April 2026 brings the most significant search capability in MarketScale's history, now live across your entire content library. Alongside search, this release delivers meaningful performance gains, tightened security, a cleaner content lifecycle, and a broad sweep of bug fixes and polish across the platform.

This release introduces platform-wide search, a new editor coaching data API, and expanded client visibility into content versions.

Your team can now search across every piece of content in your channel -- media, folders, transcripts, AI summaries, and more -- from a single interface. Results include an inline media player so your team can preview video directly from search without navigating away. Search matches on keywords, topics, speaker names, and even specific phrases spoken within a transcript, making it dramatically faster to surface the right content at the right moment.

A new read-only data endpoint now aggregates editor performance data -- including feedback, ratings, comments, QC results, and video scores by version -- into a structured feed that powers weekly coaching reports for editors. Channels using external coaching or performance review tools can now pull this data automatically without manual exports.

All content versions -- including sub-versions, editing iterations, failed takes, and approved cuts -- are now visible to clients at all times, not just inside the internal proofing room. This removes a common friction point where clients were unaware earlier versions existed and had to request them manually.

This release includes improvements to search filter accuracy, content lifecycle actions, loading performance, security hardening, dashboard speed, and recording room UI.

Several rough edges in the new search experience have been resolved in the same release cycle. Date filters now correctly restrict results to the selected range. Content type filters remain responsive after an initial selection without requiring a reset to "All types." Sidebar result counts now accurately reflect the active filter state instead of showing totals that contradict the main results area.

The actions available on content have been clarified and updated to better match how your team actually works. "Publish to Feed" now explicitly controls feed visibility -- using it manually approves content and makes it visible. "Mark as Complete" finalizes content without affecting feed visibility, so content remains accessible without being surfaced in the feed. These clearer labels reduce confusion about what each action does and prevent accidental feed changes.

Several improvements shipped this month that collectively reduce load times across the platform. The application's front-end assets are now split by page so each route only loads what it needs -- reducing the initial load time for all pages. Dashboard and approval queue data is now cached intelligently, meaning repeated loads pull from a fast local store instead of recomputing from scratch on every request. Heavily used data queries were restructured to eliminate redundant database calls, resulting in faster page rendering across content lists, feeds, and media views.

The platform was hardened against a class of injection-style security vulnerabilities found during an internal audit of raw database queries. All affected query patterns have been replaced with parameterized bindings. This change has no visible impact on day-to-day use but closes a meaningful security gap.

The proofing page received a first-pass visual update, replacing legacy UI components with the current design system. The update improves visual consistency and accessibility without changing any underlying functionality. Admins and contributors working in the proofing room will notice a cleaner, more consistent interface.

The recording room card has been updated to match the current design system. The redesigned card is visually cleaner and more consistent with the rest of the platform.

AI Writer now correctly detects when a transcript is ready, including in cases where the transcript was stored but not yet registered in the expected location. Previously, AI Writer would display "AI Writing will be available when transcript is ready" even when a transcript already existed, blocking content generation unnecessarily. This detection gap has been closed.

Toast notifications and loading indicators have been standardized across the platform. All toasts and loaders now use a single consistent style, replacing several legacy variants that had diverged over time. Notification titles no longer appear in all caps.

The following issues were resolved this month across search, dashboard, upload, recording, AI Writer, profiles, folders, the coaching feed, and the asset library.