RecomNext
Use CasesFeaturesHow it WorksWhy RecomNextPricingResourcesAPI DocsLogin

nextQL in integration

nextQL filters and boosters shape which items appear and how they rank. Configure them on scenarios and rules — not as per-request API parameters. Full syntax lives in the nextQL reference.

Where nextQL applies
SettingTypeDefaultDescription
scenario.filterfilterHard exclusion on candidates.
scenario.boostersbooster[]Score multipliers on matching items.
rules (type=filter|booster)ruleShared expressions via rulesSlugs.
segmentation (auto-nextql)expressionAssign segment names — see Segmentations guide.
hybrid searchparsed queryPrice/category extracted from natural language.

POST /recommendations does not accept runtime filter parameters. Use multiple scenarios or rules for preset combinations.

Filters

Syntax: quoted attribute names, comparisons, and / or, in / not_in.

'stock' > 0

'category' in ['pc', 'console']

'price' < 40 and 'genre' in ['roguelike', 'rpg']

('platform' = 'vr' or 'featured' = true) and 'stock' > 0
Use case
In-stock only on all placements

Create a rulein-stock with 'stock' > 0 and add to every scenario's rulesSlugs.

Use case
PC-only PDP similar

Scenario filter: 'platform' in ['pc', 'multi'] on pdp-similar-pc.

Boosters

Syntax: boost by <factor> where <filter_expression>

boost by 1.5 where 'featured' = true
boost by 1.3 where 'genre' in ['rpg', 'action']
boost by 0.7 where 'stock' < 10
Use case
Promote featured without excluding non-featured

Boosters re-rank; they do not guarantee inclusion. Pair with filters only when items must be excluded entirely.

Use case
Sale items on homepage
"boosters": [
  "boost by 1.4 where 'onSale' = true",
  "boost by 1.2 where 'discountPercent' > 20"
]
Attribute resolution

Attributes resolve from top-level category, then attributes.*. In expressions use'price', not attributes.price. Arrays (e.g. genre) work with in operators.

Related guides
RulesSegmentationsnextQL reference