Release Summary 26.16 | Aug 06, 2026
The following key features and improvements, along with bug fixes, have been released in ADA Digital Experience Personalization (DXP) in the release version 26.16.
Social Proof
Enhancements to Social Proof Coverage Analysis Report
The Social Proof Coverage Analysis Report has been enhanced with full data export support and expanded coverage for aggregated metrics. Merchandisers can now access and download the complete combination of metrics, intervals, and thresholds for a given site in one place, instead of manually stitching together multiple views for trend or comparative analysis. The report also now supports sites that use aggregated metrics, where SKU-level data is combined based on a product attribute rather than tracked individually, a scenario that was previously stored separately and not reflected in standard coverage reporting.
The report adapts to each site's configuration: non-aggregated sites see coverage at the SKU level, aggregated sites see coverage rolled up by attribute, and sites using both see a combined rollup, alongside the option to export the full data set. Together, these improvements give merchandisers a complete, accurate, and easily exportable view of coverage regardless of how a site's metrics are structured.
Enterprise Dashboard
User Segments in Recommend Segment Report
The Recommend Segment Report now includes user segments as part of the segments filter, allowing merchandisers to view recommendation performance broken down by user segment rather than only by traditional segment types. This gives merchandisers a more complete view of how different user segments interact with recommendations, without needing to rely on separate reports or manual analysis.
By bringing user segments into the same filtering experience, merchandisers can more easily compare performance across segments and tailor their recommendation strategies to specific user groups, supporting more targeted merchandising decisions directly from the report.
Jira: ENG-32865
Boost Rules Performance Report
Analytics now includes a dedicated Boost Rules Performance Report, giving merchandisers visibility into how their boosting rules are impacting recommendation performance. Boosting rules re-rank products based on categories, brands, products, attributes, price, ratings and reviews, or user attributes, and this report now makes it possible to measure their impact and compare performance with a rule applied versus not applied. This brings boost rules the same visibility already available for Advanced Merchandising rules.
The report is available under the Recommend menu and also surfaced as a card on the landing page. Users can filter by date range, aggregate or de-aggregate by page type and placement, select all or specific boost rules, and apply currency and region filters. Performance can be viewed as a bar chart or line chart, with support for Clicks, Attributable Sales, Attributable Orders, and Attributable Items, and Attributable Sales set as the default.
Jira: ENG-32891
Other Feature Enhancements
The following feature enhancements and upgrades have been made in the release version 26.16.
|
Jira # |
Module/Title |
Summary |
General Availability |
|---|---|---|---|
|
Find: Boost and Bury Rules Now Apply to Child Categories |
Boost and bury rules set at the parent category level now automatically cascade to all child categories, allowing merchandisers to define common rules once at the root category instead of repeating them across each child category. The condition limit for boost and bury rules has also been raised, supporting merchandisers with more intricate category structures. Together, these changes simplify merchandising workflows through rule inheritance and remove key barriers for complex catalog hierarchies. |
06-Aug-26 |
|
|
Enterprise Dashboard: Enhanced Rules, Layouts, and Settings on Placement Profile Page |
The Placement Profile page has received broad styling and functional improvements across its rules and configuration tabs. The Merchandising, Filters, and Boost tabs have been restyled to match the Strategies tab, with Advanced and Manual Merchandising combined into a single Merchandising tab, and Do Not Recommend and Only Recommend combined into a single Filters tab, with rules grouped by label and color-coded icons per rule type. The Layouts & Settings tab has also been redesigned, showing placement settings in a read-only format with an Edit option, and layouts as cards with a 3-dot menu for editing or removing them. Additional fixes include corrected placement names and rec group counts on the Merchandising view, more detailed Filters and Boosts views, reliable Expand All and Collapse All buttons, resolved layout and settings errors, and improved page load performance. |
|
|
|
Recommend: Audit Log Tracking for Variation and Template Actions |
The audit log now correctly tracks object IDs for actions performed on experiences, variations, and templates. Previously, creating a variation did not record the corresponding variation ID, making it harder to trace the action back to the variation it affected. This has been addressed, improving traceability when auditing configuration history. |
06-Aug-26 |
|
|
Data Engineering: Boost Rule Names Added to Dimension Tables |
Boost rule IDs and names are now included in the dimension tables, enabling boost rule names to appear in performance reporting instead of relying on IDs alone. This makes it easier for merchandisers to identify and interpret boost rule performance directly within reports. |
06-Aug-26 |
|
|
Discover: |
New fields have been added to support browse event tracking in click and view events. This includes methodName and productBrowseClickData for click events, and methodName and productBrowseViewData for view events. These additions enable more detailed tracking of browse interactions, laying the groundwork for richer reporting and analysis of browse-based user behavior. |
06-Aug-26 |
|
|
Recommend: Privacy Mode Tracking in Avro Records |
Click and view records now capture the privm parameter as a privateRequestMode field, allowing sessions to be classified based on consent status. When a request includes privm=t, the record reflects privateRequestMode:true, indicating the session was recorded without consent, and when privm is absent or set to any other value, the record reflects privateRequestMode:false, indicating consent was given. This enables more accurate and complete capture of session data for reporting purposes. |
06-Aug-26 |
|
|
Find: New Flag to Enable LLM Query Intent |
A new configuration flag, llmIntentEnabled, is now available in the tagging configuration alongside the existing intent settings. When enabled, this flag allows the catalog tagger's Solr call to retrieve LLM-based query intent, giving platform teams more flexibility in how query intent is determined during tagging. |
06-Aug-26 |
|
|
Enterprise Dashboard: Enhanced Performance View on Placement Profile Page |
The Performance tab on the Placement Profile page has been redesigned with a more interactive view. A new date range toggle (7, 14, or 30 Days, defaulting to 7 Days) updates KPI cards for Impressions, Click-Through Rate, Attributable Sales, and Revenue per 1K Impressions, each with a color-coded delta versus the previous period. A Strategy Performance section adds a toggle between Views & CTR and Attr Sales & RPMi, controlling both a bar chart and a synced table, with hover interactions that link the two for easier cross-referencing. These enhancements give merchandisers a clearer, more responsive view of placement performance without navigating away or reloading the page. |
06-Aug-26 |
|
|
Recommend: Glass Views Merchandising for Email Placements in Active Content |
Glass views merchandising, previously available for email placements through mailService, is now supported for email placements in Active Content as well. When an email placement generates recommendations, all products in the placement are automatically counted as a glass view and tracked when the site configuration is enabled. This allows the existing merchandising option that filters out products a user has already viewed in recommendations but not clicked to apply consistently across both mailService and Active Content once the relevant thresholds are met. |
06-Aug-26 |
|
|
Find: Automatic Static Configuration for Multi-Token Attributes |
When the multi-token flag is enabled for an attribute in the API, the associated static configuration is now applied automatically on the backend, rather than requiring users to pass it separately through the API or configure it in the complementary search JSON. This is particularly useful in Search Test Drive, where users can now rely on the backend to handle configuration automatically instead of setting it up manually each time. |
06-Aug-26 |
|
|
Find: Discover 2.0 Views and Clicks Pushed to Avro Logs |
Views and clicks captured in Discover 2.0 are now pushed to Avro logs, making them available in UDS_views and UDS_clicks. A new endpoint populates ProductBrowseViewData into UDS_views and ProductBrowseClickData into UDS_clicks, capturing method names for find/v1, find/v1/browse, and find/v1/content across both. This provides consistent visibility into Discover 2.0 browsing behavior alongside existing view and click data, supporting more complete downstream reporting and analysis. |
06-Aug-26 |
|
|
Platform: Job Status Reports Dashboard |
A new self-hosted dashboard is now available for exploring job execution history, making it easier to investigate issues like failed orders for a given site without needing direct database access or engineering involvement. The dashboard provides a paginated, filterable view of job status data by site, time range, application, and status, with direct lookup by job ID and detailed event information for each job. Column details can be tailored per service to surface relevant information such as order IDs and error messages, and the interface includes dark mode, date and time pickers, and load more pagination for easier investigation. |
06-Aug-26 |
|
|
Chatbot: Product Q&A Data Added to Zilliz Database |
Product question and answer attributes generated through LLM configuration are now pulled into the Zilliz database as dedicated columns, making this data available alongside other catalog attributes for downstream use. |
06-Aug-26 |
|
|
Find: Global Ranking Job Performance Optimization |
The Global Ranking job has been optimized to significantly reduce memory usage, improving reliability and stability when processing ranking data in production. The optimization also introduces the ability to skip enrichment fields based on site-level configuration, reducing unnecessary processing without changing ranking values or business output. This makes the ranking job more efficient and dependable, particularly for sites with larger data volumes. |
06-Aug-26 |
Bug and Support Fixes
The following issues have been fixed in the release version 26.16.
|
Jira # |
Module/Title |
Summary |
General Availability |
|---|---|---|---|
|
Content Ordering Fix for Multiple Content Placements |
We have fixed an issue where content in placements using a multiple content layout did not consistently follow the expected order based on campaign priority and content score, sometimes resulting in random ordering. Content is now correctly ordered based on campaign priority first, with content scores properly influencing the order of content within a given rule. |
06-Aug-26 |
|
|
Find: Autocomplete Deduplication Fix for Duplicate Product Names |
We have fixed an issue where autocomplete deduplication kept the first product it encountered for terms with identical product names, regardless of sales performance, which could surface less relevant products instead of the best-selling one. Deduplication now uses the same aggregated score already used to order autocomplete results, ensuring the product with the highest score is retained when multiple products share the same term. |
06-Aug-26 |