> ## Documentation Index
> Fetch the complete documentation index at: /docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Check forecast completeness

> Find what the last three years of books had that next year's plan does not, and decide each one: forecast it, say what covers it, or set it aside.

**Forecast completeness** holds the plan against your history. Each category in each program
that had money in the last three years becomes an expectation for the year ahead, and each one
ends in a decision. It is the quickest way to find a grant or a bill nobody forecast.

Everyone can read it; owners, finance and program managers decide.

**The result:** every expectation from the last three years is forecast, covered by another forecast, or set aside with a reason.

## How it judges

Expectations are held by the year, not the month. Grant payments move and programs change, so
a forecast need not land in the same month as last year to count.

A category counts as expected when it had money last year, or in two of the last three. It is
decided when the year's forecasts reach at least half of what is expected. Under half reads
**Forecast at $X, under half of $Y** and stays open, since a plan that small is more often
missing part of the year than planning a smaller one.

The page leads with how much is decided: for example, *82% of the money the last three years
had is decided: $1,240,000 of $1,512,000. 9 expectations are still open, \$272,000 in all.*

## Decide each one

Open expectations are under **Open**. Each card shows **Expected**, **Planned this year** and
**Last three years**, monthly bars of last year against the plan, and the payees behind it.

| Choose | When | What happens |
| - | - | - |
| **Use suggestion** | The **From history** suggestion looks right | Forecasts are made in one step. A payee paid regularly for half a year or more becomes a repeating forecast; anything else repeats last year's dates |
| **Forecast…** | You want to adjust it first | The forecast form opens, filled in from history |
| **Covered by…** | Another forecast already plans this money | Choose **The forecast that covers it**, then **Mark covered** |
| **Not this year…** | It will not happen this year | Pick a reason (**A one-off**, **The grant or contract ended**, **Forecast under another category or program**, **Another reason**), then **Set aside for this year** |

**Use every suggestion** does the first for every open card at once. Read the suggestions
before you use it.

A decision you change your mind about can be reopened from **Decided** with **Reopen**. A
"covered" decision lapses if its forecast is deleted; a "not this year" decision lapses if the
money comes back.

## Inside a locked budget

If the year is in a [locked cycle](/docs/guides/budget-cycles.md), forecasts made from history are
added to your change request instead, and the expectation reads
**Proposed in a change request** until it is decided.

## Ask your assistant

> Check forecast completeness for next year for Adult Literacy and walk me through what is
> open, largest first.

> Use the suggestions for the open expectations under \$5,000, and list the rest for me to
> decide.

Home shows the completeness figure for the year ahead, so owners and finance can follow it
while programs plan.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.