# Credits and usage: what costs credits (https://snitchfeed.com/docs/credits) Credits pay for the work SnitchFeed does for you, such as scanning for posts and scoring mentions with AI. Your plan includes a set number of credits each billing period, and that is what limits how many mentions you get. See [Plans & billing](/docs/workspace/plans-and-billing). ## What costs credits [#what-costs-credits] ### Scanning for posts [#scanning-for-posts] Reddit and Bluesky are watched in real time, so they have no scan cost. X, LinkedIn and Hacker News are checked on a schedule, and each check costs credits. See [Platforms](/docs/platforms). | Activity | Credits | Charged | | ---------------------------- | ---------------------------------------------: | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **X keyword scan** | {CREDIT_COSTS.TWITTER_POLL} | Per keyword, each time the X scan runs | | **LinkedIn keyword scan** | {CREDIT_COSTS.LINKEDIN_POLL} | Per search, each time the LinkedIn scan runs. A keyword with author job title filters runs one search per job title. | | **Hacker News keyword scan** | {CREDIT_COSTS.HACKERNEWS_POLL} | Per keyword, each time the Hacker News scan runs | | **X tracked profile** | {CREDIT_COSTS.TWITTER_USER_TWEETS_AND_REPLIES} | Per page of posts fetched for a [tracked profile](/docs/listeners/tracked-profiles) | | **LinkedIn tracked profile** | {CREDIT_COSTS.LINKEDIN_PERSON_POSTS} | Per page of posts fetched for a tracked person or company | | **Backfill** | Same as a scan | When you add a keyword, SnitchFeed searches recent posts right away. LinkedIn backfill fetches up to 2 pages per search. See [Scans & backfill](/docs/listeners/scan-intervals). | ### AI [#ai] | Activity | Credits | Charged | | ---------------------- | -----------------------------------------: | ------------------------------------------------------------------------------------------------------------------------------------------ | | **AI mention scoring** | {CREDIT_COSTS.MENTION_AI_SCORING} | Per mention scored and stored, on every platform including Reddit and Bluesky | | **AI re-scoring** | {CREDIT_COSTS.MENTION_AI_SCORING} | Per existing mention scored again | | **AI reply draft** | {CREDIT_COSTS.MENTION_RESPONSE_GENERATION} | Per reply generated. See [AI replies](/docs/features/ai-responses#generate-a-reply). | | **Snitch AI chat** | Varies | Each message is priced by how much work it takes, so short questions cost less than long tasks. See [Snitch AI](/docs/features/snitch-ai). | ### GTM tools, MCP and API [#gtm-tools-mcp-and-api] These cost the same whether you run them from **GTM Tools** in the app, the [MCP server](/docs/mcp) or the [REST API](/docs/api/introduction). | Activity | Credits | | ----------------------------------- | -----------------------------------------------------------------------------: | | **LinkedIn profile posts** | {CREDIT_COSTS.MCP_GET_LINKEDIN_PROFILE_POSTS} per page (1 to 10 pages per run) | | **LinkedIn profile comments** | {CREDIT_COSTS.MCP_GET_LINKEDIN_PROFILE_COMMENTS} per page | | **LinkedIn post comments scraper** | {CREDIT_COSTS.LINKEDIN_POST_COMMENTS} per comment (100 to 1,000 per run) | | **LinkedIn post reactions scraper** | {CREDIT_COSTS.LINKEDIN_POST_REACTIONS} per reaction (100 to 1,000 per run) | | **X search** | {CREDIT_COSTS.MCP_SEARCH_TWEETS} | | **LinkedIn search** | {CREDIT_COSTS.MCP_SEARCH_LINKEDIN_POSTS} | | **Hacker News search** | {CREDIT_COSTS.MCP_SEARCH_HACKERNEWS_POSTS} | | **Reddit search** | {CREDIT_COSTS.MCP_SEARCH_REDDIT_POSTS} | | **Query tests on MCP** | 0 | For the cost of each API endpoint, see [API credits](/docs/api/credits). ### Free [#free] * **CSV export** costs {CREDIT_COSTS.MENTIONS_EXPORT} credits, however many rows. Export is a Pro feature. See [CSV export](/docs/features/csv-export). * Browsing feeds, notifications (Slack, Discord, email, webhooks) and workflow steps use no credits. ## Estimate your monthly use [#estimate-your-monthly-use] For one listener on X, LinkedIn or Hacker News, over a 30-day month: ``` credits ≈ (24 ÷ scan_interval_hours) × 30 × keyword_count × credits_per_keyword_scan ``` * **scan\_interval\_hours**: how often the listener scans that platform, set under **Scan Intervals**: 1, 2, 4, 6, 12 or 24 hours. * **credits\_per\_keyword\_scan**: {CREDIT_COSTS.TWITTER_POLL} for X, {CREDIT_COSTS.LINKEDIN_POLL} for LinkedIn, {CREDIT_COSTS.HACKERNEWS_POLL} for Hacker News. **Example:** a listener with 5 keywords on all three platforms, using the default intervals (X every 4 hours, LinkedIn every 12 hours, Hacker News every 6 hours) and no job title filters: | Platform | Calculation | Credits per month | | ----------- | -------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | | X | (24 ÷ 4) × 30 × 5 × {CREDIT_COSTS.TWITTER_POLL} | {((24 / 4) * 30 * 5 * CREDIT_COSTS.TWITTER_POLL).toLocaleString("en-US")} | | LinkedIn | (24 ÷ 12) × 30 × 5 × {CREDIT_COSTS.LINKEDIN_POLL} | {((24 / 12) * 30 * 5 * CREDIT_COSTS.LINKEDIN_POLL).toLocaleString("en-US")} | | Hacker News | (24 ÷ 6) × 30 × 5 × {CREDIT_COSTS.HACKERNEWS_POLL} | {((24 / 6) * 30 * 5 * CREDIT_COSTS.HACKERNEWS_POLL).toLocaleString("en-US")} | | **Total** | | **{((24 / 4) * 30 * 5 * CREDIT_COSTS.TWITTER_POLL + (24 / 12) * 30 * 5 * CREDIT_COSTS.LINKEDIN_POLL + (24 / 6) * 30 * 5 * CREDIT_COSTS.HACKERNEWS_POLL).toLocaleString("en-US")}** | A few adjustments: * **LinkedIn job title filters:** multiply by the number of job titles, since each one is its own search. * **Tracked profiles:** use the number of profiles instead of keywords, at {CREDIT_COSTS.TWITTER_USER_TWEETS_AND_REPLIES} credits per page for X and {CREDIT_COSTS.LINKEDIN_PERSON_POSTS} for LinkedIn. * **AI scoring:** add {CREDIT_COSTS.MENTION_AI_SCORING} credits for each mention stored, on every platform. Scheduled scans and backfills stop until your credits reset at the start of the next billing period, or until you buy a top-up pack. Snitch AI chat also needs credits to reply. See [Buy more credits](/docs/workspace/plans-and-billing#buy-more-credits). ## Check your usage [#check-your-usage] Open **Workspace Settings → Credits Usage** ([app.snitchfeed.com/settings/usage](https://app.snitchfeed.com/settings/usage)). Owners and collaborators can both view it. * **Usage** shows credits used this period against your plan's included credits. * The chart groups usage by source. Switch between **This billing period**, **Previous billing period** or a **Custom range**. * The ledger lists every charge with its **Time**, **Source** and **Credits**. Click a row for details, such as the keyword, listener or mention. The **Credits** tab under **Workspace Settings → Plan & Billing** shows the same balance, the reset date and any top-up credits left. ## Next steps [#next-steps] * [Plans & billing](/docs/workspace/plans-and-billing): credits per plan and how to buy more. * [Scans & backfill](/docs/listeners/scan-intervals): choose how often each platform is scanned. * [API credits](/docs/api/credits): cost of each REST API endpoint.