You are here:
Scheduled and Real-Time Matching in Identity Resolution
Identity resolution can perform two types of ruleset runs: scheduled or real-time. They differ in how and when the two types of ruleset runs are performed, and how they interact with other features.
Scheduled Matching
Scheduled matching reviews new and updated source data several times in a day using match methods you select. It stores updated unified profile information in the Unified Account or Unified Individual data model object.
Scheduled matching is used the first time your identity resolution ruleset runs. After the first successful run, your identity resolution ruleset is updated incrementally using scheduled matching when data in the primary data model object changes.
You can’t change or control the timing of scheduled matching, but you can turn it off by disabling scheduling.
More match methods are available when scheduled matches are run. Scheduled matching allows you to coordinate a thorough review of your data.
Real-Time Matching
Real-time matching matches information about an active end user to a unified profile so that you can personalize your app or website within milliseconds. Real-time matching is used whenever customer identity information is received from the real-time personalization process.
Rulesets are run using real-time matching when the ruleset's output object is the primary object for a real-time data graph. In real-time matching, match methods are restricted to Exact or Exact Normalized matching only, regardless of the scheduled match selected for a given match rule. Email addresses and phone numbers can be matched using exact normalized matching, while all other fields are matched using exact matching only.
Once per day, the real-time data graph is updated with unified profiles generated by a scheduled ruleset run. To ensure real-time matching returns swift results, the number of source profiles and engagement events included in a unified profile are limited. Review the Data 360 limits and guidelines for details. Source profiles are prioritized for inclusion based on their value in real-time matching. For example, profiles with recent activity, frequent activity, and with user authentication are prioritized over less-used profiles.
Real-time matching is fast, allowing identity resolution to match your active customer to a unified profile in milliseconds. If the customer identity information isn’t an exact match for an existing unified profile, no matching profile is returned. Data that isn’t matched using real-time matching is reviewed during a scheduled match and could be matched at that time.
Reconciliation Rules in Real-Time Matching
Identity resolution uses the same reconciliation rules in all modes, regardless of whether match rules are scheduled or real-time.
Identity Resolution Output for Real-Time Matches
When you’re using real-time matching, identity resolution compares the input value to unified profile records in the ruleset’s related real-time data graph. If an existing customer is identified, these two actions are triggered:
- Real-time insights based on the unified profile are calculated. Results are saved to the real-time data graph record.
- Real-time segments based on the unified profile are built. Segmentation is saved to the real-time data graph record.

