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

# Forecast Review

> Before the forecast call: which deals changed category, and whether the deals in commit are actually on track. Pressure-tested against evidence, deal by deal.

The forecast call asks reps to defend a number they submitted three days ago. The defence is a narrative, the narrative is unfalsifiable in the room, and the meeting ends with the same number and slightly more confidence in it. Forecast review does the checking beforehand. Which deals moved category and why, and for everything sitting in commit, whether the record actually supports it being there.

## Data Sources

| Category         | Required | Recommended |
| :--------------- | :------- | :---------- |
| CRM              | ✓        |             |
| Calls            | ✓        |             |
| Email + Calendar |          | ✓           |
| Messaging        |          | ✓           |

**Starter** (required only): forecast category changes since the last call, and what the calls say about each deal in commit.

**Full** (required + recommended): the same, plus engagement between calls. Commit deals where the customer has not replied in two weeks are the ones that miss, and that signal only exists in email.

See the [Context Graph](/docs/unifying-gtm-data/customer-context-graph) for how each source feeds the review.

### Choose your path

| Path                       | Best for                                                            | Output is delivered                   |
| :------------------------- | :------------------------------------------------------------------ | :------------------------------------ |
| Build in Clearskies        | Teams with a standing forecast call                                 | To Slack DMs or channels              |
| Build in Claude or ChatGPT | Leaders pressure-testing the forecast deal by deal, in conversation | In the Claude or ChatGPT conversation |

## Build in Clearskies

Build the workflow once and the forecast is checked before the call rather than during it.

### The Workflow

In the Clearskies app, open the workflow builder and describe what you want in plain language:

```text theme={null}
Every Wednesday at 7am, find the open deals closing this quarter. Report
which ones changed forecast category since last week and what caused each
change. Then for every deal in commit, check whether the record supports
it: a confirmed next step, an engaged economic buyer, recent customer
activity, and a close date that has not moved. Post the review to the
#forecast channel.
```

The app configures the schedule, the deals in scope, and the destination.

Scheduled workflows have no meeting context, so the deals come from a record lookup rather than a calendar. Forecast category is a CRM field, so what the workflow can check depends on what your CRM tracks. If category lives in a custom field, point the lookup at that field by name.

Run it a day before the forecast call. The point is that reps see the same read leadership sees and can answer it before the meeting, rather than being surprised in it.

### Deploy to Slack

Once published, the channel gets the read before the call:

```text theme={null}
Forecast Review — week of Aug 12
Commit: $840K across 9 deals · Best case: $1.4M · Pipeline: $2.1M
Commit change vs last week: +$84K, +1 deal

Moved into commit (2)
• Acme Corp, $84K. Renewal terms settled on the Aug 4 call. Supported.
• Vantage, $210K. Moved by the rep on Aug 8. Nothing on the record
  changed that week. No call since Jul 22.

Moved out of commit (1)
• Kestrel, $60K → best case. Champion left the company Aug 4. Correct
  move, made same-day.

Commit deals the record does not support (3)
• Vantage, $210K. No customer contact in 21 days. Close date has moved
  twice. No confirmed next step. Largest deal in commit.
• Ridge Analytics, $95K. Four outbound touches since Jul 18, no reply.
  Last recorded next step was "waiting on legal," 19 days ago.
• Northwind, $120K. Third close-date slip. Each slip has had a different
  internal approval as the stated cause, which suggests the approval path
  was never actually mapped.

Commit deals the record does support (6)
$355K. Recent two-way activity, confirmed next steps on the calendar,
economic buyer engaged in the last 10 days.

The question for the call
$425K of the $840K commit sits in three deals with no recent customer
engagement. Either something is happening off the record, or commit is
overstated by roughly half.
```

### Customize the rollout

* Set the cadence to land ahead of your forecast call
* Choose the scope: whole company, one segment, one team, one rep's book
* Define what "supported" means for you: your own criteria for a commit deal, checked against evidence
* Route it to a leadership channel, or to each manager's DM with only their team
* Chain it with [pipeline review](/docs/use-cases/pipeline-review) so the movement read and the forecast read arrive together

## Build in Claude or ChatGPT

The stronger path for the deal-by-deal version, because pressure-testing a forecast is a conversation. You read the answer, push on one deal, and ask again.

### The Prompt

Open Claude or ChatGPT with the Clearskies MCP connected and paste this (or modify to your needs).

```text theme={null}
Using the Clearskies Context Graph, pressure-test my forecast for this
quarter.

First: which deals changed forecast category since last week, and what
actually changed on the record to justify each move.

Then, for every deal in commit, tell me whether the evidence supports it:
- When we last heard from the customer, and who
- Whether there is a confirmed next step with a date
- Whether the economic buyer has engaged recently
- Whether the close date has moved, and how many times

Rank the commit deals from best supported to least. Be blunt about the
weak ones. If the record is silent on something, say it is silent rather
than assuming it is fine.
```

Then push on the weakest one: `Walk me through Vantage. What would have to be true for that close date to hold?`

### Make it yours

* Add `Only [rep]'s deals` before a manager's forecast conversation.
* Add `Compare the commit deals to the deals we lost last quarter after they reached commit` to check the pattern against your own history.
* Add `Which best-case deals have stronger evidence than my weakest commit deals` to find the category assignments that are backwards.

### The Skill (Claude)

Two skills in the Clearskies plugin for Claude do the deal-level work behind a forecast read.

`clearskies:deal-review` audits a single deal against its risk signals: champion engagement, economic buyer status, unanswered communications, close-date health, deal velocity. Run it on any commit deal you doubt.

`clearskies:deal-strategy` goes further on a deal that is stuck, including whether to push the date or qualify out, which is often the real decision behind a forecast conversation.

<Tip>
  Want this packaged for your team? [**Reach out**](mailto:support@clearskies.cc) and we'll help you customize the skill to your workflow.
</Tip>

## The Plugin

**Ready to make this part of your team's workflow?** We'll set up the Clearskies plugin with you. Clearskies workflows, Claude skills, all tailored to how your team works.

[**Book 15 minutes →**](https://calendly.com/pouyan/new-meeting)

## Next steps

1. [Sign in to Clearskies](https://app.clearskies.cc/)
2. Connect your data sources
3. Get your [Clearskies MCP server](/docs/building-with-clearskies/mcp-server) and try with [Claude](/docs/building-with-clearskies/mcp-clients/claude) or [ChatGPT](/docs/building-with-clearskies/mcp-clients/chatgpt)