Skip to main content
The rules engine lets you override ML-generated rankings with business logic. Rules run after scoring and before final ranking, giving you precise control over what appears in recommendations and where.

​
Creating a rule

  1. Navigate to Console > Ranking > Rules Engine.
  2. Click New Rule.
  3. Fill in the required fields:
    • Name — a descriptive label (e.g. “Pin new series to top 3”).
    • Type — the kind of action the rule performs (see below).
    • Priority — higher numbers execute first. Use priority to control which rules take precedence.
    • Active — toggle on to make the rule live.
  4. Add conditions to control when the rule applies.
  5. Configure the action parameters.
  6. Optionally set a schedule (start and end dates).
  7. Click Create Rule.

​
Rule types

TypeWhat it doesExample use
boostMultiply an item’s score by a factorPromote seasonal content
buryReduce an item’s score by a factorDemote low-quality items
pinLock an item into a specific positionFeature a new release at position 1
filterRemove items that match conditionsHide out-of-stock products
diversifyEnsure variety across a metadata fieldNo more than 3 items from the same category
slotReserve positions for items matching criteriaAlways show one editorial pick in the top 5

​
Conditions

Conditions determine which items or users a rule applies to. A rule with no conditions applies to all items for all users.

​
Item metadata conditions

Match against fields in the item’s metadata:
FieldOperatorValueMeaning
genreequalsactionItem genre is “action”
pricegreater_than50Item costs more than 50
stock_statusequalsout_of_stockItem is out of stock

​
Segment conditions

Target a rule to users in a specific segment:
  1. Add a condition and set the field to segment_id.
  2. A dropdown appears listing all available segments.
  3. Select the segment. The rule now only applies to users who belong to that segment.
This is how you build targeted campaigns — for example, boosting a series only for users who haven’t watched it yet.

​
Scheduling

Rules support time-based activation via start and end dates.
ConfigurationBehaviour
Both dates setRule is active only within the window
Start date onlyRule activates at that time, no automatic end
End date onlyRule is active now but deactivates after the deadline
Neither date setRule is always active (subject to the Active toggle)

​
Setting a schedule

  1. In the rule editor, use the Start date and End date datetime pickers.
  2. Rules with a schedule show a scheduled badge in the table view.
  3. Click Clear schedule to remove dates and make the rule always-active.
Scheduling is enforced at the database level, so time zone handling is consistent regardless of where the request originates.

​
Priority and execution order

Rules execute in descending priority order (highest first). When multiple rules affect the same item:
  • A pin at priority 100 places the item before a boost at priority 50 can change its score.
  • A filter at any priority removes the item entirely, so later rules never see it.
Use priority deliberately:
  • 100+ for hard business constraints (compliance, filtering)
  • 50-99 for campaign promotions
  • 1-49 for soft merchandising preferences

​
How rules interact with experiments

When an A/B experiment is running, each variant can include config_overrides that specify:
  • include_rule_ids — only these rules apply for users in this variant
  • exclude_rule_ids — these rules are skipped for users in this variant
This lets you test the impact of specific rules against a control group. See A/B Testing for details.