A meta field is a field that has been automatically been added to your Schema by Lytics.

📘
Meta fields are not integration specific. For lists of fields created for specific integrations, reference the documentation for the tool of interest.

Meta Fields

Field ID	Short Description	Data Type	Summary
`_created`	Created Timestamp	date	The earliest historic timestamp found across all data sources
`_id`	Lytics ID	string	Canonical identifier from Lytics identity resolution
`_last_scored`	Last Scored Timestamp	date	Timestamp indicating when the profile was last scored
`_modified`	Modified Timestamp	date	Timestamp indicating when the profile was last modifed
`_segments`	Segment Membership	[]string	All of the segments to which a profile currently belongs
`_segs`	Computed Traits	membership	Segment history (shows in and out times for segment membership)
`_split`	Random Split 1	int	Random profile partitions (0 - 99)
`_split2`	Random Split 2	int	Random profile partitions (0 - 99)
`_num_aliases`	--	int	The number of identifiers in the profile
`_num_events`	--	int	The number of total user events
`_num_streams`	--	int	The number of streams where the user has been seen
`_num_days`	--	int	The number of days where there has been a stream event associated with the user
`_num_conflicts`	--	int	The number of merge conflicts encountered on the traversal (artifact of traversal before indexing)
`_conflicts`	--	map[string]int	The number of conflicts per identifier field
`_num_max_neighbors`	--	int	The number of fragments skipped in traversal due to max neighbors
`_max_traversals`	--	bool	Indicates the profile has hit the maximum number of traversals
`_num_nested_values`	--	int	The number of values in "nested" field types on an entity (profile)
`_total_sz`	--	int	The total byte size of all fields on the profile
`_cust_sz`	--	int	The total byte size of all non-meta fields on the profile
`_internal_sz`	--	int	The total byte size of all meta fields on the profile
`_elasticsearch_size`	--	int	The total byte size attempted to write to Elasticsearch, but failed
`_streamnames`	Stream Names	[]string	All the streams that have been seen on each profile
`_broken_profile_broke_max_size`	--	bool	If true, indicates that this is the reason for the profile's unhealthy state
`_broken_profile_broke_max_fragment_size`	--	bool	If true, indicates that this is the reason for the profile's unhealthy state
`_broken_profile_max_neighbors`	--	bool	If true, indicates that this is the reason for the profile's unhealthy state
`_broken_profile_nested_count`	--	bool	If true, indicates that this is the reason for the profile's unhealthy state

📘
On Unhealthy Profiles, only some Meta fields will continue to be surfaced. View the these fields in our Understanding Profile Health documentation

ML Powered Meta Fields

These meta fields are the result of Lytics machine learning algorithms.

Field ID	Short Description	Data Type	Description
`lytics_content`	Content Topics	map[string]number	Raw topic interests from Lytics Affinity Engine
`lytics_content_inferred`	Inferred Content Topics	map[string]number	Inferred topic interests from Lytics Affinity Engine
`lytics_rollup`	Content Affinities	map[string]number	Content affinities from Lytics Affinity Engine
`score_consistency`	Score Consistency	int	Behavior based score (Consistency)
`score_frequency`	Score Frequency	int	Behavior based score (Frequency)
`score_intensity`	Score Intensity	int	Behavior based score (Intensity)
`score_maturity`	Score Maturity	int	Behavior based score (Maturity)
`score_momentum`	Score Momentum	int	Behavior based score (Momentum)
`score_propensity`	Score Propensity	int	Behavior based score (Propensity)
`score_quantity`	Score Quantity	int	Behavior based score (Quantity)
`score_recency`	Score Recency	int	Behavior based score (Recency)
`score_volatility`	Score Volatility	int	Behavior based score (Volatility)
`segment_prediction`	Lookalike Model Predictions	map[string]number	Scores from Lytics Lookalike and SegmentML models
`segment_prediction_percentile`	Lookalike Model Percentiles	map[string]int	Percentiles of scores from Lytics Lookalike and SegmentML models
`deliver_day`	deliver day	map[string]int	The predicted best day to communicate with the user based on their next predicted event.
`deliver_hour`	deliver hour	map[string]int	The predicted best time to communicate with the user based on their next predicted event.

📘
If you have turned on Scoring by Stream, you will see additional sets of Score fields containing the stream name for each stream being individually scored.

Orchestration Meta Fields

These fields will only be available when orchestration has been turned on in an account.


`holdout_groups`	--	map[string]bool	Membership of holdout group by experience for orchestration simulation
`picked_experiences`	picked experiences	[]string	IDs of out-of-the-box integration experiences that leverage delivery optimization and have been selected at one point for the user.
`needs_message`	Needs Message Score	map[string]number	Score that represents the user's need for a message in a specific stream based on the relative distance to or from the predicted next event in the same stream.
`next_event`	Next Event Prediction	map[string]time	The predicted timestamp an event is expected in a specific stream based on interaction patterns of the user in the same stream.
`next_event_low`	--	map[string]time	Next Event Prediction (Lower Bound)
`next_event_upper`	--	map[string]time	Next Event Prediction (Upper Bound)
`decision_ttl`	--	date	Approximate time that decisioning for the profile will be reevaluated