Sprout Social API: Access, Tokens, Limits and What It Won't Return

What the Sprout API is, and who gets it

The Sprout Social API gives a developer programmatic access to the data in your own Sprout account: the profiles you connected, their analytics, the messages in your inbox, your cases and your Listening topics. It can also write, within narrow limits, by creating draft posts and uploading media. The typical jobs are dashboards, automated reporting and pushing content from a project tool into Sprout's calendar. It is not a general-purpose social network API. It reads what Sprout already holds about your owned profiles and your Listening topics, and it never reaches into competitor profiles, ad accounts or other accounts' profile data.

Access is gated twice. The account must be on the Advanced plan, and a user needs the API Permissions permission, which an Account Owner grants ^[1]; it is a user permission, not a separate purchase. Sprout's release notes date the opening to May 2025, when every Advanced customer gained access once that permission was granted. Before then the API was a narrower affair, and some of Sprout's own pages still show it. The general FAQ on the main site says there is no public API and that access is an add-on requested by emailing support. That page is outdated. We follow the newer help pages and the developer documentation.

The access checklist before your first call

Getting from a signed contract to a working request takes a few separate steps, and they are owned by different people. Sprout's marketing page about the API adds that use must be authorized by your Sprout account representative before you start, so loop them in early. Then work through the list below. None of the steps is hard. Missing one is the usual reason a first request fails.

• Confirm the account is on Advanced (or Enterprise, which is quoted).
• Ask the Account Owner to grant the API Permissions permission to whoever will build.
• Accept the Sprout Analytics API Terms of Service under Settings, Global Features, API.
• If you want X analytics data, have a user with API Permissions accept the X Content End User License Agreement; X must also approve your use case.
• Decide who inside your company owns the integration, because Sprout offers no implementation help.

That last point deserves weight. The help center says plainly that Sprout does not help implement the API and that the customer needs internal ownership. Budget developer time, not just a plan upgrade. A marketing team without an engineer or a data analyst on call will struggle to get value here, and the native integrations further down this page may serve it better.

Tokens or OAuth: choosing how you authenticate

Sprout supports two methods. OAuth 2.0 is the recommended one: it issues short-lived JWT access tokens. You create an OAuth client in Settings, Global Features, API, where machine-to-machine is the recommended flavor and user-based authentication needs allowed redirect URIs. After saving, you copy the client ID and secret from the OAuth Client Management section and store them securely. The alternative is a static API token. Either one goes in the Authorization header as a Bearer token, and every request goes to the base URL at api.sproutsocial.com. Only the major version, such as v1, sits in the path. Minor versions stay backwards compatible.

Who controls the tokens

Static tokens are simpler, and that is exactly why they need care. Sprout's token page says they have no fixed expiry, though it recommends regenerating them periodically ^[2]. Sprout also offers IP allowlisting to restrict API access to addresses you choose, and MFA has been mandatory for every Sprout user since April 2025. The pages disagree slightly on who manages tokens. The developer documentation says any user with API Permissions can create, view or invalidate them, while the help page on account roles says only the Account Owner can access, create or delete them. Plan for the stricter rule. Keep the Account Owner in the loop for every new or revoked token.

Rate limits, pagination and the errors worth handling

The ceiling is 60 requests a minute and 250,000 requests a month ^[3]. For a nightly reporting job that is generous. For a dashboard that refreshes every few seconds for a large team, it is not, so cache results instead of calling live. Analytics responses are capped at 10,000 results, and anything larger needs cursor pagination on the guid field, which is the documented way past the cap. Profile analytics take up to 100 profiles and one year per request, with 1,000 results per page. Post analytics take up to 100 profiles per request and return 50 posts per page.

Four status codes deserve handling in code rather than in a support ticket. A 429 means you went too fast or used up the monthly allowance. A 504 means the request was too large, and the fix is a smaller scope, such as fewer profiles or a shorter range. A 202 means the request was valid but the answer is not ready yet, so retry. A 401 means a missing or expired token, so reauthenticate. One more quirk catches people comparing numbers across networks: daily profile activity uses each network's own time zone, UTC for some networks and Pacific time for Facebook, Instagram and YouTube.

What you can read: profiles, analytics, messages and more

Account metadata and analytics

The metadata endpoints are read-only and return client IDs, profiles, tags, groups, users, Listening topics, teams and case queues. The tags endpoint returns active and archived tags without pagination, and deleted tags are simply gone. Analytics covers X, Facebook, Threads, Instagram, LinkedIn, Pinterest, YouTube and TikTok, with network-specific depth: Instagram and Threads follower demographics by age, gender, city and country, LinkedIn followers by job function and seniority, and TikTok impression sources. Some networks are thin. Bluesky has post metrics only. Pinterest post metrics stop at comments and saves, and YouTube profile metrics stop at followers, growth and posts sent.

Messages, cases and reviews

The messages endpoint returns inbox data and metadata, needs a filter on every request, pages forward only by cursor and returns at most 100 items per page. It can show the first reply, tag, like or complete action on a message, with who did it and when, which is the raw material for response-time reporting. Case endpoints arrived in April 2025. They need either a case ID or a date filter no wider than one week, which makes backfills slow but predictable. Since November 2025 the messages endpoint also returns reviews from the Apple App Store, Facebook, Google Play and Google Business Profile (the last 30 days only for Google Business Profile), with a rating field.

Listening topics

Listening came to the API in December 2023. Topic messages return raw posts for processing or dashboard samples, with a sentiment label and a raw score between 1 and -1. Topic metrics aggregate volume, engagement and trends, bucketed by day, month, network, sentiment, language or location, and return everything in one response. Listening is a paid add-on, so these endpoints only matter if you already buy it. Our Sprout Social Listening guide covers what the add-on itself does.

What you can write: drafts and media

The publishing endpoints, launched in May 2024, create draft posts only. Nothing publishes or schedules itself directly from the API; a person still approves and sends the post inside Sprout. That makes the API a good feeder from content and project tools and a poor fit for fully automated posting. Supported destinations include Instagram business and creator profiles, Facebook Pages, Threads, X, LinkedIn pages and personal profiles, YouTube, TikTok, Pinterest and Google Business Profile. Recent additions include LinkedIn drafts with PDF files and Instagram mixed-media carousels in April 2026, and TikTok photo posts in August 2026.

Three quirks are worth knowing before you build. A draft sent to several profiles fans out into one calendar post per profile. If the media does not suit one of those profiles, that profile is silently dropped with no error, so check the calendar after a bulk send. That one bites quietly. And a post you fetch back through the publishing endpoint always shows a delivery status of pending, even after it goes live, so use the messages endpoint to confirm what was published. Scheduled times must be in the future and are rounded down to the minute.

Media uploads have their own rules. A single upload is limited to 50 MiB; larger files go up in multipart chunks of 5 MiB, and a multipart upload must be finished and used within 24 hours. Uploaded media is kept for 24 hours unless a post uses it. Sprout does not transcode, so files must already suit the target network. Accepted formats include common image and video types plus PDF and SRT.

What the API will not give you

The list of exclusions is long, and it shapes most projects. The API carries no paid or ad account data, no competitor data, no BI connectors and no account or profile management. X analytics data needs the license step above and X's approval of your use case. X Listening data and Reddit Listening message-level data are excluded too. Review data from Yelp, Trustpilot, Tripadvisor and Glassdoor stays inside the Sprout app. Google Business Profile data from the messages endpoint, reviews included, is limited to the last 30 days. Direct message text comes through, but DM media does not, and the media links returned for DM images and videos do not work.

Metrics that need Premium Analytics

Even on Advanced, part of the metric catalog is locked behind the Premium Analytics add-on. The documentation lists X media views, many Facebook video metrics, sentiment counts and YouTube annotation and card metrics among them, and Net Sentiment Score needs Premium Analytics on every network that reports it. Premium Analytics has no published price, so ask your account representative for a quote before you promise a stakeholder a sentiment dashboard. If you only need standard engagement figures, the base Advanced access is enough.

Tableau and other BI connectors

Sprout's Tableau connector is a web data connector that pulls profile, post and tag tables from the API into Tableau. It needs API access, a valid API key and the Premium Analytics add-on, and the help page notes that only Account Owners can reach the key. It does not bring in messages data, and it does not support IPv6, which matters on some corporate networks. Marketing Cloud Intelligence, formerly Datorama, also has a Sprout connector that runs on an API key, with the same requirements: Advanced and Premium Analytics. For a team already living in Tableau, this is the shortest path to Sprout data without writing code, as long as the add-on fits the budget.

Native integrations that may beat building your own

Before writing code, check whether a native integration already does the job. The CRM and helpdesk integrations are Advanced-only, like the API. Salesforce Service Cloud lets care teams work social messages from Salesforce with automated case creation, requires Advanced plus a Professional Services implementation, supports up to 10 instances and never pushes data unless a user, rule or macro triggers it. The Agentforce add-on needs Advanced, the Service Cloud integration and an Agentforce subscription. Zendesk and HubSpot Service Hub each connect up to 10 instances, and HubSpot syncs comments both ways. Microsoft Dynamics 365 only receives data from Sprout, creating cases, contacts and leads, with no two-way link.

The workflow integrations sit on more plans. Canva and Adobe Express designs can be sent to Sprout as draft posts on all plans, and Compose accepts media from Bynder, Canva, Dropbox and Google Drive. Slack notifications cover approval and case workflows once an Account Owner switches them on. On Advanced, Opal, Optimizely and Slate feed content into Sprout for publishing. None of these integrations counts against your social profile limit. One more connection is often confused with the API: Sprout's MCP server, which is supported for ChatGPT only. It is read-only, covers TikTok publishing data alone, and is open to Professional and above per the newer help page. Our guide to Sprout Social AI covers it in detail.

Is the API worth Advanced pricing?

Advanced costs $399 per seat a month billed annually, or $499 per seat billed monthly, which is $4,788 per seat a year ^[4]. Professional costs $299 per seat a month billed annually. Because pricing is per seat, the upgrade multiplies across the team. By our arithmetic, three seats on Advanced billed annually come to 3 × $399 = $1,197 a month, or $14,364 a year, against $897 a month for the same three seats on Professional, a gap of $300 a month. All annual plans renew automatically. Annual plans paid upfront in the app are cancelled in the billing page before the renewal date, while annual plans paid monthly and invoiced plans need 30 days' notice by email. There are no refunds on cancellation or downgrade. Enterprise is quoted.

The API alone rarely justifies that gap. It does when you need Sprout data in a warehouse or BI tool, when a content system should drop drafts into Sprout's calendar, or when Advanced is already on the table for the care features it unlocks: cases, chatbots, sentiment and the CRM integrations. It does not when you want automated posting, competitor benchmarks, ad data or X Listening data, because the API returns none of them. For a full view of plans and billing rules, see our Sprout Social pricing breakdown.

The bottom line

The Sprout Social API is a solid, well-documented way to get your own Sprout data out and drafts in, with sensible limits of 60 requests a minute and 250,000 a month. It is gated to Advanced, needs an owner inside your company and leaves out paid, competitor and X Listening data, while publishing stops at drafts. Buy Advanced for the API only if you have a reporting or content pipeline that needs it; otherwise the native integrations, or the Tableau connector with Premium Analytics, will get most teams there with less work.