# Feature Overview This feature uses the item's timestamp field (supports "Launch Time" or "Timestamp" attributes) to provide two time configuration modes: dynamic (up to now) and static (before/after). It automatically adjusts the ranking of recommended items in the results. * The dynamic mode boosts the ranking of new or recent content. * The static mode boosts or demotes items before or after a specific timestamp. It suits scenarios like e\-commerce new product promotion, news timeliness ranking, and short video trending recommendations where new or specific time content needs priority display. **This feature replaces the traditional method of manually setting fixed dates and frequently updating them.** **This document details how to configure and manage dynamic time\-based boost and demotion rules.** ## Core Value * **Save Time and Labor:** **Configure once, apply indefinitely.** Eliminate the need for repetitive manual adjustments every week or month to update promotion rules. * **Boost Revenue and Engagement:** Support the promotion of your latest products or content to capture user interest immediately, driving higher levels of participation and conversion. * **Optimize Inventory and Content Lifecycles:** Ensure that outdated or low\-relevance items are automatically moved out of top\-tier recommendation slots, maintaining a sense of **immediacy and freshness** in the user experience. ## Core principle Think of this feature as a **"Freshness Rule."** Every time a recommendation request is made, the system compares the specific timestamp of an item with the current time and applies a weight adjustment based on your pre\-configured rules. > For example: if an item was published 3 days ago and you set a rule to "boost items ≤7 days by 30%", the item's ranking score will increase by 30%, improving its rank. If the item does not meet the rule, its ranking remains unaffected. You can set multiple rules, each effective in its respective recommendation scenario. --- # Configuration steps ## Add new rule The system does not support configuring priority. When an item matches multiple configurations, the platform accumulates the boost and demotion settings from all configurations and applies the final weight factor. For example, if item A matches a "promote by 50%" rule and a "demote by 30%" rule, the platform will ultimately promote the item by 20%. :::tip \*Prerequisites Ensure that the item dataset used in the recommendation configuration includes a field with a "timestamp" or "Launch time" attribute. ::: 1. In **"Experience Center"—"Recommendation Experience"—"Strategy"** , click "Boost and Demotion" to expand the boost and demotion configuration section.

2. On the rule configuration page, please refer to the table below to complete all settings including "Rule Name," "Scope", and "Promotion/Demotion Configuration."

dynamic (up to now) time configuration

![图片](https://p9-arcosite.byteimg.com/obj/tos-cn-i-goo7wpa0wc/3ee81e37ccf940d389271263376b9458 =450x)

static (earlier than, later than) time configuration

|**Configuration item** |**description** | |---|---| |Rule name |set a simple and clear name | |Scope |select a time attribute field | |Configuration logic |Supports different configurations based on the attributes of the activation field. | |^^| **“Launch Time” field:** supports dynamic (up to now) time configuration|\ | ||\ | |* Dynamic (up to now) time configuration|\ | | * Time granularity: minutes, hours, days|\ | | * Comparison operators: within, outside | |^^| **“Timestamp” field**: supports dynamic (up to now) and static (earlier than, later than) time configurations|\ | ||\ | |* Dynamic (up to now) time configuration|\ | | * Time granularity: minutes, hours, days|\ | | * Comparison operators: within, outside|\ | |* static (before, after) time configuration|\ | | * time granularity: timestamp, second level|\ | | * comparison operators: before, after | |Boost / Demotion configuration |Supports configuring integer weights from \-100% to 100%; the larger the value, the greater the escalation or de\-escalation effect. | 3. Click "OK" to create the new rule on the platform. The new boost and demotion rules will take effect on the experience interface in the next recommendation request. 4. Note that previous operations only take effect in the "Experience Center" for preview and testing purposes. After confirming the effect is correct, click the "Sync Changes to API" button to officially apply this rule change to the online service. ## View existing rules The platform displays the promotion and demotion rules active within this recommendation scenario; you can view detailed configuration information for each rule. Follow these steps to view the active promotion and demotion rules: 1. Under **"Experience Center" — "Recommendation Experience" — "Strategy" — "Boost and Demotion"** , you can view the boost and demotion configuration rules.

2. The system directly shows the core configuration of each rule in the rule list. To see more details, click the rule name or the corresponding "Edit" button (if available).

## Edit existing rules Follow these steps to edit promotion and demotion configurations: 1. Under **"Experience Center" — "Recommendation Experience" — "Strategy" — "Boost and Demotion"** , you can view the boost and demotion configuration rules.

2. In the rules list, find the rule you want to modify, then click its corresponding "Edit" button.

3. After clicking edit, you can change the existing rule configuration on the interface. On the rule editing page, modify the following information as needed:

dynamic (to date) time configuration

![图片](https://p9-arcosite.byteimg.com/obj/tos-cn-i-goo7wpa0wc/3ee81e37ccf940d389271263376b9458 =450x)

static (before, after) time configuration

|**configuration item** |**description** | |---|---| |Rule name |Set a simple and clear name. | |Scope |Select a time attribute field. | |Configure logic |Support different configurations based on the attribute of the effective field. | |^^| **“New Release Time” field:** supports dynamic (up to now) time configuration|\ | ||\ | |* Dynamic (up to now) time configuration|\ | | * Time granularity: minute, hour, day|\ | | * Comparison operators: within, outside | |^^| **“Timestamp” field**: supports dynamic (up to now) and static (before, after) time configuration|\ | ||\ | |* dynamic (up to now) time configuration|\ | | * time granularity: minutes, hours, days|\ | | * comparison operators: within, outside|\ | |* static (before, after) time configuration|\ | | * time granularity: timestamp, second level|\ | | * comparison operators: before, after | |Boost / Demotion configuration |Supports configuring an integer from \-100% to 100% as the weight factor; the larger the value, the more significant the priority adjustment effect. | 4. Click "OK" to update the rule configuration on the platform. The updated boosting and downgrading rules will take effect in the experience interface during the next recommendation request. 5. Note that previous operations only affect the "Experience Interface" and are used for preview and testing. After confirming the effect is correct, click the "Sync Changes to API" button to officially apply the rule changes to the live service. ## Delete rule Follow these steps to delete boosting and downgrading rules: 1. In **"Experience Center" — "Recommendation Experience" — "Strategy" — "Boosting and Demotion"** , you can view the boosting and demotion configuration rules.

2. Click the "Delete" button on the target rule, then click "OK" again in the pop\-up confirmation dialog.

3. After confirmation, the rule will be removed from the experience interface and will take effect in the next recommendation request. Once the rule is deleted and synchronized, it cannot be restored. 5. Note that previous operations only take effect in the experience interface for preview and testing. After confirming the effect is correct, click the "Sync Changes to API" button to officially apply the rule changes to the online service. --- # Frequently Asked Questions **Q: Why is there a difference between the recommendation results in the experience interface and online?** A: The experience interface shows preview results only; actual online user recommendations may vary due to personalized recommendations. **Q: Why don’t I notice a difference after configuring the rule?** A: If the expected effect is not observed, please check: 1. Does the "Scope" in the rules meet the business expectations? 2. Are the weight factors set reasonably (the larger the value, the more significant the effect)? **Q: Why is there no dynamic time configuration after I selected the time field?** A: Please check the item dataset to confirm whether the field is configured with the "Launch time" or "timestamp" attribute. If you need to update the field attribute configuration, please contact the Volcengine technical team. **Q: Why does my configuration no longer support editing?** A: Historical boost and demotion rules based on the "Launch time" or "timestamp" attributes are incompatible with new capabilities, so editing is not supported and only deletion is allowed. The rule will continue to take effect before the configuration is deleted. --- # Related documentation 1. For other boost and demotion capabilities, refer to [Item Attribute Boost and Demotion](/docs/85296/1925098). 2. If you want to adjust item weights based on the distance between users and items, please refer to [User geographic location-based ranking boost and suppression](/docs/85296/1963520)

Viking AI Search