Outbound Wiki

Account-level outbound attribution

Attributing outbound impact across multiple contacts and touches within the same target account or buying group.

Account-level outbound attribution starts with a join, before scoring. First decide which contacts belong to the account, roll up their touches, and apply a credit rule a rep can explain. An account may show meaningful activity across several people, and no single person may represent the whole buying group. A contact's channel history still matters, so keep each signal tied to the person who created it and to the time it appeared. The account view should answer which touch opened the account and which touch most recently moved it.

Stage What you are trying to learn Example question
Resolve identity which contacts should count in the account rollup Which contact records belong to this account?
Capture touches the source, timing, and action for each contact touch What source and action should this touch retain?
Roll up activity which contact events qualify as account activity Which user events and account traits should qualify?
Assign credit which touch gets origin, latest, or influence credit What should a reader conclude from each credited touch?

Resolve the account

Set the account boundary first. Source performance only means something once you know who belongs in the rollup.

Significant activity by a contact assigned to an account can be counted at that account.1 Create the contact-to-account map before judging outbound impact. Record the account identifier, the contact identifier, and the rule that put the contact in the account.

Check whether the contact belongs to the target account, whether a parent, subsidiary, or buying group should share one account record, and what would remove the contact from the account rollup. Credit a touch only when its account assignment is clear. Keep unresolved contacts outside account credit until the mapping is settled.

Capture touches

Keep the contact-level history intact as you build the account view. The rollup should make the work easier to read without erasing who produced each signal.

Attribution provides a breakdown of the channels or sources a contact came from.2 For every mapped contact, retain the source, touch timing, action, and contact identity. It is common to consider first and latest attribution when assigning a contact to a source.3

Use first touch to answer how the account entered the outbound motion. Use latest touch to answer which contact interaction came most recently before the account moved to the next state. Keep the other touches in the history so the account view shows the sequence without giving every event the same credit. A reviewer should be able to trace each account-level credit back to a contact, a source, and a touch.

Roll up activity

With the identity map and touch history ready, choose the signals that belong in the account view. Keep account traits, individual user behavior, and behavior shared across the account distinct.

Account-level audiences are for businesses that sell to other businesses.4 They return accounts matching a combination of account-level traits, user-level traits, and user events.5 Use account-level audience conditions when a list depends on account traits or on the traits and behaviors of individual users who belong to accounts.6

Collective behavior needs separate treatment. Creating a list from the combined behavior of all users requires an account-level computed or SQL trait, followed by an account-level audience that uses that trait as a condition.7

Keep these inputs visible in the account record. Show which account traits qualified the account, which individual user events contributed, and which collective rule was applied. A reviewer can then distinguish a qualification caused by one person's activity, a firm-level condition, or a pattern across users. The record should also explain why the account qualified and let the reviewer open the underlying contact touches.

Assign credit

Choose the credit outputs before comparing sources. One account total hides the difference between opening an account and moving an active account forward.

Use origin credit for the first qualifying touch across the contacts in the account. Use latest credit for the most recent qualifying touch before the account outcome you are measuring. Keep influence as a supporting view for the other touches, so the report shows participation without claiming that every touch opened or moved the account.

Attribution information can help identify growth opportunities and determine which channels offer the most potential.8 Review the report at the account level first, then inspect the contact sequence behind each credit. Look for repeated source patterns across accounts and check whether the same pattern appears in origin credit, latest credit, or both.

After review, accounts and their users can be synced with downstream destinations when the operating flow needs that handoff.9 Send only records that have passed the account mapping and credit checks.

What not to do

These mistakes make an account report look precise while hiding how the credit was formed.

  • Wait to count a contact's activity at an account until the contact has been assigned to that account; the account rollup depends on that relationship.1
  • Keep contact-level source history instead of replacing it with one account-level source; attribution breaks down the channels or sources a contact came from.2
  • Keep first and latest credit separate; those are common attribution views.3
  • Treat collective behavior as a separate account-level rule, using an account-level computed or SQL trait in an account-level audience.7

Sources

  1. 1
    “Any significant activity of a contact that is linked to an account can be then associated with that account.”
  2. 2
    “Attribution provides a breakdown of which channels or sources a contact came from.”
  3. 3
    “When attributing a contact to a specific source it is common to consider the first and lastest attribution.”
  4. 4
    “Account-level audiences are audiences for businesses that sell to other businesses.”
  5. 5
    “They return the set of accounts which match a combination of account-level traits, user-level traits, and user events.”
  6. 6
    “Create a list of accounts based on account-level traits, and/or the traits and behaviors of individual users associated with accounts 1. Account-level audiences Account-level audience conditions”
  7. 7
    “Create a list of accounts based on the collective behavior of all users associated with accounts 1. Account-level computed trait or SQL trait 2. Account-level audience (using computed trait as a condition) Using account-level computed and SQL traits as account-level audience conditions”
  8. 8
    “With this information, we will be able to identify growth opportunities and determine which channels offer the most potential.”
  9. 9
    “You can sync these accounts and associated users with downstream destinations.”