What: Using Breeze Assistant to audit the intent signals your HubSpot portal is currently capturing (and the ones it's missing), then producing a structured lead scoring blueprint that maps specific behavioural, firmographic, and engagement signals to score weightings - ready to implement in HubSpot's scoring properties.
Prompt of the week:
This isn't a generic "here's how lead scoring works" exercise. This is about asking Breeze to look at your portal's data model - the forms, the pages, the email engagement, the lifecycle stages - and tell you where the gaps are between what you're tracking and what actually predicts a conversion.
The buyer journey has fundamentally changed. HubSpot's own State of Marketing 2026 Report found that over half of respondents say search volume is down but searches have higher intent - meaning buyers are further along in their journeys when they arrive. Research suggests that 80% of buying decisions now happen before a seller gets involved.
And yet, most HubSpot portals are still running lead scoring models built on static rules from three years ago - or worse, no scoring model at all. The community stories are full of marketers asking how to set up segmentation, how to identify "high-intent" behaviour, and how to stop sales from wasting time on contacts who downloaded one eBook in 2023 and haven't been seen since.
The real problem isn't a lack of data. HubSpot is drowning in behavioural signals - page views, email clicks, form submissions, meeting bookings, content downloads, return visits. The problem is that nobody has sat down and mapped which of those signals actually matter, how much they should weigh, and when a score should decay because the contact has gone quiet.
That's what this prompt does.
Prompt structure
Paste this into Breeze Assistant and make sure CRM data access is enabled in your AI settings so Breeze can reference your portal's properties, forms, and content:
Role: You are a Revenue Operations Strategist specialising in
lead qualification and buyer intent modelling for B2B companies
using HubSpot.
Task: Conduct an Intent Signal Audit of our HubSpot portal and
produce a Lead Scoring Blueprint that maps behavioural,
firmographic, and engagement signals to a weighted scoring model
we can implement in HubSpot's lead scoring properties.
Context:
- Company: [COMPANY NAME]
- Industry: [INDUSTRY]
- Primary product/service: [DESCRIPTION]
- Average sales cycle: [LENGTH, e.g., "30–45 days"]
- Current lead volume: [APPROX. MONTHLY NEW CONTACTS]
- Target buyer persona(s): [ROLE/TITLE, e.g., "Marketing Directors
at mid-market SaaS companies, 50–500 employees"]
- Current scoring model: [DESCRIBE - e.g., "None" / "Basic manual
HubSpot Score with a few rules" / "Predictive scoring enabled"]
Audit the following areas:
1. HIGH-INTENT BEHAVIOURAL SIGNALS (First-Party)
Identify which of these actions are trackable in our portal
and recommend a point weighting for each:
- Pricing page visits (single vs. repeat)
- Demo/consultation request form submissions
- Case study or testimonial page views
- Product comparison page visits
- Return visits within a 7-day window
- Bottom-of-funnel content downloads
(e.g., buyer's guides, ROI calculators)
- Meeting link clicks or bookings
- Email replies to sales sequences
For each signal, categorise as:
HIGH INTENT (strong buying signal) /
MEDIUM INTENT (active evaluation) /
LOW INTENT (early research)
2. ENGAGEMENT SIGNALS (Email & Content)
Recommend scoring rules for:
- Marketing email opens (threshold before scoring)
- Email click-throughs (weight by content type)
- Blog post views (volume threshold)
- Webinar/event registrations vs. attendance
- Content re-engagement (returning after 30+ days of silence)
Flag any signals that should receive NEGATIVE scoring:
- Email unsubscribes
- Hard bounces
- Spam complaints
- Repeated visits to careers page (job seeker, not buyer)
- Visits with no interaction (single-page bounce)
3. FIRMOGRAPHIC FIT SCORING (ICP Match)
Based on our target persona, recommend point weightings for:
- Job title / seniority match
- Company size (employee count range)
- Industry match
- Geographic region match
- Technology stack indicators (if available)
Flag any firmographic signals that should DISQUALIFY
or negatively score a lead:
- Competitor domains
- Student/educational email addresses
- Company size below minimum threshold
- Geographic regions we do not serve
4. SCORE DECAY RULES
Recommend a decay model that reflects our sales cycle:
- After how many days of inactivity should scores begin
to decay?
- What percentage should decay per period?
- Should decay apply equally to all signal types, or should
firmographic fit scores be exempt from decay?
- At what score threshold should a lead be automatically
moved to a "Recycle" or "Nurture" stage?
5. THRESHOLD & ROUTING RECOMMENDATIONS
Based on the scoring model above, recommend:
- MQL threshold (score at which marketing passes to sales)
- SQL threshold (score at which sales should prioritise)
- "Hot lead" threshold (immediate alert to rep)
- Routing logic: should leads be routed by score alone,
or score + firmographic fit?
Constraints:
- All recommendations must be implementable in HubSpot's native
lead scoring properties (manual or predictive)
- Use specific point values, not vague ranges
- If a signal requires a property or tracking that may not exist
in our portal, flag it as: "SETUP REQUIRED: [what needs
configuring]"
- Do NOT recommend third-party intent data tools - keep this
to first-party signals available in HubSpot
- If data is insufficient to recommend a weighting, state:
"CALIBRATION NEEDED: Recommend running for 90 days then
adjusting based on closed-won correlation"
Output format:
### I. SIGNAL AUDIT SUMMARY
{3-sentence overview of current signal coverage and primary gaps}
### II. HIGH-INTENT SIGNAL MAP
| Signal | Intent Level | Points | Rationale |
### III. ENGAGEMENT SCORING RULES
| Signal | Points | Threshold/Condition | Decay Applies? |
### IV. NEGATIVE SCORING SIGNALS
| Signal | Points (negative) | Rationale |
### V. FIRMOGRAPHIC FIT SCORING
| Attribute | Ideal Match Points | Partial Match | Disqualifier |
### VI. DECAY MODEL
| Parameter | Recommendation | Rationale |
### VII. THRESHOLDS & ROUTING
| Threshold | Score | Action Triggered |
### VIII. IMPLEMENTATION CHECKLIST
{Numbered list of specific setup steps in HubSpot,
in priority order}
Why this prompt works - and how to adapt it
This is one of the more comprehensive prompts in the Prompt Lab series so far, and that's deliberate. Lead scoring sits at the intersection of marketing strategy, sales operations, and CRM configuration - you can't do it justice with a two-line instruction.
A few things to note about how it's constructed:
The Context section is critical. Lead scoring is entirely dependent on your business model, sales cycle, and buyer persona. A SaaS company with a 14-day trial-to-close motion scores very differently from a professional services firm with a 6-month enterprise cycle. By spelling out these variables, you give Breeze the baseline it needs to calibrate point values that actually make sense for your pipeline.
The five audit areas are designed to be modular. If you already have firmographic scoring nailed but your behavioural signals are a mess, strip out Section 3 and focus on Sections 1 and 2. If you've never thought about score decay, Section 4 alone is worth running as a standalone prompt. Take what you need.
The "SETUP REQUIRED" and "CALIBRATION NEEDED" flags are guardrails. Lead scoring models only work if the underlying data is being captured. If Breeze identifies that you're not tracking pricing page visits (because the HubSpot tracking code isn't on that page, for instance), it needs to tell you rather than silently recommending a score for data that doesn't exist. These flags turn the output from a theoretical model into a practical implementation plan.
The Negative Scoring section is the one most people forget. Everyone focuses on adding points for good behaviour, but failing to subtract points for disqualifying signals means your pipeline fills up with job seekers, competitors, and contacts who unsubscribed two years ago. Negative scoring is what keeps the model honest.
Adapting it for your portal:
If you're starting from scratch (no scoring model at all), run the full prompt and treat the output as your v1 blueprint. Implement it in HubSpot's manual lead scoring properties, run it for 90 days, then compare scores against your actual closed-won deals to see which signals are predictive and which are noise.
If you already have a scoring model but it feels stale or unreliable, add this line to the Context section:
Our current scoring rules are:
- [LIST YOUR EXISTING RULES, e.g., "Form submission = +10,
Email open = +1, Pricing page visit = +5"]
Breeze will then audit your existing model against best practices and recommend specific adjustments rather than starting from zero.
If you're on HubSpot Enterprise and have access to predictive lead scoring, this prompt is still valuable - it gives you the manual overlay that validates what the AI model is doing and helps you explain to sales leadership why a lead is scored the way it is. Predictive models are powerful but opaque; this blueprint makes the logic visible.
Beyond the prompt:
The lead scoring blueprint is the strategy layer. Once you've got it, the implementation follows a clear path inside HubSpot:
Set up the HubSpot Score contact property with your positive and negative rules. Create workflows that trigger lifecycle stage changes when score thresholds are hit (MQL → SQL → Opportunity). Configure notifications or task creation for your "hot lead" threshold so reps are alerted immediately. Build a dashboard that tracks score distribution across your database so you can spot when the model needs recalibrating.
And if you combine this with HubSpot's Buyer Intent feature (available in Sales Hub Professional and Enterprise), you can layer company-level intent signals - like which tracked companies are visiting your site, showing research activity, or going through funding rounds - on top of the contact-level scoring model for a genuinely multi-dimensional view of who's ready to buy.
This is one of those prompts that pays dividends every single week once it's implemented. Every new lead that enters your portal gets scored automatically. Every sales rep knows exactly where to focus. Every pipeline review is grounded in data, not gut feel.
So go and audit your intent signals - your sales team will thank you for it!
