feed the pipe. They do not feed the promise.
• Slippage: a deal that leaves the month without being lost. Move the date. Do not leave it “in closing” to protect the chart.

Opportunities with no date or no amount belong in none of the three. They go back to qualification or they leave.

What distorts the forecast

• Lost deals nobody will close out.
• Amounts copied from the first quote and never revised.
• A single “50 percent” on the whole pipe.
• A CRM updated outside the email flow, so it lags the conversation.

The glossary aligns the words (weighted, coverage, commit). The board makes gaps obvious: a negotiation card with no close date is not a commit.

On Pro, Colossales keeps amount, date, and probability on the Gmail thread. The forecast updates when you reply, not on Friday night.

Bottom line

Announce a commit you can walk card by card. Keep coverage as a volume warning, and weighted pipeline as a statistical reading. Both lie if the pipeline contains deals with no next action.