Personalized Recommendations Engine

Introduction

Avidon Health is introducing a Recommendations Engine that uses rule-based logic to surface personalized content for members — connecting them with the right courses, challenges, flows, and resources based on their profile, responses, and readiness.

What's Changing

1. Rule-Based Recommendation Management

Admins define recommendation rules through a dedicated management interface supporting draft, published, and archived states. Each rule targets a specific custom field response and evaluates conditional logic — including age, gender, custom fields, portal groups, subscription tier, and day of week — to determine member eligibility. Rules are prioritized numerically, and effective date ranges control when they activate.

2. Weighted, Multi-Content Recommendations

Each rule can recommend multiple content items — courses, challenges, flows, pages, habit builders, group coaching, or external URLs — each with an assigned weight that influences display order. The engine ranks all matching recommendations across rules so members always see the most relevant content first.

3. Personalized Member Dashboard Components

Two new Page Builder v2 components bring recommendations to the member dashboard:

• Top Match — A prominent card highlighting the highest-weighted recommendation, with thumbnail, description, duration, and direct-action buttons.

• Recommendations Carousel — A scrollable row of additional recommendations displayed as compact cards, swipeable on mobile and arrow-navigable on desktop.

4. Conditional Logic & Audience Targeting

Rules evaluate rich conditional expressions combining demographics, custom field values, group membership, and subscription tier. This enables precise audience segmentation — for example, recommending a stress management course only to members aged 30–50 who selected "high stress" on an intake assessment and belong to a specific portal group.

5. Multiple Trigger Sources

Recommendations can be generated from three sources:

• System rules — Evaluated automatically when members update Health Age Predictor question responses

• Admin actions — Manually assigned by coaches or administrators

• Automations — Triggered as steps within automation workflows for event-driven personalization

6. Full Lifecycle Tracking

Each member recommendation moves through a complete status lifecycle: pending, accepted, rejected, expired, or completed. Scheduled and expiration dates control visibility windows, and the system records creation, completion, and source attribution for reporting.

Why It Matters

• Personalized engagement — Members receive content matched to their profile, responses, and readiness — not generic suggestions

• Reduced manual effort — Rule-based automation replaces manual content assignment at scale

• Flexible content targeting — Seven redirect types mean recommendations can point members to any content format on the platform

• Auditable and controlled — Draft/publish workflow, priority ordering, date ranges, and full audit logging give admins precise control over what members see and when

• Scalable personalization — System rules evaluate automatically at the moment of member interaction, scaling personalized guidance without additional admin overhead