This Recommendations widget highlights items gaining momentum in your store by analyzing changes in customer engagement over time.
Tracking Popularity Growth
We track daily product view counts over a two-week period, prioritizing recent activity:
- Daily view count changes are calculated
- Recent trends carry greater weight
- Products are ranked based on growth patterns, not just total views
Delta Ranking Methodology
We compare products based on recent engagement:
Example:
Product A Daily Views: 10, 8, 6, 4, 4, 5 (most recent to oldest)
Product A Daily Change: +2, +2, +2, 0, -1
Product B Daily Views: 15, 15, 16, 14, 15 (most recent to oldest)
Product B Daily Change: 0, -1, +2, -1
Even though Product B has more total views, Product A demonstrates consistent growth. Our ranking prioritizes upward momentum.
Popularity Ranking Methodology
We apply a weighted view system to assess sustained interest:
Example:
Product A Daily Views: 10, 10, 10, 10, 10, 50 (most recent to oldest)
Product B Daily Views: 50, 10, 10, 10, 10 (most recent to oldest)
Days Weight: 1, 0.5, 0.25, 0.125, 0.0625 (most recent to oldest)
Though both products have 90 total views, Product A has 21.875 weighted views, while Product B has 58.8125 weighted views - making Product B rank higher.
Balanced Selection Criteria
We ensure accurate trend detection by considering:
- Growth Rate (Delta Ranking): How quickly a product gains views
- Baseline Popularity (Popularity Ranking): Weighted view totals over 14 days
This balanced approach ensures:
- Genuine trending products are featured
- Items with erratic low-view data are filtered out
- Reliable trends are surfaced based on substantial data
By leveraging these methodologies, our system helps customers discover products that are truly gaining traction in your store.