Skip to content
Sadra Raadfar
All work
Marketing AutomationPortfolio project2026

Marketing Lead Nurturing System

Behaviour-driven segmentation and email sequencing with measurable stage movement.

  • Workflow orchestration
  • CRM
  • Email platform
  • PostgreSQL
  • Analytics
01

Overview

A nurture system built around explicit state: every contact sits in exactly one stage, and every transition is caused by a rule you can point at.

02

The problem

Leads that are not sales-ready get one generic newsletter and then silence. There is no segmentation, no signal from behaviour, and no way to tell which content actually moves someone forward.

03

Existing process

  1. 01All non-qualified leads land on one list.
  2. 02A monthly newsletter goes to everyone, regardless of interest or stage.
  3. 03Engagement data lives in the email tool and never reaches the CRM.
  4. 04No one can answer which content precedes a sales conversation.
04

Proposed system

  • Segment on two axes: firmographic fit and observed behaviour.
  • Rolling engagement score with time decay so stale activity fades out.
  • Branching sequences per segment, each with a defined exit condition.
  • Behaviour triggers (pricing visit, repeat opens, doc downloads) can promote a contact mid-sequence.
  • Stage transitions written back to the CRM so sales sees the same state marketing does.
05 — Workflow

The system, node by node

  1. 01

    Contact intake

    New or updated contact enters the state machine

  2. 02

    Segmentation

    Fit tier × behaviour tier assigns a segment

  3. 03

    Scoring

    Engagement score recomputed nightly with decay

  4. 04

    Sequence selection

    Branch chosen per segment, cadence capped

  5. 05

    Trigger listening

    High-intent events interrupt and promote

  6. 06

    Handoff

    Threshold crossed → sales task + CRM stage change

  7. 07

    Reporting

    Stage movement, sequence performance, unsubscribes

  1. Contact intake
  2. Segmentation
  3. Scoring
  4. Sequence selection
  5. Trigger listening
  6. Handoff
  7. Reporting
06

Business logic

Frequency cap
Maximum two automated emails per contact per week across all sequences.
Intent override
A pricing-page visit plus two opens in 7 days promotes to sales-ready.
Decay
Engagement points lose 10% weekly so old activity cannot hold a stage open.
Exit conditions
Reply, meeting booked, unsubscribe or stage change ends the sequence immediately.
Suppression
Open deals and current customers are excluded from acquisition sequences.
07

Failure handling

Email provider rate limit
Queue with a token bucket; sends spread rather than dropped.
Bounce or complaint
Contact suppressed instantly and flagged in the CRM.
Conflicting sequences
Priority order resolves overlap; only one active sequence per contact.
Missing behaviour data
Contact defaults to the lowest behaviour tier rather than being skipped.
08

Demonstration

Simulation notice

This project is a portfolio simulation. Sequences were tested against a synthetic contact set with scripted behaviour events to verify branching, capping and exit conditions. No client data was used and no performance claims are made.

09

Design decisions

  • Nurture is a state machine

    Once every contact has one stage and explicit transitions, debugging becomes reading a log instead of guessing.

  • Caps protect the channel

    The frequency cap sits above every sequence, so adding a new campaign cannot accidentally double someone's email volume.

  • Write state back to the CRM

    Marketing automation that sales cannot see recreates the silo it was meant to remove.

Next

Want a process mapped like this?

Contact Me