Instagram Hashtag Scraper | $0.24/1K on Starter

Turn Instagram hashtags into structured posts and Reels for content research, trend discovery, campaign tracking, and social media intelligence. Results are relevance-ranked, as the Instagram app shows them. Paid runs support up to 10,000 items per hashtag, subject to availability, at about $0.24 per 1,000 items on Starter. No authentication or proxy setup.

What this Actor does

• Accepts one or more actual Instagram hashtags
• Collects public posts and Reels for each hashtag, relevance-ranked as the Instagram app shows them
• Returns up to 10,000 media rows per hashtag in each paid run, subject to availability
• Labels every row with media_type and product_type, so you can separate posts and Reels after export
• Supports resumable single-hashtag paid exports
• Works without Instagram credentials, cookies, session IDs, browsers, or proxy configuration
• Separates media rows, run status, continuation metadata, and aggregate reconciliation metrics

Use it for public content discovery, campaign monitoring, creator research, trend analysis, and repeatable data pipelines where an actual hashtag is the source boundary.

Quick start

1. Enter one or more hashtags. A single leading # is optional.
2. Set the maximum results for each hashtag.
3. Run the Actor, then preview or export the default dataset as JSON, CSV, or Excel.

{
  "hashtags": ["artificialintelligence"],
  "resultsLimit": 200
}

Input

Field	Rules
hashtags	Required list of actual hashtags. A single leading # is optional. Letters, numbers, and underscores are accepted after Unicode normalization. URLs, spaces, and other punctuation are rejected. The schema accepts up to 50 unique hashtags; plan-specific limits are detailed in Pricing.
resultsLimit	Required integer per hashtag from 25 through 10,000. Default and Store prefill: 200. Each hashtag returns up to this many posts and Reels, subject to availability. Plan-specific effective limits are detailed in Pricing.
continuationToken	Optional short-lived secret returned by a previous paid run. Requires the same single hashtag. It expires after 15 minutes. Every continuation run incurs a new source event.

This Actor searches actual hashtags only. It does not accept arbitrary keywords, profile URLs, direct post URLs, or date filters.

Dataset rows

Illustrative dataset item; optional facts appear only when observed:

{
  "hashtag_scrape": "artificialintelligence",
  "id": "1234567890123456789",
  "short_code": "ExampleCode",
  "url": "https://www.instagram.com/p/ExampleCode/",
  "caption": "Example public caption",
  "timestamp": "2026-08-25T10:00:00Z",
  "media_type": "image",
  "product_type": "feed",
  "like_count": 125,
  "comment_count": 4,
  "display_url": "https://...",
  "owner_id": "9876543210987654321",
  "owner_username": "example_creator",
  "owner_full_name": "Example Creator",
  "owner_is_verified": false,
  "owner_profile_pic_url": "https://..."
}

Field	Meaning
hashtag_scrape	Normalized source hashtag without #
id	Stable Instagram media ID represented as a string
short_code	Instagram post or reel shortcode
url	Public Instagram post or reel URL constructed from the shortcode
caption	Public caption text; empty string when absent
timestamp	UTC ISO-8601 publication timestamp
media_type	Normalized image, video, or carousel
product_type	Normalized Instagram product type, such as feed, clips, igtv, or carousel_container
like_count	Public like count; explicit null when hidden, omitted when unobserved; zero remains zero
comment_count	Public comment count when observed
play_count	Public play count when available
view_count	Public view count when available
child_media_count	Number of children when a child list was observed
display_url	Selected external image or thumbnail URL; it may expire
video_url	Selected external video URL when available; it may expire
owner_id	Public owner ID represented as a string when observed
owner_username	Current public owner username when observed
owner_full_name	Public owner display name when observed
owner_is_verified	Whether the owner is verified
owner_profile_pic_url	External owner image URL when available; it may expire

The same media can legitimately appear once for each requested hashtag because hashtag_scrape records customer-visible provenance. Duplicate media IDs are removed within each hashtag during a run.

Unknown facts are not synthesized as false, zero, or empty lists. Observed empty values remain explicit. Hidden like/view counts have explicit flags. Supplied locations, rich owner/coauthor profiles, user tags, full carousel children, accessibility captions, and audio metadata are retained alongside the fields above.

Pricing

Billing uses two transparent events:

1. one hashtag source event for each hashtag successfully validated and processed in a run, including a valid hashtag with no current results;
2. one hashtag media item event for each persisted dataset row.

The source event is $0.0015 per individual hashtag. Apify displays this normalized as $1.50 / 1,000 source events; it does not mean one hashtag costs $1.50. It covers starting and validating that hashtag extraction independently of how many rows are available.

Validation failures and missing or rejected hashtags do not incur a source or row event. If the run reaches its spending limit, it stops before storing an uncharged row.

Plan	Source event	Dataset item	One hashtag / 1,000 rows
Free	$0.0015	$0.00024	$0.2415
Starter	$0.0015	$0.00024	$0.2415
Scale	$0.0015	$0.00023	$0.2315
Business	$0.0015	$0.00022	$0.2215

Starter examples: one hashtag returning 25 rows costs $0.0075; 60 rows costs $0.0159; 100 rows costs $0.0255; 1,000 rows costs $0.2415. Multiple hashtags are priced independently. Every continuation run incurs another source event because it is a new source/run operation.

FREE demo limits

FREE Apify accounts can run this Actor up to 5 times per UTC calendar month. Each admitted FREE run accepts exactly one hashtag, returns up to 10 posts or reels, and cannot submit or receive a continuation token. Console, CLI, API, schedules, webhooks, MCP, and other Apify launch methods share the same account allowance. A run counts once admitted even when it is later aborted or unsuccessful, and resurrecting it does not start another extraction. Creating another task or API token does not reset the allowance. Paid Apify users are not subject to these demo limits.

Run summary

Run status, source charges, row counts, failures, continuation tokens, and non-sensitive FREE admission metadata are stored in the default key-value store's OUTPUT record rather than mixed into the dataset:

OUTPUT uses schemaVersion: 2: status is SUCCEEDED for completed work and graceful stops, or FAILED for internal faults (RUNNING denotes a checkpoint). Check outcome (COMPLETED, NO_RESULTS, PARTIAL, INPUT_REJECTED, POLICY_REJECTED, UPSTREAM_UNAVAILABLE, DEPENDENCY_UNAVAILABLE, LIMIT_REACHED, INTERNAL_ERROR) and message/issues; platform success does not guarantee data. Runtime input/policy rejections and external outages finish normally; confirmed rows remain available. Ambiguous writes or charges use reconciliation: "unknown". External reporting outages can leave summaries unavailable; platform-enforced aborts and resource failures remain outside this guarantee.

{
  "schemaVersion": 2,
  "outcome": "COMPLETED",
  "status": "SUCCEEDED",
  "stopReason": null,
  "isPaying": true,
  "requestedHashtags": 1,
  "resultsLimitPerHashtag": 200,
  "deliveredCount": 200,
  "chargedCount": 200,
  "sourceChargedCount": 1,
  "sources": [
    {
      "inputIndex": 0,
      "hashtag": "artificialintelligence",
      "status": "succeeded",
      "deliveredCount": 200,
      "chargedCount": 200,
      "sourceCharged": true
    }
  ],
  "continuations": [
    {
      "inputIndex": 0,
      "hashtag": "artificialintelligence",
      "hasNextPage": true,
      "nextContinuationToken": "<secret token>"
    }
  ]
}

Inspect each source entry even when the overall run completes. Aggregate non-PII request, retry, page, duplicate, source-charge, row-charge, storage-stop, and timing metrics are stored separately in AUDIT.

Admitted FREE runs include a freeTier object with the UTC month, claimed slot, remaining allowance, effective result limit, and policy version. Rejected FREE runs use stable reasons such as free_monthly_limit, free_run_already_consumed, or free_quota_unavailable. The record never contains an Apify user ID, API token, or internal quota-storage identifier.

Resumable paid exports

When hasNextPage is true, paying users can read nextContinuationToken from OUTPUT.continuations and start another run with the same single hashtag. Tokens expire after 15 minutes and are authenticated, product-specific, and hashtag-bound.

{
  "hashtags": ["artificialintelligence"],
  "resultsLimit": 1000,
  "continuationToken": "<token from OUTPUT>"
}

Always use the newest token. When the previous run stopped partway through a page, the continuation reads that page again, and Instagram may rank different media there. Media already delivered in the same token chain are skipped and never charged twice. A resumed run charges one new source event and charges only newly persisted rows.

Limitations

This Actor:

• Returns media in Instagram's relevance order, as the Instagram app shows them, not as a chronological feed or a complete historical archive; results can include media that are months or years old
• Returns posts and Reels together; most hashtags return mostly Reels
• Returns up to 10,000 media rows per hashtag in each paid run, subject to availability
• Does not offer arbitrary keyword search, profiles, direct post lookups, date filtering, comments, transcripts, analytics, or media downloads
• Does not bypass private-account or restricted-content access controls
• Cannot guarantee stable ordering, the same media across repeated runs, permanent CDN URLs, or that counts hidden by Instagram will be available
• May return fewer rows when Instagram shows few media for a hashtag, a source becomes unavailable, or a run reaches its spending limit
• Can return public media that later changes or is removed at its source

API and CLI use

Call the Actor through the standard Apify Actor Runs API. Read media from the run's default dataset and reconciliation metadata from the OUTPUT and AUDIT key-value-store records.

With the current Apify CLI:

apify actors call publicsignallabs/instagram-hashtag-scraper \
  --input '{"hashtags":["artificialintelligence"],"resultsLimit":200}' \
  --output-dataset

The same Actor can be called with Apify's JavaScript or Python clients, REST API, schedules, webhooks, MCP, and platform integrations. Plan-specific demo limits are detailed in Pricing.

Responsible use

Collect and use public data only where you have a lawful purpose. Respect privacy, platform rules, and applicable data-protection law. Minimize retained personal data, secure exports, and delete records when they are no longer needed.

This Actor is unofficial and is not affiliated with or endorsed by Instagram or Meta.

FAQ

Does it require an Instagram account?

No. The Actor does not require Instagram login credentials, cookies, or a session ID.

Are arbitrary keywords supported?

No. Submit actual hashtags only. Values with spaces, URLs, or punctuation other than underscore are rejected.

Are Reels included?

Yes. Posts and Reels come together in one dataset, in Instagram's relevance order. Reels have product_type: "clips" and media_type: "video", so you can filter them after export.

Can one run process multiple hashtags?

The schema accepts up to 50 unique hashtags. Plan-specific FREE demo limits are detailed in Pricing. Continuation runs require one hashtag because a token is bound to that source.

How do I export the results?

Open the run's default dataset in Apify Console and choose JSON, CSV, Excel, XML, or another supported format. Programmatic workflows can read the same dataset through the API.

Support

Use the Issues tab on this Actor's Apify Store page. Include the run ID, expected result, and observed result. Never include continuation tokens, API tokens, cookies, or other credentials.