Product updates
New APIs, platform coverage and fixes across Phyllo's Social Data and Social Screening products, and what each change means for you.
Highlights
Select a card to open the full update.
Each request now returns up to 7,000 contents on TikTok, 5,000 on Reddit and 20,000 for Instagram keyword listening.
Creator Search and Quick Search accept niche category filters, so you can pick a vertical before applying audience or engagement filters.
API reference and product docs for every product now live in the InsightIQ Docs portal.
Facebook works on the synchronous Content Fetch and Comment Fetch APIs without a separate feature being enabled.
Reddit published_at timestamps in the Content Comment API no longer include a trailing Z, matching the format used elsewhere in the API.
Quick Search keyword filters return consistent results, REPOST content paginates correctly and TikTok profile URLs no longer fail in Content Fetch.
Snapchat and RedNote are now supported in the Profile Analytics and Content Fetch APIs, returned on the same schema.
Retrieve what is said in a video, not just how it performed, across TikTok, YouTube and Instagram.
A dedicated endpoint listens to individual posts and videos by URL, alongside keyword and handle listening.
Browse the categories used to classify creators, then use them to discover creators or build category pickers.
LinkedIn and YouTube now run on an upgraded data pipeline, with more reliable pagination in Comments Fetch and Content Fetch.
LinkedIn profile URLs with special characters process correctly, and deleted YouTube Shorts no longer appear in the Public Content API.
See and search Brand Safety and background verification requests by Job ID. Instagram reposts are now screened end to end.
Set minimum and maximum engagement rates in Creator Search. Content Fetch also supports TikTok Stories.
Fixes across the Audience Demographic, Audience Overlap, Estimated Metrics and Risk Details APIs.
Screening includes LinkedIn follower and following counts, and adverse media flags appear on the dashboard even when no other platform is flagged.
The Social Data APIs support the Model Context Protocol, so agents and MCP-compatible clients can query social data directly.
The Social Listening Filters API accepts a date filter, and jobs process faster on every supported platform.
Monitor and analyse Reddit conversations in the Social Listening APIs and the Public Comments Fetch API.
Send screening requests back for rework through the public Rework API.
Flagging criteria can now catch toxic language in text that appears inside images and videos.
Choose whether screening workflows run automatically or pause for manual review.
Upload up to 6,000 profiles in one go for screening.
Add tags to screening reports and filter by them.
Screening models return fewer false positives across supported platforms.
Personal information in background verification is now stored encrypted.
Instagram Profile Analytics now includes creator GMV.
Flag or unflag content in one call with the new Content Review endpoints.
Combine keywords with boolean operators in Social Listening queries.
HEIC images are now supported.
Archive
Everything we shipped before July, month by month.
Read all previous updates →The Rework API is now public, OCR language toxicity is available in the Flagging Criteria API, and screening workflows can run automatically or with manual review.
Read June updates →Bulk upload for up to 6,000 profiles, tags to organise and filter reports, more accurate screening with fewer false positives, and encrypted PII in background verification.
Read May updates →Content Review endpoints to flag and unflag in one call, boolean keywords in Social Listening, Creator GMV in Instagram Profile Analytics, and HEIC image support.
Read April updates →Every few weeks. Social Data and Social Screening ship on their own cycles, and every update is listed here with its publish date.
Most are available as soon as they ship. Some capabilities, such as continuous monitoring, are enabled for your account by our support team. Contact us if you do not see a feature you expect.
Changes that can affect how you parse responses are labelled Format change and highlighted at the top of this page. Subscribers are notified by email.
Public updates affect data available on any profile with no account connection. Consented updates affect data that is only available after a creator connects their account and approves what is shared.
All API reference and product documentation is in the InsightIQ Docs portal at docs.insightiq.ai.