Skip to main content

Description

  • Log Level MTA captures every marketing touchpoint on each individual user path to conversion.
  • Log Level MTA is the synthesis of several types of data:
    • User context on the conversion event as well as conversion timestamp.
    • User, device, and referrer context for each marketing event.
    • UTM parameter tracking and custom URL parameters for each marketing event
    • Marketing ad object context for each marketing event.
    • Attribution credit assigned to each marketing event per four attribution methodologies: (1) first touch (2) last touch (3) even weight (4) Rockerbox custom multi-touch attribution model.

Table Creation

  • One log level MTA table is created for a single conversion event tracked in Rockerbox.
  • The table is manually created in the Rockerbox UI.

Partition Keys

  • date
💡 Note: Leverage partition keys when querying the table to improve query efficiency.

Primary Key

  • There is no primary key.
  • Use the concatenation of conversion_key and the event sequence_number as a composite primary key.

Field Reference

NameDescriptionType
reportThe name of the report (only visible in Snowflake integrations)str
advertiserRockerbox Account ID (only visible in Snowflake integrations)str
identifierConversion segment ID (only visible in Snowflake integrations)int
dateDate when the conversion event occurreddate
timestamp_convTimestamp of when the conversion occurred (ISO 8601 UTC)timestamp
conversion_hash_idA hash identifier for a unique conversion (based on conversion_key)str
conversion_keyUnique delivered conversion ID (e.g., your Order ID)str
actionName of the conversion eventstr
new_to_file1 if new customer (first-time conversion seen), else 0int
onsite_countTotal number of times the user appeared on your siteint
total_eventsNumber of marketing touchpoints before conversionint
event_idUnique identifier for each marketing touchpointstr
timestamp_eventsTimestamp of the marketing event (ISO 8601 UTC)timestamp
sequence_numberOrder of the touchpoint in the user journey (1 = earliest)int
hash_ip_eventsHashed IP of the user for that eventstr
typeReport type (e.g., platform_data, attribution)str
uidRockerbox user ID cookiestr
user_agent_eventsUser agent string from the marketing eventstr
base_idPrimary user identifierstr
currency_codeReporting currency for revenue and spendstr
fx_rate_to_usdExchange rate to USD (1 if already USD)float
platformName of platform (always Rockerbox for this dataset)str
spend_keyPlatform spend identifier (usually Ad ID or composite key)str
matchesList of categorization rules matched (last one is used)dict
tier_1Five-level categorization tiers aligned to UI taxonomy (most broad)str each
tier_2Five-level categorization tiers aligned to UI taxonomystr each
tier_3Five-level categorization tiers aligned to UI taxonomystr each
tier_4Five-level categorization tiers aligned to UI taxonomystr each
tier_5Five-level categorization tiers aligned to UI taxonomy (most granular)str each
original_urlURL of the marketing touchpointstr
request_referrerReferrer of the page where the user came fromstr
utm_parametersDictionary of all UTM query parameters passed in the URLdict
url_parametersOther non-UTM URL parametersdict
tier_oneSource data fields for taxonomy_lookup categorizationstr each
tier_twoSource data fields for taxonomy_lookup categorizationstr each
tier_threeSource data fields for taxonomy_lookup categorizationstr each
tier_fourSource data fields for taxonomy_lookup categorizationstr each
tier_fiveSource data fields for taxonomy_lookup categorizationstr each
normalizedFractional attribution assigned by multi‑touch modelfloat
revenue_normalizedConversion revenue assigned by modelfloat
first_touch1 if first touchpoint, else 0int
revenue_first_touchFull revenue if it’s the first touch, else 0float
last_touch1 if last touchpoint, else 0int
revenue_last_touchFull revenue if it’s the last touch, else 0float
evenEqual fractional credit across touchpointsfloat
revenue_evenRevenue attributed under even weight distributionfloat
rb_sync_idRockerbox internal sync identifierint
updated_atTimestamp when the record was last updatedtimestamp
marketing_typeTouchpoint category (e.g., onsite, creative, mail, etc.)str

Marketing Event Type Reference

The marketing_type column indicates the nature of each tracked touchpoint:
Marketing TypeDefinition
onsiteA click drove a user to site where a marketing touchpoint was tracked
creativeA display impression (view-based touchpoint)
postlogA linear TV touchpoint from postlog spike analysis
mailA direct mail touchpoint via address matchback
facebook_clickIn-app Facebook click that leads to conversion across devices
facebook_viewFacebook view-based touchpoint (synthetically modeled)
ottOTT view-based touchpoint that leveages an IP address matchback (includes Reddit, modeled using OTT methodology)
externalTouchpoint via custom matching of external file to Rockerbox data
viewthrough_events_tiktokTikTok view-based touchpoint via deterministic matchback
tiktok_viewTikTok view-based touchpoint (extrapolation of determinsitic matchback to account for low user tracking opt-in rate)
adwords_viewView-based touchpoint for YouTube/Demand Gen campaigns (synthetically modeled)
conv_onlyA direct (unattributed) touchpoint