For about three years the correct answer to this question was no, and the internet has not noticed that it changed. We had it wrong in our own product copy until last week, which is part of why this article exists.
Does LinkedIn have an analytics API?
Quick answer
Yes. LinkedIn has at least seven documented analytics endpoints. For your own personal posts the one you want is memberCreatorPostAnalytics, launched 8 July 2025 behind the r_member_postAnalytics permission. It returns eleven metrics including impressions, members reached, saves, sends and link clicks. It sits inside the approval-gated Community Management API.
The endpoint is real and it is maintained. LinkedIn's own changelog carries three separate entries for it: the 202506 launch, six new metrics at 202604, and a breaking change at 202605 that altered the response format. That is not a token gesture left to rot.
| Added | Metrics |
|---|---|
| 202506, at launch | IMPRESSION, MEMBERS_REACHED, RESHARE, REACTION, COMMENT |
| 202604 | POST_SAVE, POST_SEND, LINK_CLICKS, PREMIUM_CTA_CLICKS, FOLLOWER_GAINED_FROM_CONTENT, PROFILE_VIEW_FROM_CONTENT |
| Never | Engagement rate, and any demographic breakdown of who saw it |
It is not one API, it is at least seven
Most of the confusion in this topic comes from treating LinkedIn analytics as a single thing. It is not. Different endpoints serve different entities, need different permissions, and sit in different approval programmes. Landing on the wrong one is the most common way people conclude nothing exists.
| What you want to measure | Endpoint | Permission | Access |
|---|---|---|---|
| Your own posts | memberCreatorPostAnalytics | r_member_postAnalytics | Community Management, approval |
| Your own follower count | memberFollowersCount | r_member_profileAnalytics | Community Management, approval |
| Your own video posts | memberCreatorVideoAnalytics | Member analytics set | Community Management, approval |
| A company page's posts | organizationalEntityShareStatistics | rw_organization_admin | Community Management, approval |
| A company page's followers | organizationalEntityFollowerStatistics | rw_organization_admin | Community Management, approval |
| A company page's page views | organizationPageStatistics | rw_organization_admin | Community Management, approval |
| Ad performance | adAnalytics | r_ads_reporting | Advertising API, approval |
| Reading your own post content | Posts API | r_member_social | Closed, not accepting requests |
‼️ Note the last row. You can publish to LinkedIn self-serve and you can now measure what you published, but reading your own posts back is the one thing still shut. That inversion is the most consequential fact in LinkedIn's API surface, and the next section is about what it does to every tool built on it.
LinkedIn shipped the analytics without the discovery
This is the part that no other article on the subject covers, and it matters more than the endpoint itself.
The analytics endpoint's per-post finder takes a post URN as its input. To ask how a post performed, you must already know which post to ask about. LinkedIn does publish an endpoint that lists a person's posts, and its documentation is explicit about the catch: retrieving all posts authored by a person requires r_member_social. That is the permission LinkedIn describes as restricted, available to approved users only, and which the Community Management overview says plainly is closed, with requests not being accepted due to resource constraints.
Put those two facts together and the consequence is concrete. An approved tool can only report on posts it published itself, because those are the only post URNs it holds. It has no sanctioned way to enumerate the posts that were already on your profile when you signed up.
| Official API route | Browser session route | |
|---|---|---|
| Posts you publish from now on | Yes | Yes |
| Posts you published before connecting | No, the URNs are unreachable | Yes, your whole history |
| Terms exposure | The posting clause, see the terms article | The browser plugin clause, plus the posting clause |
You can watch this surface in the wild. One official launch partner tells its users that post statistics are only available for content published through it, and attributes that squarely to LinkedIn's API limitations. That reads like a product shortcoming. It is not. It is the permission model showing through.
This also rescues the conventional wisdom, which turns out to be wrong on the letter and broadly right on the effect. When practitioners said LinkedIn had no analytics API, what they meant, without quite having the words, was that no compliant tool could show them their own back catalogue. That was true before July 2025 and it is still true today.
How we know the endpoint is genuinely in use
Vendor marketing is not evidence. Three things are, and none of them is a claim anyone made on purpose.
- Metricool's help centre reproduces a LinkedIn internal caveat almost word for word, warning that impressions, members reached, reshares, reactions and comments are not consistent with the user interface at the moment. Those are the five launch metric constants, in LinkedIn's own order. You do not write that sentence unless you are reading that doc.
- Hootsuite's July 2026 changelog adds exactly six metrics: sends, saves, link clicks, premium CTA clicks, followers from a post, and profile views from a post. That is a one to one match with LinkedIn's 202604 additions, about three months later.
The third proof is the cleanest, because you can see it in a product rather than a changelog. In July 2024, Buffer described personal-profile LinkedIn analytics as likes and comments. In July 2025, calling itself an official launch partner, it listed impressions, watch time and unique viewers. Likes and comments are printed on the page for anyone to read. Impressions and unique viewers are private to the creator and were never on the page to scrape. A tool does not acquire those by trying harder.
Fingerprints like these are worth more than a press release, because they are involuntary. They also tell you something useful: if a tool's LinkedIn metrics change three months after LinkedIn's version notes change, that tool is on the official API rather than reading your screen.
There is a faster version of that test you can run yourself in about ten seconds, and it is the most practical thing in this article.
‼️ Look at the metric labels a tool shows for a personal profile. If one of them is members reached, it is reading LinkedIn's official API. That is not a phrase a designer writes. It is LinkedIn's own metric constant, MEMBERS_REACHED, surfacing through the interface. One launch partner's help page lists it next to impressions, video plays and watch time, in the API's vocabulary rather than its own.
The pairing is what makes it reliable. Reactions, comments and reposts are printed on the page for anyone to read, so any tool can show you those by any method. Impressions and members reached are private to the author and were never on the page to take. A product showing you both, for a personal profile, is on the official endpoint, whatever its marketing says. A product showing you only the public three is not, whatever its marketing says. Those public three are exactly what an engagement rate is built from, so if you have the numbers in front of you our free LinkedIn engagement rate calculator turns them into a percentage and is open about which benchmarks it bands against.
So why does everyone say it does not exist?
We checked rather than assumed. We pulled the pages that rank for this question and its close variants, read all of them, and counted.
- Three of seventeen name memberCreatorPostAnalytics or r_member_postAnalytics: LinkedIn's own documentation, one vendor blog, and an open-source issue thread.
- Nine present themselves as complete LinkedIn API guides and never mention it. Two of those were updated in August 2026 and run to roughly 5,500 and 15,000 words.
- Three cover the July 2025 launch accurately as news, without naming any endpoint, so a developer searching later never finds them.
- One states outright that metrics are not available for posts on a personal page. It was published in December 2022, it is still live, and it still ranks.
So the honest version is not that the category shouts a falsehood. It is that the category is quiet, and quiet is enough. A reader who works through three long guides on LinkedIn's API without once seeing member analytics mentioned will reasonably conclude they do not exist.
There is also a simple structural reason. The folklore is older than the endpoint. Every guide written before July 2025 was correct when it was published, and search rewards the pages that have been accumulating links since 2022.
The real barrier is eligibility, not existence
This is where the story gets less encouraging, and where the phrase "LinkedIn has no analytics API" survives as a rough approximation of a truth that is actually about paperwork.
- Access is limited to registered legal organizations for commercial use cases only. LinkedIn says so in those words.
- Personal email addresses will not pass vetting. A creator applying as themselves is refused at the form.
- A super admin of the LinkedIn Page associated with your organization has to verify the application.
- Rejection is terminal for that app. You cannot re-apply for Development tier access with the same app, so you create a new one and start again. One clean attempt each.
- Moving from Development tier to full Standard access requires a technical sign off that LinkedIn tells you to initiate by contacting your point of contact on its Business Development team. A cold applicant does not have one.
Worth stating as an unknown rather than a fact: we could find no public account, anywhere, of a developer outside the original launch cohort obtaining this permission. Fourteen months after launch, the most prominent open-source LinkedIn scheduling tool has an unanswered issue about applying for it. That is absence of evidence, not evidence of refusal, and we are not going to dress it up as either.
The Development tier also has a limit that shapes what any approved tool can show you. It allows 500 calls per app and 100 per member per 24 hours, and the endpoint takes one metric per call. Pulling all eleven metrics for a single post costs eleven calls, which works out at roughly nine posts per member per day before retries.
What the API will not tell you about your own posts
Even with access, personal profiles are deliberately thinner than company pages, and this asymmetry explains almost every limitation you will hit in a tool.
| Company page | Personal profile | |
|---|---|---|
| Follower demographics | Seven breakdowns: country, region, function, industry, seniority, company size, association type | None. A single integer |
| Historic window | Rolling 12 months | Not specified in the docs |
| Engagement rate | Returned | Not returned, tools compute it themselves |
There are three more limits worth knowing before you plan around this, all documented by LinkedIn and mentioned by essentially nobody else.
- LinkedIn's own doc warns that on the aggregate finder, reshares, reactions and comments are not consistent with what the interface shows. If a tool's numbers disagree with your LinkedIn page, that may be the API being honest rather than the tool being broken.
- Daily granularity is not supported for members reached, link clicks, followers gained from content, or profile views from content. Those come back as totals only.
- The counts are described as best-effort accurate and explicitly not for billing.
How long is anyone allowed to keep your numbers?
This is the most interesting unanswered question in the area, and we have not seen it raised anywhere else.
LinkedIn's data storage rules cap members' social activity data, defined to include posts and the metadata relating to them, at 48 hours. Organization page reporting data gets a year. There is no row in that table for member post analytics. The page's own last-modified stamp predates the endpoint by more than two years, and LinkedIn's tiebreaker says that where two rules could apply, the shorter retention wins.
Read strictly, that points at a 48 hour cap on the very data an analytics product exists to trend over months. We are not going to assert that reading, because LinkedIn has not confirmed it and the gap looks more like a document nobody updated than a deliberate rule. But it is unresolved in the published text, and it is a plausible structural reason why official-route tools often show you a thinner personal history than a browser extension that keeps its own copy.
What you can get today with no API at all
If you only want your own numbers rather than a programmatic feed, the API question is a detour. LinkedIn gives every member an export, and the common advice about it is out of date.
- Post analytics and creator analytics are available to all LinkedIn members. Creator mode is not required, whatever older guides say.
- You will find them under Analytics and tools on your profile, covering the past 7 days up to the past 365 days.
- There is an export button, upper right, that gives you an XLSX file. It covers impressions, engagements, reactions, comments, reposts, saves and sends, plus follower demographics by job title, location, industry, seniority and company size.
- The "Get a copy of your data" archive does not contain analytics. It gives you your posts, their text, dates and URLs, but no impressions, reach or engagement counts. People reach for it expecting performance data and it is not in there.
For a lot of people that XLSX is the whole answer, and it costs nothing but a few clicks.
What this means for the tools you already use
Once you know the shape of the API, the limitations tools apologise for stop looking like bugs. Several say so directly in their own help centres. One says its personal profile analytics are limited by LinkedIn's API. Another says post stats are only available for content published through it, because historic and externally published posts are not reachable. A third warns that reshare data is not available at a reliable level of accuracy for profile posts. A fourth keeps post performance data for two weeks.
Those are not different products being careless in the same way. They are four descriptions of the same set of constraints, which is a good sign they are all reading the same official surface. If you are choosing between them on features rather than plumbing, we keep a separate comparison in the best AI tools for LinkedIn personal branding.
The other side of the trade is visible too, and 2025 and 2026 were unkind to it. Shield, the tool that effectively defined personal LinkedIn analytics, wound down in May 2026 after seven years. Its own notice says that Google and LinkedIn made it clear it could not continue operating as it was built. Its co-founder put it more bluntly, saying both companies made that clear over a period of weeks and that they decided not to fight it. A year earlier, Kleo removed its Chrome extension after more than 70,000 users, with its founder saying simply that LinkedIn told them to, and adding in the same breath that Kleo itself was not going anywhere. It rebuilt as a web app. That distinction matters: the thing that died was the access method, not the company.
Note that Shield names Google as well as LinkedIn. Anyone weighing browser-based tooling should understand there are two enforcement surfaces, not one, and the extension store is the faster of them. We cover the LinkedIn terms side in will a LinkedIn MCP get your account restricted.
One last observation, offered as a test rather than an accusation. Several tools say they are built on LinkedIn's official APIs. Not one of the vendors we checked names the member post analytics endpoint. That is a fair question to put to any of them, ours included, and the answer tells you more than the marketing page does.
Where VoiceMoat stands, and what we get wrong
We do not hold r_member_postAnalytics. Our LinkedIn numbers come from our own browser extension reading your own analytics in your own session, which is the route described in the terms article above, with all the exposure that carries.
That means our coverage has a specific shape, and you should know it rather than discover it. We capture impressions, unique impressions, reactions, comments and shares, and derive engagement rate and a breakdown by reaction type. We do not capture saves, sends, link clicks, followers gained from a post, or profile views from a post. Those five are visible to the official API and not to us.
We also said the wrong thing about this in our own copy until last week. Our engineering notes had it right, with the qualifier that analytics needed Community Management approval. The qualifier got dropped when the claim moved into marketing copy, and it became a flat "LinkedIn gives no analytics API". It was wrong, we have corrected it in four places, and this article is the longer version of that correction.
If you would rather not connect anything, our Claude skills never touch LinkedIn at all, and our MCP connector is where the posting and reading tools live if you do.
What to do next
- If you just want your numbers, export the XLSX from Analytics and tools. It is free and it is better than most people assume.
- If you are building, work out which of the seven endpoints matches the entity you care about before you apply for anything.
- If you are applying for Community Management access, get the company details right first. You get one clean attempt per app.
- If you are evaluating a tool, ask whether its LinkedIn metrics come from the official API or from your browser session. The answer changes what it can show you and what it exposes you to.
- Discount any guide on this topic that does not name an endpoint. It is probably describing 2022.
The question in the title has a clean answer, which is unusual for this subject. Yes, it exists. The interesting part is that existing and being reachable turn out to be very different things, and almost nobody writing about it has noticed either half.
The other platform has the opposite problem. Its API is wide open and meters everything, which we worked through in what the Twitter API now costs.