Instagram Brand Mentions & UGC Monitor

Turn public tagged posts, campaign hashtags, and brand keyword results into one deduplicated feed of newly observed Instagram content, with match evidence and first-seen timestamps ready for review or automation.

What this Actor delivers

Each bounded scan can check one public brand account's tagged posts and Reels, up to five keyword phrases, and up to five recent campaign hashtag feeds. The Actor joins observations by stable Instagram media ID, preserves every configured source that found an item, and remembers what it has seen for the selected monitor.

Scheduled runs return only newly observed content when you reuse the same monitorId. A different monitor ID starts a separate state history.

Input example

{
  "monitorId": "patagonia-fall-2026",
  "brandAccount": "patagonia",
  "keywords": ["patagonia", "worn wear"],
  "hashtags": ["patagonia", "wornwear"],
  "firstRunBehavior": "emit_current",
  "keywordMatchPolicy": "include_configured_matches",
  "maxPagesPerInput": 1,
  "maxItemsPerInput": 50,
  "maxItemsPerRun": 250
}

Add at least one account, keyword, or hashtag. The maximum is one account, five keywords, and five hashtags.

First and recurring runs

• emit_current returns currently observed posts and saves the baseline.
• baseline_only saves the current baseline without emitting content rows.

State keys combine the authenticated run user and monitor ID. Media IDs unseen for more than 180 days expire from the monitor state.

Concurrent runs for the same user and monitorId do not scan in parallel: one run holds a Request Queue lock and another returns a busy outcome. A platform lock expires after a crashed run so later recovery remains possible.

Match confidence

Every delivered row keeps configured-match evidence. It also reports whether the configured term was literally confirmed in the caption, structured hashtag, or structured @mention.

Instagram keyword search can surface related content without the exact phrase. Keep those labeled results with include_configured_matches, or select literal_matches_only for a narrower feed. Tagged-account and hashtag evidence is preserved even when the caption does not repeat the term.

Output

The default Dataset contains one content row for each newly delivered, deduplicated media ID. Rows include:

• stable media, row, deduplication, monitor, and scan IDs;
• the public Instagram post or Reel URL and public author metadata;
• caption, posting and observation times, and first-seen time;
• engagement values when available;
• public media URLs and dimensions when available;
• normalized hashtags and mentions;
• ordered configured-source matches and literal-confirmation locations.

Unavailable fields are null; missing metrics are not converted to zero.

Every run also writes RUN_SUMMARY to the default key-value store. It reports completed, unavailable, and unscanned inputs; pages and retries; observed, filtered, known, and new item counts; state status; limits; warnings; and billed events. The Dataset contains content rows only.

Dataset delivery is at least once across process crashes. If a process stops after appending rows but before the final checkpoint, retrying the same run can append those rows again. Consumers should deduplicate with dedupeKey or mediaId. Billing recovery compares authoritative charged-event counts for the current run before retrying an event; live-cloud recovery remains part of the private smoke gate.

Limits and failure handling

• One brand account.
• Up to five keyword phrases.
• Up to five hashtags.
• One to three pages per monitored input.
• Up to 100 observed items per input.
• Up to 500 observed items per run.

A successful scan with no new content is distinct from a retrieval failure. If every source is unavailable before usable state can be committed, the run fails and no completed-monitor-scan event is charged.

Free users are limited to three monitored pages per run, one page per input, and 30 completed scans per calendar month. Paid tiers receive the full published limits.

Pricing

Customers pay for bounded monitoring coverage, not for new-item count.

Plan	Completed monitor scan	Additional monitored page
Free	$0.060	$0.007, unreachable under the current Free page cap
Bronze	$0.050	$0.006
Silver	$0.045	$0.0055
Gold	$0.040	$0.005
Platinum	$0.040	$0.005
Diamond	$0.040	$0.005

One completed-monitor-scan event includes the first three completed monitored pages. Each completed page after the first three uses an additional-monitored-page event. Retries and returned items are not billed separately. The automatic $0.00005 Actor-start event applies to every plan.

These are the coordinator-selected launch-candidate prices and are not live until the Actor is created and published.

Scheduling

Create an Apify Task with a stable monitorId, then add an Apify Schedule. Use webhooks or integrations to send the new-item Dataset and RUN_SUMMARY to your destination. The Actor does not create or manage schedules itself.

Coverage and responsible use

This Actor observes bounded public Instagram results. It is not an exhaustive social-listening index and does not promise every mention, complete historical backfill, deterministic keyword relevance, comments, Stories, deleted content, or private-account access. Media URLs can expire.

Instagram content can contain personal data and copyrighted material. Use results only for a lawful purpose, follow applicable rules, and do not use this Actor to access private content or make automated legal, employment, credit, or eligibility decisions about individuals.