Best LinkedIn APIs in 2026: partner vs public

"LinkedIn API" or "LinkedIn data API" usually means LinkedIn's partner and Marketing Developer Platform products. In practice, teams shopping those phrases need one of three things: member-authorized apps LinkedIn has approved, public person or company enrichment without a partner deal, or browser automations that scrape Sales Navigator-shaped workflows.

Those are different products. Mixing them up is how you end up buying Phantom hours when you needed a profile GET, or waiting months for partner access when a public-page lookup would have shipped the feature. Longer walkthrough of that fork: LinkedIn partner API vs public data.

Proxycurl shut down in July 2025 after LinkedIn litigation. Rate cards from a dead vendor still float around in RFPs, so treat those numbers as history.

This is a coverage and pricing comparison for 2026. Social Fetch is the product I build, so weigh that. It leads the list because it fits sync REST over public LinkedIn people, companies, posts, jobs, Ad Library, and search, especially if you already normalize creators across TikTok and Instagram. Use LinkedIn's official APIs when you need member OAuth or partner features.

The 7 LinkedIn data APIs at a glance

1. Social Fetch: synchronous REST for public LinkedIn data, with the same envelope as 22 other platforms.

2. LinkedIn official APIs: the authorized route for member scopes and LinkedIn-approved products.

3. Enrich Layer: a LinkedIn-focused continuation path after Proxycurl shut down.

4. Apify: marketplace Actors when you need custom fields or scheduled runs.

5. Bright Data and Oxylabs: enterprise proxy and dataset infrastructure at volume.

6. PhantomBuster: cloud automation for sales playbooks, not a profile schema API.

7. RapidAPI: a discovery marketplace of third-party LinkedIn listings, not one production contract.

Full LinkedIn route list and examples live on the LinkedIn platform page. Partner versus public enrichment decision: LinkedIn partner API vs public data.

How I compared these APIs

I apply the same criteria to every vendor and weigh production behavior over landing-page adjectives. Each entry below carries the same labeled fields: Best for, Strengths, and Limitations.

The criteria are what you can actually call (people profiles, companies, organizations and schools, posts, people search, jobs, Ad Library, transcripts), whether the route is official or public enrichment, whether the response is synchronous, how person, company, and job entities are separated, how errors are typed, and what a second platform costs in schema work. Read LinkedIn's terms for your use case, because I am not your lawyer.

I also ask about vendor concentration risk, since Proxycurl showed that a LinkedIn-only enrichment business can vanish. Competitor figures are limited to what each vendor publishes, and I left out numbers I could not tie to a public source.

1. Social Fetch

Best for: product teams and agents that need public person cards, company pages, posts, jobs, and search without a LinkedIn partner app, and that already pull other social platforms.

Strengths: One key and one parser across 23 platforms and 232 documented operations, synchronous JSON in a single envelope, free failed lookups, credits that never expire, and an MCP server for agents.

Limitations: Public data only, so there is no Sales Navigator, InMail, or partner analytics. It covers 23 platforms, not forty, and heavier routes like LinkedIn profiles cost 3 credits.

Social Fetch exposes LinkedIn routes under /v1/linkedin/*: person profiles (batch up to 50 URLs), profile posts, people search, organizations, company pages, company posts, individual posts, post transcripts, post search, Ad Library get and search, jobs, and job search. Responses use the shared { data, meta } envelope, and follower-style metrics follow the same conventions as other platforms.

Person, company, and organization routes are separate, so you do not feed a /company/… URL into a person endpoint and hope. Batch profile lookups accept up to 50 URLs per request at 3 credits per URL requested. Every response includes meta.requestId and meta.creditsCharged, and the same key works for TikTok, Instagram, YouTube, and the rest of the catalog.

People search and job search are slower and metered differently than a single profile GET. Plan for about a minute and a half on those searches, and read the credit notes, which combine an attempt fee with per-record credits. Coverage follows what is publicly visible, and gated or logged-in-only sections are out of scope.

Pricing: 100 free credits with no card, then from $1.35 per 1,000 credits on the Scale annual plan, or $1.64 per 1,000 on the Scale pack. Starter is $14 for 1,000 credits. These are the published prices as of October 2026, and all packs are on pricing. Profile GETs are 3 credits per URL requested, company pages, posts, and many other flat routes are 1 credit on success, and people search bills a 2-credit attempt fee plus 2 credits per person returned. Credits charge on completed lookups, so trust meta.creditsCharged on every response.

Verdict: The sensible default when LinkedIn is one surface in a multi-platform product, or when you want synchronous public JSON without waiting on partner access.

Example person lookup.

Example company lookup.

For a one-off paste-and-copy person lookup without wiring keys, use the free LinkedIn profile viewer. Full route list: LinkedIn hub.

2. LinkedIn official APIs

Best for: apps that manage member-authorized experiences, use Marketing Developer Platform features, or must stay inside LinkedIn's partner access model.

Strengths: The first-party terms path when you qualify, stable docs for the surfaces you are allowed to call, and member OAuth flows.

Limitations: Access is gated by partner and MDP review, and many enrichment use cases never receive the scopes teams assume exist.

LinkedIn's official APIs are the correct product when LinkedIn has to approve what you build. Partner and MDP processes gate access. That is the right path for products that act as a LinkedIn-connected app, and the wrong one for looking up any public /in/… card for CRM enrichment.

Access is not "sign up and GET any profile." Teams that watch a general LinkedIn API tutorial often assume scopes that review never grants, so expect review timelines.

Verdict: Use it when ownership and compliance require LinkedIn partner access. For public enrichment volume, a public-data API usually ships sooner if you never get, or should not apply for, those scopes.

3. Enrich Layer

Best for: teams that want a Proxycurl-shaped migration and whose roadmap is LinkedIn enrichment only.

Strengths: An explicit migration story for former Proxycurl customers, a LinkedIn-deep focus, and less remapping if your clients already speak Proxycurl-shaped requests.

Limitations: It concentrates enrichment risk on one network and one operator lineage, and field depth and people-search behavior differ from other options.

Proxycurl shut down in July 2025 after LinkedIn litigation. Enrich Layer positions itself for former customers, with similar endpoints, key and credit carryover for many users, and a base URL change. If wire compatibility beats multi-platform breadth, evaluate that path.

Compare against your actual payloads rather than nostalgia for old Proxycurl rate cards. Longer migration notes: Social Fetch vs Proxycurl.

Verdict: A good match for LinkedIn-only stacks that refuse a schema remap. It is a weaker fit if TikTok or Instagram is next quarter.

4. Apify

Best for: scheduled profile dumps, niche field extraction, and scrapers you want to configure as Actors rather than maintain as a fleet of browsers.

Strengths: Marketplace breadth when no managed REST route exposes the field you need, plus built-in scheduling.

Limitations: Schemas differ by Actor author, results arrive through run polling, and Actor quality varies.

Apify hosts many LinkedIn Actors covering profiles, companies, posts, jobs, and sales-adjacent scrapers. You start a run, poll it, and download a dataset. Billing mixes compute and event fees, and monthly plan credits expire.

You are buying a runtime, not one LinkedIn schema. Interfaces that need a single HTTP round trip fit poorly, and LinkedIn anti-bot pressure hits Actor authors unevenly.

Verdict: Worth a look for batch and custom scrapes. It is a weak fit as your only integration if you want one typed client for person and company data in a request handler.

5. Bright Data and Oxylabs

Best for: high-volume teams with budget for proxy networks, dataset pipelines, and sales-led contracts.

Strengths: Hard-target reliability at volume, SLAs and legal processes that enterprises already staff for, and warehouse delivery patterns.

Limitations: There is no unified multi-social schema by default, and procurement cycles and spending floors do not suit a product engineer who needs public profile JSON in an onboarding flow.

These vendors sell unblock and collection infrastructure. LinkedIn scrapers and dataset products exist on top of residential IPs, browser farms, and async exports. You are buying a platform with spending limits, domain tiers, and procurement cycles that large data organizations already understand.

Floors and spending caps matter as much as the unit rate. Hit a cap on launch day and requests stop until someone with admin access raises the limit. You also still own parsers unless you buy a higher-level dataset product.

Verdict: Justified when volume and hard-target success rates dominate the decision. See Social Fetch vs Bright Data.

6. PhantomBuster

Best for: growth and sales ops teams that need cloud browser playbooks, such as scrapes into Sheets or outreach sequences, billed in execution hours and slots.

Strengths: Pre-built sales workflows that a human or an ops tool can own end to end.

Limitations: You think in hours, slots, and exports rather than typed profile responses, and export caps on free and trial tiers constrain volume.

PhantomBuster is easy to put on LinkedIn API lists because LinkedIn Phantoms are popular. The product is an automation runtime, and paid plans expose an API to drive those automations. That is awkward inside a SaaS request handler that needs 500 profile cards during onboarding.

Verdict: Keep it for outreach playbooks, not as your production enrichment API. Longer take: Social Fetch vs PhantomBuster.

7. RapidAPI

Best for: finding a niche LinkedIn endpoint over a weekend without signing an enterprise contract or a partner deal.

Strengths: Fast discovery under one billing dashboard and a test console that shows quickly whether the JSON looks like a person card.

Limitations: Listing quality varies, person, company, and job endpoints may use incompatible schemas, and support tickets route through a middleman.

RapidAPI hosts third-party LinkedIn scrapers under one billing dashboard. Prototypes move fast because you are not waiting on Marketing Developer Platform review. LinkedIn is a hard target, so listing quality varies more than on softer networks, and anti-bot pressure hits publishers unevenly.

Production teams that stitch several LinkedIn marketplace APIs usually inherit several schemas, several rate limits, and several places a publisher can vanish after a terms scare. Check the publisher's meter before you promise CRM volume.

Verdict: Keep it for exploration, and prefer a first-party public-data API when enrichment is load-bearing. Longer take: Social Fetch vs RapidAPI.

Official API vs public enrichment

Most LinkedIn API comparisons bury this fork. In 2026 it decides the vendor list before pricing does. If you want the access-friction version of the same decision, read LinkedIn partner API vs public data. For DIY versus partner and MDP versus Social Fetch with runnable GETs, see How to scrape LinkedIn data.

• Member OAuth and LinkedIn-approved app features: official partner and MDP APIs.

• Public /in/… or /company/… card for CRM enrichment: a managed public-data REST API such as Social Fetch or Enrich Layer.

• Nightly custom scrape with odd fields: an Apify Actor.

• Sales outreach sequences in the browser cloud: PhantomBuster.

• One person URL, no API key yet: /tools/linkedin-profile-viewer.

Social Fetch only returns publicly visible data, and content gated beyond public visibility is out of scope. Official APIs do not unlock any profile on the internet either. They unlock what LinkedIn authorizes for your app and members.

Pass a company URL into the person profile route and you will not get a company card. Use GET /v1/linkedin/companies, or GET /v2/linkedin/organizations/resolve then GET /v2/linkedin/organizations, for the right page type. Branch on lookupStatus before you upsert.

Which API fits which job

• Multi-platform enrichment product, synchronous JSON: Social Fetch (LinkedIn hub).

• LinkedIn partner or member-authorized app: the official LinkedIn APIs.

• Proxycurl-shaped LinkedIn-only migration: Social Fetch or Enrich Layer, where Enrich Layer keeps the Proxycurl request shape and Social Fetch adds the other platforms behind one key.

• Nightly scrape with custom fields: an Apify Actor.

• Enterprise proxy volume: Bright Data or Oxylabs.

• Outreach automation playbooks: PhantomBuster.

• One-off person profile in the browser: /tools/linkedin-profile-viewer.

If you only need LinkedIn today but Instagram next quarter, pick the option whose envelope you can reuse. Rewriting parsers per platform is where most cheap LinkedIn APIs get expensive.

Questions before you buy

1. Do you need member-authorized scopes, or public enrichment? That fork decides official versus third-party first.

2. Person, company, job, or Ad Library? Confirm the vendor has separate, documented paths rather than one HTML dump.

3. Will people and job search latency fit your UI? Some search routes take well over a minute, so product handlers that need sub-second cards should not share a path with deep search.

4. Synchronous or batch? Request handlers prefer synchronous calls, and overnight dumps can live on Actors or datasets.

5. One schema across platforms? Call LinkedIn and TikTok with the same parser, because needing if (platform === "linkedin") for basic identity fields means you do not have unification.

6. Do credits expire? Spiky enrichment workloads hate monthly resets.

7. What is your single-vendor LinkedIn risk? Ask what you would do if this provider disappeared the way Proxycurl did.

8. Can you preview without shipping? Use a playground, free credits, or the free LinkedIn profile viewer before you lock a vendor.

Frequently asked questions

Smoke-test a person or company URL in the playground, or start with 100 free credits on pricing. For one public person profile without an API key, open the free LinkedIn profile viewer. Partner versus enrichment path: LinkedIn partner API vs public data.