Datahub

Product quality scoring

Vendably scores each source and destination on a 0-100 scale across six quality dimensions. Understand what each dimension measures and how to improve your score.

Last verified 29 Jun 2026

No results found

Try different keywords or browse all help articles.

    The product quality score is a 0-100 composite score that Vendably calculates daily for each connected source and each configured destination. It measures how ready your product data is for distribution across channels. A higher score reduces the likelihood of product disapprovals and validation failures.

    The score is distinct from the feed health score, which is a source-level measure. Product quality scoring applies the same six-dimension framework to both sources (based on import logs) and destinations (based on export logs).

    The six dimensions

    Dimension Weight What it measures
    Completeness 30% Average coverage of the price and category fields across active products
    Accuracy 20% Percentage of valid records across recent import or export logs
    Freshness 15% How recently a successful import completed
    Error rate 15% Ratio of successful imports to total imports over the lookback window
    Consistency 10% Stability of product count across recent imports
    Enrichment 10% Coverage of rankings and priority data

    Scores are calculated over a rolling 7-day lookback window.

    Score thresholds

    Score Status
    80 and above Healthy (green)
    50 to 79 Warning (amber)
    Below 50 Error (red)

    Understanding each dimension

    Completeness (30%)

    Vendably checks whether each active product has a value for price and category. The score is the average coverage of those two fields across your catalogue. Missing price or category is the single most common cause of a low overall score, and it is also the most impactful to fix.

    To improve: ensure every active product has a price and a category assigned. The Rules Engine can fill missing categories automatically using the Set Value or Lookup Table actions. For missing prices, check whether the source feed is supplying price data and that the field mapping in your source configuration is correct.

    Accuracy (20%)

    Accuracy reflects the percentage of records in recent import logs that contain no errors. Field type mismatches, malformed prices, and invalid values all reduce this score.

    To improve: open the import log for your source and filter for error records. Resolve field mapping issues in the source configuration or use the Rules Engine to clean values before they reach the destination.

    Freshness (15%)

    Freshness decays linearly from 100 to 0 over 7 days since the last successful import. A source that last imported today scores 100 on freshness; one that has not imported in 7 or more days scores 0.

    To improve: increase the import frequency in your source settings, or enable webhooks if your platform supports them. If a scheduled sync has stopped running, check the source for connection errors.

    Error rate (15%)

    Error rate measures the proportion of imports over the last 7 days that completed without errors. A source with consistent import failures will score low here even if individual records are valid.

    To improve: investigate recent import failures in the import log. Repeated failures often indicate an authentication issue, a platform-side change to the feed format, or a network problem reaching the source.

    Consistency (10%)

    Consistency measures how stable your product count is across recent imports. Large swings in product count reduce this score. Some fluctuation is normal during initial setup; unexpected drops in an established catalogue may indicate a sync problem.

    To improve: enable the product count failsafe in your source settings. When active, the failsafe pauses imports automatically if the product count drops by more than a configurable threshold, preventing a partial or malformed sync from propagating to your destinations.

    Enrichment (10%)

    Enrichment measures the coverage of rankings and priority data across your products. Products without enrichment data score lower on this dimension.

    To improve: connect a Google Analytics 4 source in DataHub. Vendably uses GA4 data to derive sales and performance signals, which feed into the enrichment score and also improve product prioritisation in feeds.

    Quality alerts

    Vendably can notify you when quality conditions change. Alerts are available for:

    • Import failure
    • Export failure
    • Quality score drop
    • Product count drop
    • Error spike
    • Disapproval spike
    • Data freshness becoming stale

    Alerts have a default cooldown of 60 minutes to avoid repeated notifications for the same condition. Notifications are sent by email. Configure alerts in the source or destination settings panel.

    See also