Company-level attribution and ROI engine

The situation

A B2B medical supplier earns much of its revenue from reorders, and the buying relationship sits with the practice rather than any one contact. Native HubSpot attribution couldn’t answer which marketing and sales activities actually drove revenue across the full relationship.

What I built

• A company-level attribution model, deliberately not contact-level, because the sale belongs to the account.
• A nightly pipeline on a managed server that extracts the full HubSpot database, rebuilds company, contact and touch relationships, applies recency and reorder logic, joins a cost sheet, and writes finished numbers back into HubSpot reports. It runs off-platform because the job spans the entire database.
• Four reports inside HubSpot: acquisition (first-order revenue, crediting the touch that opened the relationship), full value (total revenue including every reorder), ROI by initiative (credited revenue against cost per event, campaign and channel), and activities to closure (activities ranked by correlation with closed deals).
• A cost sheet with an automated one-way sync, so the client keeps maintaining costs somewhere familiar.
• Data hygiene: fuzzy and domain-based matching of unassociated contacts to companies, with contacts that match several companies flagged for a person to review instead of guessed.

Stack

HubSpot CRM APIHubSpot custom reportingNightly server jobsFuzzy matching

Related service: Attribution & Reporting.