Stale Opportunity Reviver

Stale Opportunity Reviver

Find open HubSpot deals that have gone stale, verify contacts are still in their roles, research company news for fresh angles, and generate revival outreach to re-engage dormant pipeline.

Instructions

When a user wants to identify and revive stale opportunities, systematically pull deal data from HubSpot, enrich contacts, research companies, and produce a prioritized revival plan.

Steps

1. Gather revival parameters by asking the user:

   (a) "How many days of inactivity makes a deal stale?" Offer defaults:

   • 14 days (aggressive, for fast sales cycles)
   • 30 days (standard, for mid-market deals)
   • 60 days (conservative, for enterprise deals)
   • Custom number of days

   (b) "Which deal stages should I look at?" Common options:

   • All open stages
   • Specific stages (e.g., "Qualification", "Proposal Sent", "Negotiation")
   • Exclude early-stage (e.g., skip "Appointment Scheduled")

   (c) "Is there a minimum deal amount?" Accept a dollar threshold or "any amount".

   (d) "Filter by deal owner?" Accept an owner name/email or "all owners".

2. Retrieve the current user's identity by calling mcp__claude_ai_HubSpot__get_user_details to get the authenticated user's name, email, and HubSpot user ID. This is needed if the user wants to filter by "my deals" or if you need to resolve owner references.

3. Discover relevant deal properties by calling mcp__claude_ai_HubSpot__search_properties with:

   • object_type: "deals"
   • query: "activity" or "last" to find date fields like notes_last_updated, hs_last_sales_activity_date, hs_lastmodifieddate

   Also search for stage and amount properties to confirm their field names:

   • query: "stage" to confirm dealstage
   • query: "amount" to confirm amount

   This ensures you use the correct property names for filtering and retrieval.

4. Search for stale deals by calling mcp__claude_ai_HubSpot__search_crm_objects with:

   • object_type: "deals"
   • filter_groups: Build filters combining:
     • Deal pipeline stage is not "closedwon" and not "closedlost" (open deals only)
     • Last activity date is before the staleness threshold (today minus X days)
     • Amount is greater than or equal to the minimum (if specified)
     • Owner ID matches the specified owner (if specified)
   • properties: ["dealname", "amount", "dealstage", "pipeline", "hubspot_owner_id", "hs_lastmodifieddate", "notes_last_updated", "hs_last_sales_activity_date", "closedate", "createdate"]
   • limit: 50

   If the user specified deal stages, add a dealstage filter with operator: "IN" and the matching stage IDs.

5. Calculate staleness for each deal. For each returned deal:

   • Determine the last activity date from whichever is most recent: hs_last_sales_activity_date, notes_last_updated, or hs_lastmodifieddate
   • Calculate days since last activity
   • Sort deals by staleness (most stale first)

6. Retrieve associated contacts by calling mcp__claude_ai_HubSpot__get_crm_objects with:

   • object_type: "contacts"
   • object_ids: IDs of contacts associated with the stale deals
   • properties: ["firstname", "lastname", "email", "jobtitle", "company", "phone", "hs_lead_status"]
   • associations: ["deals"]

   Alternatively, for each deal, call mcp__claude_ai_HubSpot__search_crm_objects with object_type: "contacts" and filter by associated deal ID to find related contacts.

7. Resolve deal owners by calling mcp__claude_ai_HubSpot__search_owners with the hubspot_owner_id values from the deals to get owner names and emails. This maps owner IDs to human-readable names for the output.

8. Enrich contacts to verify current status by calling mcp__claude_ai_Amplemarket__enrich_person for the primary contact on each stale deal (limit to top 10-15 deals to manage credit usage):

   • Use email or linkedin_url from the HubSpot contact record
   • Set reveal_email: true
   • Compare the enriched current company and title against the HubSpot record
   • Flag contacts who have left the company. These deals need a new contact strategy

9. Research company news for revival hooks by calling WebSearch for the top 10 stale deal companies:

   • Query: "[Company Name] news [current year]" or "[Company Name] announcement [current quarter]"
   • Look for trigger events: new funding, product launches, leadership changes, expansion, earnings reports
   • These provide fresh conversation starters for re-engagement

10. Build the prioritized revival list. Score each stale deal on revival priority:

• High Priority: Large deal + contact still there + recent company news (fresh hook available)

• Medium Priority: Moderate deal + contact still there but no news, OR large deal but contact has left

• Low Priority: Small deal + no news + contact may have left

  Present the output with:

• Executive summary (total stale deals, total pipeline value at risk, distribution by priority)

• Priority-ordered deal table with all dynamic fields

• Per-deal revival recommendations with suggested openers

• Flagged deals where the contact has left (need new champion)

Important Notes

• Limit enrich_person calls to the top 10-15 deals by amount to manage Amplemarket credit usage. For larger batches, warn the user: "I found X stale deals. Enriching all contacts would use X credits. Shall I focus on the top 10-15 by deal size?"
• If the HubSpot account uses custom deal stages, the stage names may not match standard labels. Use search_properties to discover the actual stage values.
• Deals with no associated contacts should be flagged separately: "This deal has no contacts. It may need manual review in HubSpot."
• The staleness calculation depends on which activity date fields are populated. If hs_last_sales_activity_date is empty, fall back to notes_last_updated, then hs_lastmodifieddate.
• When a contact has left the company, do not suggest re-engaging them at the old company. Instead, recommend finding a replacement contact or using the deal-contact-gap-analysis skill.

Dynamic Fields Generated

The following dynamic fields are populated for each stale deal and can be used in revival outreach templates:

Examples

Example 1: Standard 30-Day Stale Deal Review

User prompt: "Find my deals with no activity in the last 30 days. My email is jrodriguez@ourcompany.com"

What the skill does:

1. Calls mcp__claude_ai_HubSpot__get_user_details to confirm the user identity.
2. Calls mcp__claude_ai_HubSpot__search_properties with object_type: "deals", query: "activity" to find the correct last-activity date field.
3. Calls mcp__claude_ai_HubSpot__search_crm_objects with object_type: "deals", filtering for open deals owned by jrodriguez@ourcompany.com with last activity before 2026-02-07.
4. Calls mcp__claude_ai_HubSpot__get_crm_objects for associated contacts.
5. Calls mcp__claude_ai_HubSpot__search_owners to resolve owner names.
6. Calls mcp__claude_ai_Amplemarket__enrich_person for the primary contact on the top 10 deals.
7. Calls WebSearch for recent news on each deal's company.
8. Produces the prioritized revival list.

Example output:

STALE OPPORTUNITY REPORT - James Rodriguez

Executive Summary

• Stale deals (30+ days no activity): 8
• Total pipeline at risk: $742,000
• High priority: 3 | Medium priority: 3 | Low priority: 2
• Contacts who left their company: 1 (needs new champion)

Priority Revival Table

Action Items:

1. FinServ deal ($65K) needs a new contact. Rachel Kim has left the company. Use deal-contact-gap-analysis to find a replacement.
2. Top 3 deals represent $520K in pipeline. Prioritize revival outreach this week.
3. TechFlow and DataSync both have strong trigger events. They are the best candidates for immediate re-engagement.

Example 2: Enterprise Deals Over $100K Stale for 60+ Days

User prompt: "Which opportunities over $100K have been stuck for more than 60 days? Show me across all reps."

What the skill does:

1. Calls mcp__claude_ai_HubSpot__get_user_details to get authenticated user context.
2. Calls mcp__claude_ai_HubSpot__search_properties to discover activity date fields.
3. Calls mcp__claude_ai_HubSpot__search_crm_objects filtering for open deals with amount >= 100000 and last activity before 60 days ago, without owner filter.
4. Resolves all deal owners via mcp__claude_ai_HubSpot__search_owners.
5. Enriches top contacts and searches for company news.
6. Groups results by deal owner for management review.

Example output (abbreviated):

STALE ENTERPRISE DEALS - All Owners (60+ Days, $100K+)

Summary: 5 deals | $1.2M total pipeline at risk | 3 owners affected

Owner Breakdown:

• Lisa Chen: 2 stale deals ($630K). GlobalTech contact has left. Assign new champion immediately. MegaCorp has a strong hook with the new VP Eng hire.
• David Park: 2 stale deals ($385K). EnterpriseOne EMEA expansion is a timely hook. BigScale needs a creative re-approach.
• James Rodriguez: 1 stale deal ($185K). Series C funding provides a natural re-entry point.

Recommended Management Actions:

1. Schedule pipeline review with Lisa Chen. $630K at risk across 2 deals, one with a departed contact.
2. Prioritize MegaCorp and EnterpriseOne for immediate outreach. Both have strong trigger events.
3. BigScale ($140K, 91 days stale) may need to be reclassified. Consider whether it is still a viable opportunity.
4. Consider running a team-wide stale deal review weekly to catch deals before they exceed 60 days of inactivity.

Example 3: Stale Deals in Specific Stage

User prompt: "Find deals stuck in Proposal Sent stage for more than 14 days"

What the skill does:

1. Calls mcp__claude_ai_HubSpot__get_user_details.
2. Calls mcp__claude_ai_HubSpot__search_properties with query: "stage" to find the stage property and its options.
3. Calls mcp__claude_ai_HubSpot__search_crm_objects filtering for dealstage = "proposalpendingreview" (or equivalent ID) and last activity > 14 days ago.
4. Enriches contacts and researches companies.
5. Presents results focused on proposal follow-up strategies.

Example output (abbreviated):

STALE PROPOSALS - Deals in "Proposal Sent" (14+ Days)

Summary: 6 deals stuck in proposal stage | $430K total value

Troubleshooting