Pipeline Review Workspace

Inside this Project, Claude carries the structure that makes comparison possible:

• Your team and territory structure — who owns what.
• Your stage definitions and exit criteria — so "stalled" means the same thing every week.
• Your forecast category definitions — commit, best case, pipeline, as your organisation defines them.
• Movement history — what was flagged last week and what happened to it, which is what turns a snapshot into a trend.

It does not retain the full export from previous weeks. Paste the current export each time; the Project remembers the flags and the structure, not every row.

Reach for this Project when:

• You run a recurring pipeline review and want the prep to take minutes.
• You need at-risk deals flagged on consistent criteria rather than on whoever spoke loudest.
• You are preparing a forecast conversation with your own leadership and need the coverage maths and the narrative ready.
• You want to know what moved since last week and what did not.

Use something else when:

• You are working one deal in depth — that is MEDDPICC Deal Strategy Workspace.
• You are preparing a coaching conversation about a person rather than a pipeline — that is Manager Coaching Studio, and the separation is deliberate.

Each review, you provide:

• The pipeline export — CSV or a pasted table. Useful columns: opportunity, account, owner, amount, stage, close date, last activity date, forecast category.
• The review period — this week, this month, the quarter to date.
• Any deals you already want examined, and why.

A last-activity date is the single most valuable column. Without it, stall detection is guesswork.

Upload the roster first. Everything else can follow.

The forecast-versus-actual file is what calibrates optimism. With it, Claude can tell you that your commit category historically lands at 78 per cent rather than treating it as certain.

Claude accepts files up to 30MB each, and there is no fixed limit on how many you add, but everything in a Project's knowledge has to fit the context window. Fewer, sharper files beat a full archive.

Prepare files this way:

• Strip cover pages, legal boilerplate, and navigation chrome — they consume context and carry no signal.
• Prefer text formats. A clean export beats a scanned PDF.
• Name each file for what it contains, not when you exported it. Claude reads filenames.
• One subject per file. A merged reference document is harder to cite precisely.
• Date anything that goes stale, inside the file itself, so Claude can tell you when it is working from old material.

This Project is deliberately staged so you get value from the first session without completing every field. Do the first pass now and the refinement later.

1. Create the Project. Name it for the team and cadence — for example Pipeline Review — West Team, Weekly.
2. Paste the Project Instructions.
3. First pass — fill one placeholder. Complete [FILL IN: Your Team Roster] only, and upload the roster file. Leave the other placeholders as they are. Run a review. You will get a usable narrative, coverage analysis, and at-risk list immediately, using standard stage and stall assumptions that Claude will state explicitly.
4. Second pass — refine after your first review. Once you have seen where the standard assumptions do not match how you work, fill [FILL IN: Your Stage Definitions], [FILL IN: Your Forecast Categories], and [FILL IN: Your Stall Thresholds], and upload the definition files. This is the pass that makes the flagging yours rather than generic.
5. Fill the company name whenever you get to it — it only affects framing.
6. Run it on the same cadence every week. The movement history is the reason this is a Project and not a prompt.

The bracketed fields below are placeholders you replace with your own details before you save the instructions. Leaving them unfilled produces generic output — that is the single most common reason a Project underperforms in its first week.

This Project has five, and they are deliberately staged — see the setup steps. Only the roster is needed for a useful first review.

• [FILL IN: Your Team Roster] — names, territories, and quotas. Required for the first pass; without it, per-owner analysis is impossible.
• [FILL IN: Your Stage Definitions] — what each stage means and what has to be true to leave it. Second pass.
• [FILL IN: Your Forecast Categories] — how your organisation defines commit, best case, and pipeline. Second pass.
• [FILL IN: Your Stall Thresholds] — the number of days without activity at which a deal is stalled, by stage. Second pass. Until you set it, Claude uses 14 days for early stages and 21 for late, and says so.
• [FILL IN: Company Name] — framing only.

Role

You are a pipeline inspection analyst for a sales manager at [FILL IN: Company Name]. You turn a pipeline export into a written review: what the coverage is, what moved, what is at risk, and what to raise. You analyse deals, never people.

Operating rules

• Never predict a close date the data does not support.
• Distinguish stalled from slow but progressing, and state the threshold used.
• Never characterise a rep's performance, effort, or capability from pipeline data.
• Never rank named reps against each other.
• State every assumption inside the output — thresholds, close rates, medians.

Module 1 — Coverage

Total open pipeline against target for the period. Coverage ratio. Weighted figure using the close rate from the knowledge files, with the arithmetic shown. Distribution across territories, naming any that sit outside the band.

The full instructions continue with stall detection, movement tracking, the risk rubric, the five commands, output formats, and escalation.

This Project produces four deliverables:

• Review narrative — a written summary of pipeline health for the period, with the coverage maths and what changed.
• At-risk list — deals flagged with the specific signal that flagged each, ordered by exposure.
• Movement summary — what advanced, what slipped, what was added, what was lost since the last review.
• Action list — what to raise with whom, in the next review.

A weekly review narrative for a fictional six-rep team.

Pipeline review — West Team — week ending 6 March

Coverage. $4.1M open against a $1.15M quarterly target: 3.6x. Against your trailing four-quarter close rate of 24 per cent, that implies roughly $984k — about 86 per cent of target. Coverage is adequate but not comfortable; the gap sits in one territory.

Distribution. Four of six reps are between 2.9x and 4.2x. Two are outside: one at 6.1x, carrying two large early-stage deals that dominate the number, and one at 1.4x, who is the source of the team-level gap.

Forecast. $620k in commit across nine deals. Last quarter your commit category landed at 78 per cent, which would put this at roughly $484k.

At risk

1. Cascade Manufacturing — $145k — Negotiation. No activity in 26 days, against your 21-day late-stage threshold. Close date has moved twice, most recently by three weeks. Two slips plus a stall is the pattern that most often precedes a push. Raise: what specifically has to happen this week, and who owns it on their side.
2. Beacon Health — $210k — Proposal. Largest deal in the quarter, in stage 41 days against a 19-day median. Activity is current, so this is slow rather than stalled — a different problem needing a different question. Raise: what is the actual approval path.
3. Trellis Logistics — $58k — Discovery. Close date is 11 days away while still in Discovery. The date and the stage are inconsistent; one of them is wrong. Raise: which.

Moved this week

• Halden Group — $92k — Discovery to Proposal. Flagged stalled last week; the flag can close.
• Ironwood Partners — $37k — closed won.

Action list

• Cascade and Trellis: date and ownership questions in this week's review.
• Beacon Health: approval path, given it carries 18 per cent of the quarter.
• The 1.4x territory: a pipeline-generation conversation, separate from this review.

Assumptions stated: stall thresholds of 14 and 21 days are the defaults, not yours. Median stage duration is calculated from this export only.