The following document outlines the different input and outputs utilized in Attribution AI.
Attribution AI works by analyzing the following datasets to calculate algorithmic scores:
You can add multiple datasets from different sources if each of the datasets shares the same identity type (namespace) such as an ECID. To learn more about adding multiple datasets, visit the Attribution AI user guide.
The Adobe Analytics source connector can take up to four weeks to backfill data. If you recently set up a connector, you should verify that the dataset has the minimum length of data required for Attribution AI. Please review the historical data section to verify you have enough data to calculate accurate algorithmic scores.
For more details on setting up the Consumer Experience Event (CEE) schema, please refer to the Intelligent Services data preparation guide. For more information on mapping Adobe Analytics data, visit the Analytics field mappings documentation.
Not all the columns in the Consumer Experience Event (CEE) schema are mandatory for Attribution AI.
The following 9 columns are mandatory, additional columns are optional but recommended/necessary if you want to use the same data for other Adobe solutions such as Customer AI and Journey AI.
|Mandatory columns||Needed for|
|Primary Identity Field||Touchpoint / Conversion|
|Timestamp||Touchpoint / Conversion|
Typically, attribution is run on conversion columns such as order, purchases, and checkouts under “commerce”. The columns for “channel” and “marketing” are used to define touchpoints for Attribution AI (for example,
channel._type = 'https://ns.adobe.com/xdm/channel-types/email'). For optimal results and insights, it is highly recommended that you include as many conversion and touchpoint columns as possible. Additionally, you are not limited to just the above columns. You can include any other recommended or custom columns as a conversion or touchpoint definition.
If you are using Adobe Analytics data in your CEE schema, the touchpoint information for Analytics is typically stored in
channel.typeAtSource (for example,
channel.typeAtSource = 'email').
The columns below are not required but it is recommended that you include them in your CEE schema if you have the information available.
Additional recommended columns:
The minimum amount of data that is needed for Attribution AI to function is as follows:
Attribution AI requires historical data as input for model training. The data duration required is mainly determined by two key factors: training window and look-back window. Input with shorter training windows are more sensitive to recent trends, while longer training windows help produce more stable and accurate models. It’s important to model the objective with historical data that best represents your business goals.
The training window configuration filters conversion events set to be included for model training based on occurrence time. Currently, the minimum training window is 1 quarter (90 days). The lookback window provides a time frame indicating how many days prior to the conversion event touchpoints related to this conversion event should be included. These two concepts together determine the amount of input data (measured by days) that is required for an application.
By default, Attribution AI defines the training window as the most recent 2 quarters (6 months) and lookback window as 56 days. In other words, the model will take into consideration all of the defined conversion event(s) that have occurred in the past 2 quarters and look for all the touchpoints that have occurred within 56 days prior to the associated conversion event(s).
Minimum length of data required = training window + lookback window
The minimum length of data required for an application with default configurations is: 2 quarters (180 days) + 56 days = 236 days.
Attribution AI outputs the following:
Example output schema:
Attribution AI outputs attribution scores in the most granular level possible so that you can slice and dice the scores by any score column. To view these scores in the UI, read the section on viewing raw score paths. To download the scores using the API visit the downloading scores in Attribution AI document.
You are able to see any desired reporting column from the input dataset in the score output dataset only if either of the following are true:
The following table outlines the schema fields in the raw scores example output:
|Column Name (DataType)||Nullable||Description|
|timestamp (DateTime)||False||The time when an conversion event or observation occurred.
|identityMap (Map)||True||identityMap of the user similar to the CEE XDM format.|
|eventType (String)||True||The primary event type for this time-series record.
Example: “Order”, “Purchase”, “Visit”
|eventMergeId (String)||True||An ID to correlate or merge multiple Experience Events together that are essentially the same event or should be merged. This is intended to be populated by the data producer prior to ingestion.
|_id (String)||False||A unique identifier for the time-series event.
|_tenantId (Object)||False||The top level object container corrisponding to your tentant ID.
|your_schema_name (Object)||False||Score row with conversion event all the touchpoint events associated with it and their metadatas.
Example: Attribution AI Scores - Model Name__2020
|segmentation (String)||True||Conversion segment such as geo segmentation which the model is built against. In case of the absence of segments, segment is same as conversionName.
|conversionName (String)||True||Name of the conversion that was configured during setup.
Example: Order, Lead, Visit
|conversion (Object)||False||Conversion metadata columns.|
|dataSource (String)||True||Globally unique identification of a data source.
Example: Adobe Analytics
|eventSource (String)||True||The source when the actual event happened.
|eventType (String)||True||The primary event type for this time-series record.
|geo (String)||True||The geographic location where the conversion was delivered
|priceTotal (Double)||True||Revenue obtained through the conversion
|product (String)||True||The XDM identifier of the product itself.
Example: RX 1080 ti
|productType (String)||True||The display name for the product as presented to the user for this product view.
|quantity (Integer)||True||Quantity purchased during the conversion.
Example: 1 1080 ti
|receivedTimestamp (DateTime)||True||Received timestamp of the conversion.
|skuId (String)||True||Stock keeping unit (SKU), the unique identifier for a product defined by the vendor.
|timestamp (DateTime)||True||Timestamp of the conversion.
|passThrough (Object)||True||Additional Score dataset Columns specified by user while configuring the model.|
|commerce_order_purchaseCity (String)||True||Additional Score dataset Column.
Example: city: San Jose
|customerProfile (Object)||False||Identity details of the user used to build the model.|
|identity (Object)||False||Contains the details of the user used to build the model such as
|id (String)||True||Identity ID of the user such as cookie ID or AAID or MCID etc.
|namespace (String)||True||Identity namespace used to build the paths and thereby the model.
|touchpointsDetail (Object Array)||True||The list of touchpoint details leading to the conversion ordered by|
|touchpointName (String)||True||Name of the touchpoint that was configured during setup.
|scores (Object)||True||Touchpoint contribution to this conversion as score. For more information on the scores produced within this object, see the aggregated attribution scores section.|
|touchPoint (Object)||True||Touchpoint Metadata. For more information on the scores produced within this object, see the aggregated scores section.|
You can view the path to your raw scores in the UI. Start by selecting Schemas in the Platform UI then search for and select your attribution AI scores schema from within the Browse tab.
Next, select a field within the Structure window of the UI, the Field properties tab opens. Within Field properties is the path field that maps to your raw scores.
Aggregated scores can be downloaded in CSV format from the Platform UI if the date range is less than 30 days.
Attribution AI supports two categories of attribution scores, algorithmic and rule-based scores.
Attribution AI produces two different types of algorithmic scores, incremental and influenced. An influenced score is the fraction of the conversion that each marketing touchpoint is responsible for. An incremental score is the amount of marginal impact directly caused by the marketing touchpoint. The main difference between the incremental score and the influenced score is that the incremental score takes the baseline effect into account. It does not assume that a conversion is caused purely by the preceding marketing touchpoints.
Here is a quick look at an Attribution AI schema output example from the Adobe Experience Platform UI:
See the table below for more details about each of these attribution scores:
|Influenced (algorithmic)||Influenced score is the fraction of the conversion that each marketing touchpoint is responsible for.|
|Incremental (algorithmic)||Incremental score is the amount of marginal impact directly caused by a marketing touchpoint.|
|First Touch||Rule-based attribution score that assigns all credits to the initial touchpoint on a conversion path.|
|Last Touch||Rule-based attribution score that assigns all credit to the touchpoint closest to the conversion.|
|Linear||Rule-based attribution score that assigns equal credit to each touchpoint on a conversion path.|
|U-Shaped||Rule-based attribution score that assigns 40% of the credit to the first touchpoint and 40% of the credit to the last touchpoint, with the other touchpoints splitting the remaining 20% equally.|
|Time Decay||Rule-based attribution score where touchpoints closer to the conversion receive more credit than touchpoints that are farther away in time from the conversion.|
Raw Score reference (attribution scores)
The table below maps the attribution scores to the raw scores. If you wish to download your raw scores, visit the downloading scores in Attribution AI documentation.
|Attribution scores||Raw score reference column|
Aggregated scores can be downloaded in CSV format from the Platform UI if the date range is less than 30 days. See the table below for more details about each of these aggregate columns.
|customerevents_date (DateTime)||User defined & fixed format||False||Customer Event Date in YYYY-MM-DD format.
|mediatouchpoints_date (DateTime)||User defined & fixed format||True||Media Touchpoint Date in YYYY-MM-DD format
|segment (String)||Calculated||False||Conversion Segment such as geo segmentation which the model is built against. In case of absence of segments, segment is same as conversion_scope.
|conversion_scope (String)||User defined||False||Name of the Conversion as configured by the user.
|touchpoint_scope (String)||User defined||True||Name of the Touchpoint as configured by the user
|product (String)||User defined||True||The XDM identifier of the product.
|product_type (String)||User defined||True||The display name for the product as presented to the user for this product view.
Example: gpus, laptops
|geo (String)||User defined||True||The geographic location where the conversion was delivered (placeContext.geo.countryCode)
|event_type (String)||User defined||True||The primary event type for this time-series record
Example: Paid Conversion
|media_type (String)||ENUM||False||Describes whether the media type is paid,owned or earned.
Example: PAID, OWNED
|campaign_group (String)||User defined||True||Name of the campaign group where multiple campaigns are grouped together like ‘50%_DISCOUNT’.
|campaign_name (String)||User defined||True||Name of the campaign used to identify marketing campaign like ‘50%_DISCOUNT_USA’ or ‘50%_DISCOUNT_ASIA’.
Example: Thanksgiving Sale
Raw Score reference (aggregated)
The table below maps the aggregated scores to the raw scores. If you wish to download your raw scores, visit the downloading scores in Attribution AI documentation. To view the raw score paths from within the UI, visit the section on viewing raw score paths within this document.
|Column Name||Raw Score reference column|
Once you have prepared your data and have all your credentials and schemas in place, start by following the Attribution AI user guide. This guide walks you through creating an instance for Attribution AI.