RB2B Intent Triage
Full-export triage - route every RB2B row to the right action, prioritise by intent signals, and output a single outreach-ready contact list. Every email returned is graded A-F for deliverability risk, and you pick which grades to keep.
example prompts
-
Triage my RB2B visitors by intent and tell me who to prioritize -
Rank these RB2B website visitors by buying signals
about this skill
What this skill does
Takes your full RB2B visitor data - pulled via the Profiles API or from a CSV export - and routes each row to the correct enrichment action:
| Row type | Action |
|---|---|
| Person - is ICP | Enrich directly with email + phone |
| Person - not ICP, company fits | Find ICP contacts at their company |
| Company row | Find ICP contacts at that company |
Rows are prioritised by intent signals before processing: AllTimePageViews, RecentPageCount, NewProfile, and Tags (e.g. "Hot page"). The output is a single ranked outreach list.
Chain
RB2B Profiles API or full CSV export
to define ICP + company fit criteria
to triage each row into buckets
to enrich ICP visitors directly (linkedin_to_business_email per-URL parallel + linkedin_to_mobile_phone batch)
to find ICP contacts at company-only + non-ICP-visitor accounts
(search_people to linkedin_to_business_email per-URL parallel + linkedin_to_mobile_phone batch)
to merge + rank by intent signal
Inputs
- RB2B visitor data - two ways in:
- Connect to the RB2B Profiles API using RB2B's Profile API skill - pulls visitors straight from RB2B, no export needed (requires an RB2B plan with API access)
- Upload an RB2B CSV export - upload or paste (full export)
- Either way, works on any mix of Person/Company rows
- ICP definition - role/title, seniority, department criteria
- Company fit criteria - optional industry, size, or revenue filter
- Acceptable risk scores - which A-F deliverability grades to keep (recommended: A, B, C and F; D excluded)
Output
| Priority | Name | Company | Signal | Source | Title | Business Email | Risk | Mobile |
|---|
Rows processed: X (X Person, X Company)
ICP visitors: X to enriched directly
Non-ICP visitors: X to X unique companies to X ICP contacts found
Company rows: X to X ICP contacts found
Total outreach list: X contacts
Business emails: X/N (XX%)
Mobile numbers: X/N (XX%)
Email risk scores:
A: X B: X C: X D: X F: X
kept (accepted grades): X
filtered (excluded): X
Search filters
search_people filters on query, company, company_domain, country, state, city, seniority, department or functional_area, industry, linkedin_industry, naics_code, employee_range, and revenue_range.
- Filters are free precision - execution cost comes from the free-text
query, so a filters-only search runs ~40x cheaper and lands more precisely. A barequeryis never the right shape; at minimum it getscountryattached (~99% fill rate). - Size and revenue apply to the person's current employer - "VPs at 51-200 employee software companies" is one call, not a company search followed by a people search.
- Three industry vocabularies - broad
industrybuckets, LinkedIn's ~150 niche labels, or a NAICS code at any depth. Shortening the code widens the net. - Sparse filters are the usual cause of a thin result set -
seniority/industry/departmentare ~60% filled andnaics_code/linkedin_industry~50%, and every filter is exact-match, silently dropping records where the field is empty. - Probe before you pull -
search_peoplehas no count-only mode, so run it atlimit: 1first and readresults.totalbefore paginating. (count_only: trueis free, but only onsearch_linkedin_profile.)
Tips
NewProfile = true= first visit - high urgency signal- Tags like "Hot page" from RB2B surface in the priority ranking automatically
AllTimePageViews+RecentPageCounttogether indicate sustained vs. spiking intent- For a single workflow (not triage), use focused skills: Enrich RB2B Visitors, RB2B Companies to ICP Contacts, or RB2B Non-ICP Visitors to ICP Contacts
- Every returned email carries an A-F deliverability risk score; you're asked up front which grades to keep, and out-of-range grades are shown and flagged rather than silently dropped