QualityNavboostCrapsCrapsData
Quality NavboostQuality SignalGoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsCrapsData
SEO Analysis
AI GeneratedThis model is part of Google's NavBoost system, which uses click-through data and user interaction signals to adjust search rankings. NavBoost is one of the most significant confirmed ranking systems, processing user engagement data to determine which results best satisfy search queries. It likely tracks metrics related to user clicks, dwell time, and satisfaction signals that directly influence where pages appear in search results.
Actionable Insights for SEOs
- Monitor for changes in rankings that may correlate with updates to this system
- Consider how your content strategy aligns with what this signal evaluates
Attributes
22badClicksfloat(nilclicksfloat(nilcountrystringnilFull type: String.tThe two-letter uppercase country slice of the CrapsData. Examples: "US", "FR", "BR"
nilFull type: GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsCrapsDevice.tThe device interface and os slice of the CrapsData.
nilFull type: list(GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsFeatureCrapsData.tContains CrapsClickSignals for specific features. (i.e. for mobile, US, metro id - 123")
goodClicksfloat(nilimpressionsfloat(nilThese fields may become legacy fields; we may retire them and use the squashed field (below) instead, to allow for some nesting.
languagestringnilFull type: String.tThe language slice of the CrapsData. Examples: "en", "fr", "pt-BR",
lastLongestClicksfloat(nilThe number of clicks that were last and longest in related user queries.
mobileDataQualityNavboostCrapsCrapsData →nilFull type: GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsCrapsData.tDO NOT USE: Use the above mobile_signals fields instead. DO NOT REMOVE: Field is present in legacy protos in golden tests.
mobileSignalsQualityNavboostCrapsCrapsClickSignals →nilFull type: GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsCrapsClickSignals.tThe portion of this CrapsData aggregated on data from tier 1/2 mobile interfaces in QSessions.
packedIpAddressstringnilFull type: String.tContains a packed string in network byte order, as expected by CrapsIpPrior. Only populated if we looked up the ip_prior_bad_fraction at retrieval time.
patternLevelinteger(nilLevel of pattern. More general patterns get higher values. For URL patterns this field = 0. For example, if we have "http://abc.def.ghi/xyz.html" level 0 pattern will be "http://abc.def.ghi/xyz.html" level 1 pattern will be "p://abc.def.ghi" level 2 pattern will be "p://def.ghi"
patternSccStatsQualityNavboostCrapsStatsWithWeightsProto →nilFull type: GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsStatsWithWeightsProto.tFor pattern data, this will contain stats of the SCC's of the individual urls contributing to the pattern.
querystringnilFull type: String.tsliceTagstringnilFull type: String.tThis field can be used by the craps pipeline to slice up signals by various attributes such as device type, country, locale etc. The slice_tag can be an arbitrary string, and the CrapsData values for each slice_tag are aggregated separately, together with the default empty slice_tag.
nilFull type: GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsCrapsClickSignals.tNot used yet - we will probably move the impressions / clicks / good_clicks bad clicks / last longest clicks into here from top level, and rename those fields to legacy.
unscaledIpPriorBadFractionfloat(nilUsed to assign a prior based on IP address. See quality/navboost/craps/craps-ip-prior.h. This value is prior to the linear transformation (scaling / offset / min / max) that's applied in craps-penalty.cc.
nilFull type: GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsCrapsClickSignals.tWe will start using this one for the retuning rollout.
unsquashedMobileSignalsQualityNavboostCrapsCrapsClickSignals →nilFull type: GoogleApi.ContentWarehouse.V1.Model.QualityNavboostCrapsCrapsClickSignals.turlstringnilFull type: String.tvoterTokenCountinteger(nilThe number of distinct voter tokens (a lower bound on the number of distinct users that contributed to the entry, used for privacy-related filtering).