What: Using Breeze Assistant to run a structured governance audit of your deal pipeline — identifying stale deals, stage definition violations, probability misalignment, missing required fields, and forecast-distorting patterns - and outputting a prioritised action plan to fix them.
Prompt of the week:
This isn't a cosmetic CRM tidy-up. This is about asking Breeze to look at your pipeline the way a revenue integrity auditor would: where are deals stuck, where has the process been bypassed, and where is the forecast lying to you?
The root causes keep coming up across forums, partner blogs, and CRM best practice guides:
The default HubSpot pipeline comes with generic stages that fit almost no actual sales process, and companies that leave these in place end up with pipelines that don't reflect how deals actually progress. When stage definitions are ambiguous, probability scores are fiction. A messy pipeline full of stale deals doesn't just slow things down - it skews reporting, hurts forecasting, and wastes the team's time.
That tension - between what reps do and what governance requires - is exactly where this prompt lives. Instead of manually combing through your pipeline board trying to spot violations, you can hand the job to Breeze and get a structured audit back in seconds.
Prompt structure
Paste this into Breeze Assistant and make sure CRM data access is enabled in your AI settings so Breeze can reference your live deal records:
Role: You are a Revenue Operations Analyst conducting a pipeline
governance and revenue integrity audit for [COMPANY NAME].
Task: Analyse the current deal pipeline and produce a structured
Pipeline Health Audit Report that identifies governance violations,
revenue risk, and forecast integrity issues.
Context:
- Pipeline name: [PIPELINE NAME, e.g., "New Business"]
- Sales cycle: [TYPICAL LENGTH, e.g., "45–60 days"]
- Number of active reps: [NUMBER]
- Current quarter close date: [DATE]
- Our deal stages and expected progression:
1. [STAGE 1, e.g., "Discovery Completed"]
2. [STAGE 2, e.g., "Qualified Opportunity"]
3. [STAGE 3, e.g., "Solution Presented"]
4. [STAGE 4, e.g., "Proposal Sent"]
5. [STAGE 5, e.g., "Negotiation"]
6. [STAGE 6, e.g., "Closed Won / Closed Lost"]
Audit the following areas:
1. STALE DEAL DETECTION
- Flag any open deal that has not changed stage in more than
[THRESHOLD, e.g., 21 days]
- Flag any deal with a close date in the past
- Categorise each as: Needs Update / Likely Dead / Requires
Manager Review
2. STAGE INTEGRITY CHECK
- Identify deals that appear to have skipped stages
(e.g., jumped from Discovery to Proposal with no Qualification)
- Identify deals sitting in early stages with close dates
within 14 days (unrealistic progression)
- Flag any deals created directly in late stages
(bypassing entry criteria)
3. DATA COMPLETENESS
- For each stage, check whether key properties are populated:
- [STAGE 2]: Deal Amount, Decision Maker Contact
- [STAGE 4]: Proposal Number/Reference, Close Date
- [STAGE 6 — Closed Lost]: Closed Lost Reason
- List deals with missing critical fields, grouped by rep
4. FORECAST RISK ASSESSMENT
- Calculate the ratio of deals with close dates this quarter
vs. deals showing no activity in the last 14 days
- Highlight the top 5 highest-value deals at risk of slipping
- Note any rep whose pipeline is disproportionately weighted
in a single stage (concentration risk)
5. PROBABILITY CALIBRATION CHECK
- Compare the default stage probabilities against the pattern
of closed-won vs. closed-lost deals visible in the pipeline
- Flag if any stage probability appears significantly
misaligned with actual outcomes
- Recommend adjusted probabilities where data supports it
Constraints:
- Be clinical and specific — cite deal names or rep names where
possible
- Do NOT soften findings. This is an audit, not a coaching session
- If data is insufficient for a section, state explicitly:
"INTEL GAP: Insufficient data to assess [area].
Recommend manual review of [specific records]."
- Prioritise findings by revenue impact (highest first)
Output format:
### I. EXECUTIVE SUMMARY
{3-sentence overview of pipeline health and primary risk}
### II. STALE DEALS REGISTER
| Deal Name | Owner | Stage | Days in Stage | Close Date | Status |
### III. STAGE INTEGRITY VIOLATIONS
| Deal Name | Expected Path | Actual Path | Risk Level |
### IV. DATA COMPLETENESS GAPS
| Rep | Deals Missing Fields | Most Common Gap | Impact |
### V. FORECAST RISK FLAGS
{Top 5 at-risk deals with rationale}
### VI. PROBABILITY CALIBRATION
| Stage | Current Probability | Suggested Probability | Basis |
### VII. RECOMMENDED ACTIONS (Priority Order)
{Numbered list of specific, actionable next steps}
Why this prompt works - and how to adapt it
This prompt is deliberately long and structured compared to the email marketing prompts earlier in this series. That's intentional. Pipeline governance is a more complex domain - there are multiple audit dimensions, each with different data requirements and different implications. The prompt needs to guide Breeze through a systematic inspection.
A few things to note about how it's constructed:
The Context section does the heavy lifting. By spelling out your stage names, sales cycle length, and team size, you're giving Breeze the baseline it needs to spot deviations. Without this, it would have to guess what "normal" looks like - and guesses aren't what you want in an audit.
The numbered audit areas act as a checklist. Each one is a distinct lens on pipeline health: stale deals, stage integrity, data completeness, forecast risk, and probability calibration. You can use all five, or strip it back to just the two or three that matter most for your current quarter. The structure is modular - take what you need.
The Constraints section enforces rigour. The instruction to "not soften findings" matters. Without it, Breeze will default to diplomatic language that might sound pleasant but buries the actual risk. In an audit context, you want direct, clinical output. The INTEL GAP fallback prevents Breeze from fabricating analysis where the data is thin - a critical guardrail.
The Output Schema keeps it actionable. Tables force specificity. Named deals, named reps, numbered actions. This isn't a report that sits in a drawer - it's designed to be dropped straight into a pipeline review meeting or a RevOps standup.
Adapting it for your portal:
You'll want to adjust the stage names and required properties to match your actual pipeline configuration. If you're running multiple pipelines (e.g., New Business and Renewals), run the prompt once per pipeline - don't try to audit both in a single pass.
The stale deal threshold (21 days in the example) should reflect your sales cycle. A 90-day enterprise cycle might use 30 days; a high-velocity SMB motion might use 10. Match it to what "stuck" actually means for your team.
If you're running this regularly (and you should - quarterly at minimum, monthly if you can), save the prompt as a template and update only the quarter close date and any structural changes to your pipeline. Consistency in the audit format makes it easy to spot trends over time: are the same reps showing up with missing fields every quarter? Are the same stages consistently leaking deals? That pattern recognition is where the real governance value lives.
Beyond the prompt:
Think of this prompt as the AI-assisted version of the pipeline review that your RevOps team probably wishes they had time to do properly every month. Hiring plans, board reporting, and quota setting all rely on stage conversion integrity. If your CRM feels like the Wild West, a lack of policy is usually the culprit.
The good news is that HubSpot has been steadily shipping the features that make governance enforceable - pipeline rules now let you restrict which stages deals can be created in, prevent stage skipping, and restrict backward movement. This prompt helps you find the problems; those features help you prevent them from recurring.
Combine this audit prompt with a custom Breeze Assistant (like we covered in the last instalment) configured specifically for RevOps - give it a clinical tone, access to your CRM records, and a Knowledge Vault containing your pipeline definitions and governance policy - and you've got yourself a repeatable, team-wide audit tool that anyone can run without needing to be a RevOps specialist.
So go and audit your pipeline - the forecast will thank you for it!
