Sign in

Blog · Forecast and pipeline · Finance and FP&A teams

How to read a forecast bridge: the lines that sum, and the deal behind each one

A reading guide for a forecast bridge from last period's number to this one: the check that the lines sum before anything is read, the order to read them in, closed, slipped, lost, new, resized, the deal list behind each line as the thing to open, the line with no deals behind it as the one to distrust, the repeat slippers across bridges, and the two lines that set next period's calibration.

The short answerCheck first that the bridge sums: last period's forecast plus the lines equals this period's. Then read the lines in order: closed won, slipped out, lost, new in, resized. Each line has a deal list behind it, and the line is read by opening the list, not by its total. A line with a total and no deals behind it, a haircut or an adjustment, is the one to distrust. Across several bridges, the deals that appear on the slipped line repeatedly are the calibration finding, and the slipped and lost lines together set how much to discount next period's roll-up.

A forecast bridge is a table of five or six lines between last month's number and this one, and its value is entirely in what is behind each line. This guide is the reading order: the sum, the lines, the deal list behind each, the line to distrust, the repeat slippers, and the calibration.

The check before reading

last period's forecast + Σ lines = this period's forecast every deal that changed is on exactly one line

If either fails, stop; the bridge is broken.

The lines, in reading order

Line Deals behind it Read for
Closed won Deals that closed since last period What was right
Slipped out Deals whose close date moved past the period Who pushed, and how many times
Lost Deals closed lost Where and why
New in Deals created or pulled into the period Whether they are real at this stage
Resized Deals whose value changed Up or down, and by whom
Adjustment None Distrust; require an explanation

A worked bridge

Line Value Deals
Last month's forecast $4.6m
Closed won −$1.2m to actual; +$0.0m to forecast 9 deals closed, in the forecast at $1.3m: $0.1m under
Slipped out −$0.9m 7 deals; 4 of them slipped for the second time
Lost −$0.6m 4 deals; 3 at the same competitor
New in +$0.4m 5 deals, all proposal stage
Resized +$0.2m 2 deals up, 1 down
Adjustment −$0.3m "Management haircut"
This month's forecast $3.4m

Sums: 4.6 − 0.0 − 0.9 − 0.6 + 0.4 + 0.2 − 0.3 = 3.4. Read.

Slipped: seven deals, four for the second time. Open the list; those four are the repeat slippers.

Lost: three of four to one competitor. Open the list; that is a pattern.

New in: five proposal-stage deals pulled into a period with eight weeks left. Open the list; at the team's proposal conversion, they are worth a third of face.

Adjustment: three hundred thousand with no deals. Ask what it is.

Across bridges

Deal Bridges it appeared on the slipped line
O-8821 3
O-8834 3
O-8902 2

Two deals have slipped on three consecutive bridges. They are in the forecast at proposal weight and the team's own history says they close at a twentieth of that.

The calibration

Over four periods Share of starting forecast
Slipped out 18%
Lost 12%
Together 30%

Next period's roll-up, at this point, carries a thirty percent discount, derived from the bridges, with the deals behind it.

Where it goes wrong

Sum unchecked. A deal on two lines; the reading is on a broken table.

Lines read by total. Seven hundred thousand slipped, and no names.

Adjustment accepted. A number someone wanted, uncheckable.

Bridges not compared. The repeat slippers invisible.

Every period, the sum then the lines

Covirage builds the bridge with every deal on one line, the deal lists behind each, and the adjustment line flagged. The forecast bridge guide covers how the bridge is built, and the slip count guide covers the repeat slippers.

Questions people ask

Why check the sum first?

Because a bridge that does not sum has a line missing or a deal on two lines, and every reading after that is on a broken table. The identity is that last forecast plus the lines equals this forecast, exactly, and every deal is on one line.

Why is the line with no deals the one to distrust?

Because a bridge is an attribution: every dollar of movement to a deal. A line that says 'management adjustment, minus $300,000' has attributed nothing; it is a number someone wanted. It may be right, and it cannot be checked, and it should be rare and explained.

What are the two calibration lines?

Slipped and lost, together. Their sum over several periods, as a share of what was in the forecast at the start, is how much the roll-up overstates at this point in the period. That share is the discount next period's roll-up carries, and it is derived, not chosen.