Analyze LinkedIn Performance Metrics

LinkedIn Analytics via Publora pulls detailed performance data for LinkedIn posts and profiles through the Publora MCP server, and can also act on posts - reacting and commenting - rather than just reading about them. linkedin_post_stats returns impressions, unique reach (members reached), reactions, comments, shares, and engagement rate for a specific post; linkedin_account_stats aggregates the same kind of metrics across the whole account (daily or total); linkedin_followers reports follower count or growth over a date range; and linkedin_profile_summary combines followers and stats into one overview.

Use it for a weekly or ad hoc performance review (which posts did best, engagement trends, follower growth and recommendations), for finding patterns across several recent posts, or for engaging directly with a colleague's or prospect's post via a reaction or comment. Don't expect real-time numbers right after publishing: LinkedIn's analytics can take up to 24 hours to fully populate, so querying immediately after posting returns partial data, and responses may be served from cache (indicated by a cached field).

Setup: create a free Publora account, connect LinkedIn via OAuth in the dashboard, get an API key, and configure an MCP server entry pointing at https://mcp.publora.com with Authorization: Bearer YOUR_API_KEY (LinkedIn access works on the free Starter plan). Post-level and account-level stat calls need a platformId (format linkedin-{id}) and, for post stats, a postedId - LinkedIn URNs from post URLs look like urn:li:activity:xxx, but the API needs urn:li:share:xxx or urn:li:ugcPost:xxx instead, best obtained from Publora's own get_post response. Engagement tools cover the full loop: linkedin_create_reaction (six types - LIKE, PRAISE, EMPATHY, INTEREST, APPRECIATION, ENTERTAINMENT), linkedin_delete_reaction, linkedin_create_comment (up to 1,250 characters, with optional nested replies via parentComment), and linkedin_delete_comment.

If the MCP server is unavailable, the same calls work over REST at https://api.publora.com/api/v1, authenticated with an x-publora-key header instead of a bearer token. The skill documents rough engagement-rate benchmarks by follower count (3-5% under 5K followers, 2-3% at 5K-50K, 1-2% above 50K) to help interpret whether a post's numbers are actually good, and notes LinkedIn's own rate limit of roughly 200+ API calls per hour, requiring backoff on 429 responses.

Creators, marketers, and sales professionals who want an AI assistant to report on LinkedIn post and profile performance in plain language, and to react to or comment on specific posts, without opening LinkedIn's own analytics dashboard.