Does Braze Remember Between Runs?
Does Braze Remember Between Runs? No. A Braze run holds no memory of its own between runs. When a campaign sends or a Canvas step fires, it works from what is in front of it at that moment: the user profile, the trigger event, and the journey design. Anything an earlier run worked out survives only if that run wrote it back into a profile attribute, a custom event, or a segment that a later run can read. That point is easy to miss because a Braze workspace feels continuous. User profiles are s
Does Braze Remember Between Runs?
No. A Braze run holds no memory of its own between runs. When a campaign sends or a Canvas step fires, it works from what is in front of it at that moment: the user profile, the trigger event, and the journey design. Anything an earlier run worked out survives only if that run wrote it back into a profile attribute, a custom event, or a segment that a later run can read.
That point is easy to miss because a Braze workspace feels continuous. User profiles are still there tomorrow. Canvases stay live. Segments keep updating. Reports keep adding up. The continuity lives in those stored records, not in a memory that one run hands to the next.
What does a Braze user profile actually hold?
A Braze user profile is the center of the system. Each profile is tied to a user ID, called an external_id, and holds standard attributes such as name, email, city, and language. It also holds custom attributes that a team defines for its own needs, such as loyalty tier, favorite category, or account type. Messages can read those values through Liquid tags at send time.
The profile also holds behavior. Custom events record actions a user took, such as a purchase, a click, or a completed step in an app. Purchases, devices, subscription status, and engagement data sit on the same record. Braze states that data stored on the user profile, including custom event counts and first or last use, is kept for as long as the profile stays active.
Read that record closely and its job is clear. It answers present questions. What is this user's tier today? What did this user do last? Which segment does this user match now? It does not store what a run decided last week, or why it decided it.
Even the history view is short. The Event History tab on a profile lists custom events and purchases from the past 30 days, up to the 100 most recent events. That is enough to check recent behavior. It is not a record of past work.
Canvas: a journey design, not a memory
Canvas is Braze's visual journey builder. A Canvas holds its design: entry rules, steps, branches, delays, and exit rules. That design persists, and it should. A welcome journey built once can run for months.
What happens inside one journey is a different matter. Canvas Context steps let a journey create and update variables as a user moves through it. Braze is clear about the limit: this data exists only within that Canvas journey. It does not persist across different Canvases, and it does not change the user's profile for good. Each new entry redefines the context variables from the latest entry data. The total size of all variables in a step also has a cap of 50 KB.
Entry properties follow the same pattern. Data that arrives with the trigger event can shape the first steps of a journey. Connected Content can pull live data from an outside system at send time. Both are useful. Both are about the current run. Once the journey ends, the working notes from that journey end with it, unless a step wrote a final value back to the profile.
Campaigns, segments, and Currents: logs and lists, not reasoning
Segments group users by rules over profile and behavior data. A segment can show who matches a set of conditions right now. It cannot show why a group was built, which version of the group was used last month, or what the team planned to test next.
Campaign and Canvas reporting shows what was sent and how people responded. Opens, clicks, and conversions prove a message went out and how it performed. Messaging interaction data for a stopped campaign or Canvas also has a shelf life for retargeting: after three months it moves to long-term storage unless an active filter still uses it.
Currents streams event data out of Braze into a warehouse or another tool. That gives a team a raw feed of what happened. A feed is still a log. It records customer behavior, not operator reasoning. It cannot bring back which audience was almost used, which message variant was rejected, or what the next step was supposed to be.
An outside agent that builds segments, drafts messages, or reviews journey performance faces the same gap. The agent can read the current profile and the latest numbers. It cannot read its own earlier reasoning, because that reasoning was never stored in Braze. If the agent does not save its decision somewhere else at the time, the next run starts from current state and works the case out again.
That is the pattern across the whole platform. Braze remembers the customer. It does not remember the work.
What a shared memory layer adds
A shared memory layer sits outside any one tool and holds the full context of the work itself. Vilix AI is built for exactly that job. It is cloud-hosted, so there is zero infrastructure to run or maintain. The same memory is available everywhere over MCP, so a scheduled agent, a chat tool, and a coding tool all read the same shared state instead of separate copies.
Vilix AI stores full conversation history, not just extracted facts. That matters when the next run needs the reason, not only the result. The full thread shows what was checked, what was ruled out, and what was promised, in the order it happened. A custom attribute can hold a final value. Full history holds the path that led to it.
For a lifecycle workflow, that changes each run. A weekly run that reviews a win-back Canvas can save why each audience was kept or cut, which variant was retired, and what to watch next week. The next run reads that record first, checks only what changed in Braze, and continues instead of rebuilding the review from raw counts.
Plans and control
Vilix AI has a free plan forever, and a 7-day Pro trial with no credit card required. Data can be exported or deleted at any time, so the memory stays under the operator's control. Getting started is at Vilix AI.
So, does Braze remember between runs? Braze remembers user profiles, events, segments, and journey designs. It does not remember decisions, reasoning, or next steps between runs. Keep Braze as the system that holds the customer and sends the messages. Give the agents and automations around it one shared memory, and each run starts where the last run actually left off.