HubSpot Lead Scoring & Routing (Fit + Engagement) Workflow (HubSpot + Clay + Apollo + Zoho SalesIQ + n8n)

Short answer: HubSpot lead scoring works when it combines two scores: fit (how closely a company matches your ICP) and engagement (what they are doing right now). Score both, set a threshold sales agrees on, and let the score trigger the lifecycle stage change and routing, instead of asking reps to eyeball a single number.

Why most HubSpot lead scoring fails

Most scoring models I see fail for the same three reasons:

• One blended number. A student who reads ten blog posts and a VP at a perfect-fit account who visits pricing once can end up with the same score. A single score hides the difference between fit and intent.
• Points nobody agreed on. Marketing sets the weights, sales never sees them, and within a month reps ignore the score.
• Scores that trigger nothing. A score that sits on the record without changing a stage, an owner or a task is a report, not a system.

The model: fit score + engagement score

Fit score (who they are). Built from firmographics that match your ICP: industry, company size, region, tech stack, and the job title of the contact. Fit changes slowly, so it can be enriched once and refreshed on a schedule.

Engagement score (what they are doing). Built from recent behavior: visits to high-intent pages (pricing, services, comparisons), form fills, email replies, and site intent signals from a tool like Zoho SalesIQ. Engagement should decay over time so old activity doesn’t keep a lead hot forever.

The rule that makes it work: a contact only becomes an MQL when both scores clear their threshold. High fit with low engagement goes to a nurture or outbound list; high engagement with low fit gets a lighter-touch path.

How it looks end to end

flowchart LR
  A["New or updated contact"] --> B["Enrich firmographics"]
  B --> C["Fit score (ICP match)"]
  A --> D["Engagement score (intent + activity, with decay)"]
  C --> E{"Both above threshold?"}
  D --> E
  E -->|yes| F["Lifecycle stage: MQL"]
  E -->|no| G["Nurture / outbound list"]
  F --> H["Assign owner + task + sequence"]
  H --> I["SQL when sales accepts"]

When it fails

• Empty fit fields: if enrichment misses, the contact is flagged for research instead of scoring as zero-fit.
• Score inflation: engagement decay stops old activity from keeping leads hot.
• Duplicate records: dedupe before scoring, or the same person gets routed twice (see CRM write-back).
• No owner available: fall back to a default owner and alert the team so nothing sits unassigned.

Stack

HubSpot (scoring, lifecycle stages, workflows, sequences), Clay and Apollo for enrichment, Zoho SalesIQ for site intent, n8n for anything HubSpot workflows can’t do natively.

Where I used this

At Mavlers I scored partner-agency visitors against the ICP and moved them through the same MQL to SQL stages sales already trusted, so internal sales worked SQL-ready accounts. Read the Mavlers case study, or see how visitors are captured in the website visitors to HubSpot workflow.

Need this built in your HubSpot? Work with me as your RevOps consultant or hire me for this.