> ## 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. # Pipeline Review > Week over week: what progressed, what stalled, what closed, won or lost, and why. The Monday review prepared before anyone walks into it. Monday pipeline reviews run on a spreadsheet exported Friday and a set of verbal updates. The meeting spends forty minutes establishing what happened and ten deciding what to do, which is the wrong ratio. Pipeline review inverts it. What progressed, what stalled, what closed and why, prepared from the record before the meeting starts, so the hour goes to the deals that need a decision. ## Data Sources | Category | Required | Recommended | | :--------------- | :------- | :---------- | | CRM | ✓ | | | Calls | ✓ | | | Email + Calendar | | ✓ | | Messaging | | ✓ | **Starter** (required only): week-over-week movement across the pipeline, stage and close-date changes, and what the calls say about the deals that moved. **Full** (required + recommended): the same, plus what happened between calls. A deal can look unchanged in CRM while the actual signal, a champion who stopped replying, is sitting in an email thread. 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 review on a fixed day | To Slack DMs or channels | | Build in Claude or ChatGPT | Leaders who want to read into the pipeline off-cadence, or RevOps shaping the format first | In the Claude or ChatGPT conversation | Pipeline review is the week-over-week movement read on a cadence. For deals silently going wrong between reviews, see [Pipeline Risk Detection](/docs/pipeline-risk-detection), which fires without being asked. ## Build in Clearskies For teams with a standing review. Build the workflow once and the read is in the channel before the meeting, every week, without anyone assembling it. ### The Workflow In the Clearskies app, open the workflow builder and describe what you want in plain language: ```text theme={null} Every Monday at 7am, find the open deals closing this quarter. Compare this week against last week: what progressed a stage, what slipped a close date, what stalled with no activity, and what closed won or lost. For each material change, say what caused it based on calls and email. Post the review to the #pipeline 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. Set the schedule far enough ahead of the meeting that people read it. An hour before is too late; the night before or first thing that morning gets read. Record lookups cap at 100 records per step, so a large pipeline needs the scope narrowed rather than widened. In practice a review of the deals that can still close this quarter is more useful than a review of everything. ### Deploy to Slack Once published, the channel gets the read before the meeting: ```text theme={null} Pipeline Review — week of Aug 10 Open deals closing this quarter: 34 · $2.1M Change vs last week: +$96K Progressed (4) • Acme Corp — Negotiation → Verbal, $84K. Renewal terms settled on Tuesday's call; expansion into the analytics module still open. • Brightline — Discovery → Evaluation, $45K. Security review cleared. • Two others, both single-stage moves with no notable cause on the record. Slipped (3) • Northwind — close date Aug 31 → Sep 30, $120K. Third slip. Cause on the record each time has been a different internal approval step. • Kestrel — Aug 15 → Sep 15, $60K. Champion left the company Aug 4. • Vantage — Aug 31 → Oct 15, $210K. No cause on the record. Stalled (5) No activity in 21+ days, $340K total. Two have no next step recorded at all. Ridge Analytics has had no customer reply since Jul 18 despite four outbound touches. Closed (2) • Won: Harborline, $52K. Competitive against the incumbent. What separated it on the record: we got to their security team in week two rather than week six. • Lost: Tessera, $38K. Lost to no decision, not a competitor. Budget was never confirmed at any point in the cycle. Worth the meeting Vantage slipped six weeks with nothing on the record explaining it, and it is the largest deal in the quarter. ``` ### Customize the rollout * Set the cadence and the day to match your review, weekly or biweekly * Choose the scope: the whole pipeline, one segment, one team, or one forecast category * Change the comparison window from week over week to since the last review * Set what counts as stalled for you, in days without activity or without a next step * Add the sections your review actually uses and drop the ones it does not * Route it to the review channel, or to each manager's DM with only their team's deals * Chain it with [forecast review](/docs/use-cases/forecast-review) so the pipeline read and the forecast read land together ## Build in Claude or ChatGPT For reading into the pipeline between reviews, and for shaping the format before you automate it. Run the prompt a few times, change what you ask for, then build the workflow around the version people actually want. ### 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, review my pipeline for this week against last week. Break it into: - Progressed: what moved forward, and what caused each move - Slipped: close dates that moved, by how much, and whether this deal has slipped before - Stalled: no meaningful activity, with how long and whether a next step exists - Closed: won and lost, with what the record says separated them Then tell me the one or two deals most worth spending the review meeting on, and why. Ground causes in calls and email. Where a change has no explanation in the record, say so rather than inventing one. ``` That last instruction matters more than it looks. A close date that moved with no reason anywhere in the record is itself the finding. ### Make it yours * Add `Only deals I own` or `Only [rep]'s deals` to narrow it. * Add `Compare this week's stall rate to the last four weeks` to see a trend rather than a snapshot. * Add `Which of these deals look like the ones we lost last quarter` when you have enough closed-lost history to pattern-match against. ### The Skill (Claude) The `clearskies:pipeline-review` skill in the Clearskies plugin for Claude reviews pipeline health and prioritizes accounts by urgency rather than close date, surfacing what needs attention this week. #### Trigger phrases * "Review my pipeline" * "What changed in the pipeline this week" * "What needs attention this week" #### Customize it for your team * Match the sections to your review agenda * Set your own definition of stalled * Name the segments or teams you always want broken out separately Want this packaged for your team? [**Reach out**](mailto:support@clearskies.cc) and we'll help you customize the skill to your workflow. ## 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)