On this page: Review the system, journey, audience, channel, and content limits of Adobe Journey Optimizer so you can plan deployments that scale without hitting failures.
Below you will find guardrails and limitations when using Adobe Journey Optimizer.
Entitlements, product limitations and performance guardrails are listed in Adobe Journey Optimizer product description page.
-
Guardrails for Real-time Customer Profile data and segmentation also apply to Adobe Journey Optimizer.
-
See also Guardrails for Data Ingestion in Real-time Customer Profile
System & Platform system-platform
Supported browsers browsers
Adobe Journey Optimizer interface is designed to work optimally in the latest version of Google Chrome. You might have trouble using certain features on older versions or other browsers.
Datasets guardrails datasets-guardrails
As of February 2025, a time-to-live (TTL) guardrail is rolled out to Journey Optimizer system-generated datasets in new sandboxes and new organizations as follows:
- 90 days for data in the profile store
- 13 months for data in the data lake
This change will be rolled out to existing customer sandboxes in a subsequent phase. Learn more about datasets Time-To-Live (TTL) guardrails
Journeys journeys-guardrails
This section covers guardrails and limitations for journeys, including general journey limitations, journey components (actions, events, data sources), journey activities, and specific features like custom actions and expression editor.
General journey guardrails journeys-guardrails-journeys
-
The number of activities in a journey is limited to 50. The number of activities is displayed on the upper left section of the journey canvas.
As journeys near this limit, editing and publishing performance may degrade, and save or validation failures can occur. If this happens, split your journey into smaller sub-journeys using jump activities or recreate it in a new version. The activity limit cannot be increased.
-
By default, the number of live/paused/dry run journeys at one time is limited to 100. The current number of journeys is displayed above the journey canvas.
As you publish journeys, we automatically scale and adjust to ensure maximum throughput and stability. As you near the milestone of 100 live journeys at one time, you will see a notification appear in the UI on this achievement. If you see this notification and have a need to extend your journeys beyond 100 live journeys at a time, please create a ticket for customer care and we will help you reach your goals.
-
When using an audience qualification in a journey, that audience qualification activity may take up to 10 minutes to be active and listen to profiles entering or exiting the audience.
-
A journey instance for a profile has a maximum size of 1 MB. All data gathered as part of the journey execution is stored in that journey instance. Therefore, data from an incoming event, profile information retrieved from Adobe Experience Platform, custom action responses, etc. are stored in that journey instance and impact the journey size. It is advised, when a journey starts with an event, to limit the maximum size of that event payload (e.g., below 800 KB) to avoid reaching that limit after a few activities, in the journey execution. When that limit is reached, the profile is in error status and will be excluded from the journey.
-
For each profile and journey version, the journey runtime keeps an internal queue of up to 10 pending events while one is being processed. If this limit is reached, additional events are discarded with the
maxInstanceStackEventsReachedreason until the stack drains. See Events discarded due to a blocked journey instance. -
In addition to the timeout used in journey activities, there is also a global journey timeout which is not displayed in the interface and cannot be changed. This global timeout stops the progress of individuals in the journey 91 days after they enter. Read more
Journey payload size validation journey-payload-size
When you save or publish a journey, Journey Optimizer validates the total journey payload size to preserve stability and performance.
Default configuration
- Default maximum request size: 2 MB (2,000,000 bytes). Some organizations may have custom limits configured by Adobe.
- Warning threshold: 90% of the maximum limit.
- Error threshold: 100% of the maximum limit.
Troubleshooting and recommendations
- Review the largest node highlighted in the warning or error.
- Simplify conditions, reduce data mappings, and remove unnecessary steps or parameters.
- Consider splitting the journey into smaller journeys if needed.
- If you believe your organization needs a higher limit, contact your Adobe representative.
To monitor the current payload size of your journey before publishing, use the Current journey payload size indicator in the journey properties panel. Learn how to check the size of your journey payload
License package comparison select-package-limitations
For customers using the Select license package, the following additional limitations apply specifically to unitary journeys (journeys starting with an event or an audience qualification):
ERR_PKG_SELECT_8ERR_PKG_SELECT_7ERR_PKG_SELECT_6ERR_PKG_SELECT_2Journey versions journey-versions-g
The following guardrails apply to the Journey versions:
- A journey starting with an event activity in v1 cannot start with something else than an event in further versions. You cannot start a journey with a Audience Qualification event.
- A journey starting with a Audience Qualification activity in v1 must always start with a Audience Qualification in further versions.
- The audience and namespace chosen in Audience Qualification (first node) cannot be changed in new versions.
- The reentrance rule must be the same in all journey versions.
- A journey starting with a Read Audience cannot start with another event in next versions.
- You cannot create a new version of a read audience journey with incremental read. You must duplicate the journey.
Journeys and profile creation journeys-limitation-profile-creation
There is a delay associated to API based profile creation/update in Adobe Experience Platform. The Service Level Target (SLT) in terms of latency is < 1 min from ingestion to Unified Profile for 95th percentile of requests, at a volume of 20K Requests per second (RPS).
If a journey is triggered simultaneously to a profile creation and immediately checks/retrieves information from Profile Service, it might not work properly.
You can choose from one of these two solutions:
-
Add a wait activity after the first event, to give Adobe Experience Platform the time it needs to perform the ingestion to Profile Service.
-
Set up a journey that does not immediately leverage the profile. For example, if the journey is designed to confirm an account creation, the experience event could contain information needed to send the first confirmation message (first name, last name, email address, etc.).
Events events-g
The following guardrails apply to the Events in your journeys:
- Journey Optimizer supports a peak volume of 5,000 inbound journey events per second, across all sandboxes. Learn more about this limitation on this page.
- Event-triggered journeys may take up to 5 minutes to process the first action in the journey.
- For system-generated events, streaming data used to initiate a customer journey must be configured within Journey Optimizer first to get a unique orchestration ID. This orchestration ID must be appended to the streaming payload coming into Adobe Experience Platform. This limitation does not apply to rule-based events.
- Business events cannot be used in conjunction with unitary events or audience qualification activities.
- Unitary journeys (starting with an event or an audience qualification) include a guardrail that prevents journeys from being erroneously triggered multiple times for the same event. Profile reentrance is temporally blocked by default for 5 minutes. For instance, if an event triggers a journey at 12:01 for a specific profile and another one arrives at 12:03 (whether it is the same event or a different one triggering the same journey) that journey will not start again for this profile.
- Journey Optimizer requires events to be streamed to Data Collection Core Service (DCCS) to be able to trigger a journey. Events ingested in batch, events inserted via Query Service, or events from internal Journey Optimizer datasets (Message Feedback, Email Tracking, etc.) cannot be used to trigger a journey. For use cases where you cannot get streamed events, you must build an audience based on those events and use the Read Audience activity instead. Audience qualification can technically be used, but is not recommended as it can cause downstream challenges based on the actions used.
Data sources data-sources-g
The following guardrails apply to the Data Sources in your journeys:
- External data sources can be leveraged within a customer journey to lookup external data in real time. These sources must be usable via REST API, support JSON and be able to handle the volume of requests.
- Internal Adobe addresses (
.adobe.*) are not allowed in URLs and APIs.
General actions general-actions-g
The following guardrails apply to the Actions in your journeys:
- Three retries are systematically performed in case of an error. You cannot adjust the number of retries according to the error message received. Retries are performed for all HTTP errors except for HTTP 401, 403 and 404.
- The built-in Reaction event allows you to react to out-of-the-box actions. Learn more on this page. If you want to react to a message sent via a custom action, you must configure a dedicated event.
- You cannot place two actions in parallel, you must add them one after the other.
- A profile cannot be present multiple times in the same journey, at the same time, for all active versions of the journey. If reentrance is enabled, a profile can reenter a journey, but cannot do it until he fully exited that previous instance of the journey. Read more
Custom actions custom-actions-g
The following guardrails apply to the Custom Actions in your journeys:
- A capping limit of 300,000 calls over one minute is defined for all custom actions, per host and per sandbox. This cap is enforced as a sliding window per sandbox and per endpoint for endpoints with response times less than 0.75 seconds. For endpoints with response times greater than 0.75 seconds, a separate limit of 150,000 calls per 30 seconds (also a sliding window) applies.
- The custom action URL does not support dynamic parameters.
- POST, PUT and GET call methods are supported.
- The name of the query parameter or header must not start with “.” or “$”.
- IP addresses are not allowed in URLs. Use hostnames instead.
- Internal Adobe addresses (
.adobe.*) are not allowed in URLs and APIs. - Built-in custom actions cannot be removed.
- Custom actions support JSON format only when using request or response payloads. See this page.
- Any endpoint targeted by a custom action must support at least 200 TPS. Be cautious that a throttling configuration cannot go below 200 TPS. Depending on your expected throughput, having a high response time could impact the actual throughput.
Supplemental identifiers supplemental
Specific guardrails apply to the use of supplemental identifiers in journeys. They are listed in this page.
Expression editor expression-editor
The following guardrails apply to the journey expression editor:
- Experience event field groups can not be used in journeys starting with a Read audience, an Audience qualification or a business event activity. You must create a new audience and use an
inaudiencecondition in the journey. timeSeriesEventsattributes cannot be used in the expression editor. To access Experience Events at a profile level, please create a new field group based on aXDM ExperienceEventschema.
Journey activities activities
Audience Qualification activity audience-qualif-g
The following guardrail applies to the Audience Qualification journey activity:
- The Audience qualification activity cannot be used with Adobe Campaign activities.
- Supplemental identifiers are not supported for Audience qualification journeys.
Learn more about journey processing rates and throughput limits in this section.
Additional guardrails — including recommendations on streaming vs. batch audiences and composition audience limitations — are listed on this page.
Campaign activities ac-g
The following guardrails apply to the Campaign v7/v8 and the Campaign Standard activities:
- Adobe Campaign activities cannot be used with a Read audience, or an Audience qualification activity.
- Campaign Standard activities cannot be used with other channel activities: Card, Code-based Experience, Email, Push, SMS, In-app messages, Web.
- Campaign v7/v8 activities can be used alongside native channel activities in the same journey.
Reaction events reaction-events-g
Specific guardrails apply to Reaction events, including the requirement to place the activity immediately after a channel action and the inability to track messages sent in a different journey. They are listed on this page.
In-app activity in-app-activity-limitations
The following guardrails apply to the In-app message action. Learn more about In-app messages on this page.
-
This feature is currently not available for Healthcare customers.
-
Personalization can only contain profile attributes.
-
The In-app activity cannot be used with Campaign Standard activities.
-
In-app display is tied to the journey lifespan, meaning that when the journey ends for a profile, all In-app messages within that journey will cease to be displayed for that profile. Consequently, it is not possible to stop an In-app message directly from a journey activity. Instead, you must end the entire journey to stop the In-app messages from being displayed to the profile.
-
In test mode, the In-app display depends on the journey’s lifespan. To prevent the journey from ending too early during testing, adjust the Wait time value for your Wait activities.
-
Reaction activities can not be used to react to an In-app open or click.
-
An activation delay may happen between the moment a user profile reaches an In-app activity in the canvas and the time they start seeing that In-app message.
-
In-app message content size is limited to 2 MB. Including large images can hinder the publishing process.
Content decision activity content-decision-g
Specific guardrails apply to the Content decision activity, including a 48-hour delay before updated consent policies take effect in decision policies. They are listed on this page.
Jump activity jump-g
Specific guardrails apply to the Jump activity. They are listed on this page.
Read audience activity read-segment-g
The following guardrails apply to the Read Audience journey activity:
- Streamed audiences are always up-to-date but batch audiences will not be calculated at retrieval time. They are only evaluated every day at the daily batch evaluation time.
- At journey entry, profiles use attribute values from the batch audience snapshot. However, when a profile reaches a Wait activity, the journey automatically refreshes profile attributes by fetching the latest data from Unified Profile Service (UPS). This means profile attributes may change during journey execution.
- The Read Audience activity cannot be used with Adobe Campaign activities.
- The Read Audience activity can only be used as a first activity in a journey, or after a business event activity.
- A journey can only have one Read Audience activity.
- The Read Audience activity can target only one audience per journey. If multiple audiences are required, merge them into a single audience first. Learn how to combine audiences using composition workflows.
- Each organization can run up to five Read Audience instances concurrently (scheduled or business-event triggered), across all sandboxes and journeys. Avoid having more than five journeys with Read Audience starting at the exact same time; spread them 5 to 10 minutes apart. Learn more about journey processing rates in this section.
- Sandbox throughput: the system manages processing per sandbox with a maximum of 20,000 profiles per second shared across all Read Audience activities. Individual activities can be configured from 500 to 20,000 profiles per second. If sandbox limits are reached, jobs may be queued.
- Job processing timeout: Read Audience jobs that cannot be processed within 12 hours are automatically cleaned up and will not execute.
- Retries are applied by default on audience-triggered journeys while retrieving the export job. If an error occurs during export job creation, retries will be made every 10 minutes, for a maximum of 1 hour. After that, the journey is considered failed and can therefore be executed up to 1 hour after the scheduled time.
- For journeys using supplemental IDs, the reading rate of the Read Audience activity for each journey instance is limited to a maximum of 500 profiles per second.
See also recommendations and configuration for the Read Audience activity.
Update profile activity update-profile-g
Specific guardrails apply to the Update profile activity. They are listed on this page.
Journey Pause pause-g
Specific guardrails apply to pausing journeys, including a maximum pause duration of 14 days and a 10-million-profile cap across all paused journeys in your organization. They are listed on this page.
Journey Dry run dry-run-g
Specific guardrails apply to Journey Dry run, including counting toward engageable profile and live journey quotas. They are listed on this page.
Journey Fragments fragments-journey-g
Specific guardrails apply to Journey Fragments, including a maximum of 20 nodes per fragment and 200 active fragments per sandbox. They are listed on this page.
Send using waves waves-g
Specific guardrails apply to wave sending in journeys, including a 2–10 wave range and a 30-minute minimum interval between waves. They are listed on this page.
Journey simulation simulation-g
Specific guardrails apply to journey simulation. They are listed on this page.
Audiences & Profiles audiences-profiles
This section covers guardrails for audience management, profile handling, and engageable profile considerations.
Audience and profile guardrails audience
-
You can publish up to 10 audience compositions in a given sandbox. If you have reached this threshold, you need to delete a composition to free up space and publish a new one.
Learn more about audience compositions on this page.
-
When ingesting data, emails are case-sensitive. It means that duplicate profiles may be created (for example, one profile for John.Greene@luma.com, another profile for john.greene@luma.com) and used when targeting the corresponding recipient in your Journey Optimizer journeys and campaigns.
-
When targeting pseudonymous profiles (unauthenticated visitors) with inbound channels, consider setting a Time-To-Live (TTL) for automatic profile deletion to manage your engageable profile count and associated costs. Learn more
Channels & Messaging channel-guardrails
This section covers guardrails for all communication channels including email, SMS, inbound channels (web, in-app, code-based, content cards), and transactional messages.
Email guardrails message-guardrails
The following guardrails apply to the email channel:
- You cannot use the same sending domain to send out email messages from Adobe Journey Optimizer and from another product, such as Adobe Campaign or Adobe Marketo Engage for example.
When designing email messages, the system checks for key settings and displays alerts for warnings (recommendations and best practices) and errors (blocking issues that prevent testing or activation). Learn more about email alerts and validation requirements in this section.
Message content size for journey publication message-content-size
When publishing journeys that contain email messages, the total message content size must not exceed 2 MB after backend processing. During publication, the system automatically processes message content by patching links, images, and applying transformations, which increases the payload size beyond the authored content size.
Best practices to prevent publication failures:
- Keep authored email content under 1 MB
- Minimize the number of content variants
- Optimize and compress images before adding them to messages
- Remove unused assets and unnecessary HTML elements
- Test message size before publishing journeys to production
If journey publication fails due to content size, reduce your message content and republish the journey.
SMS guardrails sms-guardrails
The following guardrails apply to the SMS channel:
- Media files for MMS can be included through a supported URL. Please ensure that the media file is uploaded separately.
- Message feedback syncing is not currently available for MMS.
- Consent management operates at the SMS channel level for MMS.
Inbound channel guardrails inbound-guardrails
-
To use code-based experience actions in Journey Optimizer and deliver code content payload that can be used by your applications, follow the prerequisites detailed on this page.
-
To be able to access and author web pages in the Journey Optimizer user interface, follow the prerequisites listed on this page.
-
To send In-app messages in your journeys and campaigns with Journey Optimizer, follow the delivery prerequisites listed on this page.
-
For Adobe Journey Optimizer to correctly display content cards, you must configure the Adobe Experience Platform settings listed on this page.
-
Journey Optimizer supports a peak volume of 5,000 inbound requests per second. This guardrail applies to all inbound requests, which can originate from any of the Journey Optimizer supported inbound channels (web, In-app, code-based experiences, content cards).
-
Journey Optimizer supports a maximum of 500 active inbound actions at any moment in time. These inbound actions are counted if they are part of a live campaign or if they are a node used in a live journey. Once you reach this number, you need to deactivate older campaigns or journeys that are using inbound actions before being able to launch new ones.
Profile management with inbound channels profile-management-inbound
Journey Optimizer inbound channels can target pseudonymous profiles, meaning profiles that are not authenticated or not known yet because they have not been engaged before on other channels. This is the case for example when targeting all visitors or audiences based on temporary IDs like ECID.
This increases your total engageable profile count, which may have cost implications if the contractual number of engageable profiles you purchased is exceeded. License metrics for each package are listed on the Journey Optimizer Product Description page. You can check the number of engageable profiles in the license usage dashboard.
To keep your engageable profiles within reasonable limits, Adobe recommends setting a Time-To-Live (TTL) to automatically delete pseudonymous profiles from the Real-Time Customer Profile if they haven’t been seen or engaged within a specific time window. Adobe recommends setting the TTL value to 14 days to match the current Edge profile TTL.
Transactional message guardrails transactional-message-guardrails
Journey Optimizer supports a peak volume of 500 transactional messages per second in campaigns.
Content & Assets content-assets
This section covers guardrails for content creation and management, including landing pages, subdomains, and fragments.
AI Assistant guardrails ai-assistant-g
Guardrails and limitations for AI Assistant content generation — including supported channels (email, push, web, SMS) and personalization editor limitations — are listed on this page.
Landing pages guardrails lp-guardrails
The following guardrails apply to the landing pages:
- Only one Form component can be used in a single primary page.
- The Form component cannot be used in subpages.
- You cannot add a preheader to a landing page.
- You cannot select the Code your own option when designing a landing primary page.
Subdomains guardrails subdomain-guardrails
The guardrails and limitations applying to subdomain delegation in Journey Optimizer are detailed on this page.
Fragments guardrails fragments-guardrails
The following guardrails apply to the fragments:
- To create, edit, archive, and publish fragments you need the Manage library items and [Publish Fragment] permissions included in the Content Library Manager product profile. Learn more
- Visual fragments are only available for the Email channel.
- Expression fragments are not available for the In-app channel.
- Visual fragments cannot exceed 100 KB. Expression fragments cannot exceed 200 KB.
- To use a fragment in a journey or campaign, it must be in the Live status.
- Contextual attributes are not supported within fragments.
- Visual fragments are not cross-compatible between the Use Themes and Manual Styling modes. To be able to use a fragment in a content where you want to apply a theme, this fragment must be created in Use Themes mode. Learn more on themes
- When tracking is enabled in a journey or a campaign, if you add links to a fragment and if this fragment is used in a message, these links are tracked such as all other links included in the message. Learn more on links and tracking
Decision Management decision-management
Decisioning & Decision management guardrails decisioning-guardrails
Guardrails and limitations to keep in mind when working with Decisioning or Decision management are detailed in these the Decisioning & Decision management sections:
Campaign Orchestration campaign-orchestration
Campaign Orchestration guardrails orchestration-guardrails
Guardrails and limitations to keep in mind when working with Campaign Orchestration are detailed in this section: Guardrails & limitations.