Platform: Enolytics DTC

Summary:

What it is: Suggested Final Confidence Score combines machine learning predictions (60%) with business rules (40%) to create an overall recommendation strength score from 0 to 1.0, where 0.75 means 75% confidence this is a strong product recommendation for the customer.

How to use it:
• Prioritize outreach by focusing on customers with higher confidence scores, as they indicate stronger purchase signals
• Set confidence thresholds for automated marketing campaigns or staff follow-up workflows to optimize your team's time

Tip: This blended scoring approach captures both data patterns and proven wine business logic like vintage continuation preferences, so trust higher scores for your most promising recommendation opportunities.

Note: Parts of this article were generated with AI and may not be perfect. If something looks off or could be better, click the 😞 below — it opens a quick chat so you can let us know.

Quick Stats:

The overall recommendation strength score ranging from 0 to 1.0, combining both machine learning predictions (60% weight) and rule-based signals (40% weight). A score of 0.75 represents 75% confidence that this is a strong product recommendation for this customer. This blended approach leverages the pattern-recognition power of ML while incorporating proven business rules like vintage continuation patterns and repurchase frequency. Higher scores indicate stronger purchase signals and should be prioritized for outreach.

Suggested Next Purchase

Type: float

Format: ##0.0%