在此頁面上：瞭解Journey Optimizer系統產生的資料集的存留時間限制，以便您可以規劃追蹤、回饋和歷程資料能保留多久的時間，並在資料過期之前保留重要資料。
自 2025 年 2 月起，將在 新沙箱和新組織 中，向 Journey Optimizer 系統產生的資料集，開放使用存留時間 (TTL) 護欄功能，如下所示：
- 輪廓存放區中的資料為 90 天
- 資料湖中的資料為 13 個月
此變更將從 2026 年 10 月 1 日 開始，強制在 現有客戶沙箱 中執行。
受影響的資料集 datasets
下表列出所有Journey Optimizer系統產生的資料集，以及資料湖和設定檔存放區中的各自存留時間。 時間序列資料集受TTL限制；列出記錄型別資料集以供參考，並在兩欄中標示n/a。 「可用性」欄會指出資料集是否預設包含，或需要特定的附加元件或授權。
常見問題 faq
您可以在下方找到資料集存留時間(TTL)的常見問題集。
需要更多詳細資料？ 使用此頁面底部的意見回饋選項來提出您的問題，或與 Adobe Journey Optimizer 社群聯絡。
n/a。目前不支援TTL副檔名。 不過，我們已計畫最佳化TTL流程，以便在2025年下半年開始的某個時候允許這些擴充請求。
| note |
|---|
| NOTE |
| 儲存在設定檔中的資料受限於「總資料量」權益。 因此，設定檔上因TTL延伸所增加的任何資料儲存都將計入「總資料量」權益。 了解更多 |
-
查詢存放區：否
-
歷程上限：否
-
優惠方案上限：否
-
傳送時間最佳化(STO)：否
-
訊息頻率上限 （亦即商務規則）：否
-
報告：否
note NOTE 已在Customer Journey Analytics (CJA)連線上實作TTL，這會將受影響資料集資料的有效回顧期間縮短為13個月。 -
Experience Platform資料來源：不適用 — 不支援透過資料來源擷取體驗事件。
-
計算屬性：是 — 初始回填計算將限製為過去90天的資料；計算屬性將根據後續更新的增量事件進行更新。 後續更新一旦達到回顧期間（最多6個月），TTL基本上就不會再影響運算屬性。 更多詳情。
-
分段和重新目標定位：是 — 分段取決於設定檔存放區中的資料；因此，受影響資料集資料的回顧時間限製為90天。
-
追蹤：是 — 將受影響資料集資料的有效回顧期間縮短至90天。 來自受影響資料集的資料會在Data Lake中存留13個月。
- 檢閱您的使用案例，並找出任何需要保留新TTL以外的資料的使用案例。
- 設定自動化查詢，以在刪除資料之前將關鍵資料複製到衍生的資料集。
- 與您的Adobe代表合作，討論任何額外的需求或潛在的TTL擴充功能（預計於未來版本推出）。
Journey Optimizer系統產生的資料集受到保護，且無法透過標準Adobe Experience Platform UI刪除。 這些資料集是Journey Optimizer功能的必要專案，由系統管理。
如果您需要永久移除Journey Optimizer系統資料集（例如，針對QA環境、沙箱清理或特定資料衛生要求），請聯絡Adobe工程或Adobe客戶服務。 這些資料集需要專門的後端程式，以確保完全安全地移除。
| note |
|---|
| NOTE |
| 若要在這些系統資料集中進行例行資料清理，請使用Privacy Service提供的 資料生命週期 作業來刪除特定記錄或身分。 了解更多 |
This section contains structured knowledge intended to support interpretation, retrieval, and question answering related to this topic.
For complete understanding, this information should be combined with the documentation on this page. Neither source is intended to stand alone; the page describes the feature, while this section provides additional context that helps disambiguate terminology, intent, applicability, and constraints.
- TL;DR: This page explains the time-to-live (TTL) retention guardrails on Journey Optimizer system-generated datasets in the profile store (90 days) and the data lake (13 months), including rollout timing and options for retaining data longer.
Intents:
- Understand how long system-generated dataset data is retained in the profile store versus the data lake.
- Identify which datasets are subject to TTL.
- Plan for the enforcement on existing customer sandboxes starting October 1, 2026.
- Retain data beyond the TTL using dataset export through Destinations or Data Distiller derived datasets.
- Understand the impact on segmentation, computed attributes, tracking, and Customer Journey Analytics.
Glossary:
- Time-to-live (TTL): retention guardrail after which system-generated dataset data is dropped (product-specific)
- Profile store: store subject to the 90-day TTL for impacted datasets (product-specific)
- Data lake: store subject to the 13-month TTL for impacted datasets (product-specific)
- Time-series dataset: dataset type that is subject to TTL
- Record-type dataset: dataset type not subject to TTL (marked
n/ain both TTL columns) - Data Distiller: entitlement that allows creating derived datasets stored in the data lake without a TTL (product-specific)
Guardrails:
- Profile store TTL: 90 days for Journey Optimizer system-generated dataset data (hard limit; TTL extensions are not currently supported).
- Data lake TTL: 13 months for Journey Optimizer system-generated dataset data (hard limit; TTL extensions are not currently supported).
- AJO Message Export Dataset and AJO Message Event Metadata Dataset: 30 days data lake TTL (hard limit; both require the Message Export add-on).
- TTL is enforced on new sandboxes and new organizations as of February 2025, and on existing customer sandboxes starting October 1, 2026 (hard enforcement date).
- TTL applies only to time-series datasets; record-type datasets are
n/ain both the Data Lake TTL and Profile Store TTL columns. - TTL enforcement uses the event timestamp, not the ingestion date.
- Journey Optimizer system-generated datasets are protected and cannot be deleted through the standard Adobe Experience Platform UI.
- Computed attributes: initial backfill calculation is limited to the last 90 days of data; subsequent look-back is a maximum of 6 months.
- Segmentation and retargeting look-back is limited to 90 days on affected profile store data.
Terminology:
- Canonical name: Time-to-live — Acronym: TTL
- Do not confuse: “profile store TTL” (90 days) ≠ “data lake TTL” (13 months)
- Do not confuse: “time-series dataset” (subject to TTL) ≠ “record-type dataset” (
n/a, not subject to TTL) - Do not confuse: dataset TTL (drops system-generated dataset data in the profile after 90 days) ≠ deletion of the profiles themselves (the profiles are not dropped)
FAQ:
- Q: Which datasets are subject to TTL? — Only time-series datasets; record-type datasets (such as entity and classification datasets) are not subject to TTL and are marked
n/a. - Q: Does the 90-day profile store TTL delete profiles? — No; the system-generated dataset data in the profile is dropped after 90 days, not the profiles themselves.
- Q: Can I increase the TTL? — TTL extensions are not currently supported; you can export data through Destinations, or, with a Data Distiller entitlement, create derived datasets stored without a TTL.
- Q: When does enforcement reach existing sandboxes? — Starting October 1, 2026.
- Q: Which timestamp is used for enforcement? — The event timestamp, not the ingestion date.
- Q: Can I delete Journey Optimizer system-generated datasets? — No; they are protected and cannot be deleted through the standard Adobe Experience Platform UI; contact Adobe Engineering or Adobe Customer Care for permanent removal.