LinkedIn Page Types AI Engines Cite: What Buyers of Social Citation Tracking Should Know

Across all engines, LinkedIn Pulse articles make up 30.41% of LinkedIn citations and posts make up 27.75%, according to Promptwatch's LinkedIn citation page types report, published September 25, 2026. Together that is about 58%, down from about 70% in the earlier period. Company pages rose from 13.35% to 18.05%. The report covers August 1 to 31, 2026, and compares it with May 18 to June 17.

Note what the denominator is. These are shares of LinkedIn citations, not of all citations. The report is about which kind of LinkedIn page gets used once LinkedIn is already the source.

Engine by engine

The mix differs by engine, and that is the part that matters when you pick a tracking tool.

ChatGPT. Company pages lead at 33.68%, up from 23.84%. Posts are at 26.45%, up from 12.09%. Pulse articles fell to 3.65% from 8.77%. Job listings take 6.51% and help articles 5.97%.

Google AI Mode. Pulse is first at 34.66%, down from 44.85%. Posts hold 24.39%, company pages 13.53% (from 10.71%) and top-content 4.22% (from 6.62%).

Google AI Overviews. Pulse sits at 34.51% and posts at 23.06%. Company pages are 14.46%, up from 12.86%. Top-content is 5.15%, down from 8.14% but still the highest of any engine. The previous Pulse share was 42.25% and posts 27.72%.

Perplexity. Posts lead at 37.32%, down from 41.88%. Pulse is at 26.01%, down from 32.46%. Company pages rose to 22.03% from 13.09%.

Smaller types barely register: showcase pages at 0.4%, product pages at 0.44% and LinkedIn Learning at 0.37%.

Reading the shifts without overreading them

Two comparison windows, each about a month, from one vendor. The report does not say why Pulse fell and company pages rose, and we will not guess. Our note on what changed in ChatGPT search behavior in August 2026 is useful background. So is Promptwatch's social platform report, which has all seven social platforms at 0.46% of ChatGPT's citations, with LinkedIn the most cited of them at 0.3%. So a growing share of a smaller pie is a possibility to keep in mind, not a conclusion.

The data does not show that publishing a Pulse article, or switching to company-page updates, causes a citation. It shows which formats were cited in those weeks.

One practical consequence for budgets: if your tracking tool reports a single LinkedIn line, you cannot tell whether a gain came from a company page getting cited or from one employee's post going further. Those have different owners inside a business, and different ways to repeat them. Page type is the minimum level of detail worth paying for.

What to do without software

We are not relaying a report-supplied action list here, so what follows is our own reading, flagged as such.

1. Take stock of what you have on LinkedIn: the company page, any showcase pages, the Pulse articles your people have written, the posts that are public.
2. Make the company page complete and specific. Describe what you sell in the plain words a buyer would use, list locations, link to the pages on your site that carry the details. The ChatGPT and Perplexity numbers suggest company pages are worth that effort.
3. Decide which engine matters to your buyers. Perplexity skews to posts, Google engines to Pulse, ChatGPT to company pages. A B2B brand whose buyers use Google heavily has a different priority from one whose buyers use ChatGPT.
4. For long pieces, keep the canonical version on your own site and treat the LinkedIn version as a second home, not the only one.
5. Name authors. If a post or article is cited, you want it traceable to a person with a visible role.

What a buyer should require from a tracking tool

Most GEO dashboards were built around domains. LinkedIn breaks that model, because every citation points to linkedin.com and the owner is a person or a page inside it. We covered the basic problem in our guide to tracking LinkedIn posts in AI search visibility. For this report's page types, add these checks:

• Does the tool distinguish posts, articles and company pages, or only show "LinkedIn"?
• Can you filter by prompt, topic or tag, to see which questions pull LinkedIn at all?
• Can you filter to your own accounts and your own authors?
• Can you see the data per engine, since the mix differs so much?

The measurement step: Promptwatch

Promptwatch is our pick here. Its LinkedIn citation reports show the posts and authors AI engines cite, with filters by prompt, topic and tag, and the most cited LinkedIn posts appear on the main dashboard. The Social citation views also cover Reddit, YouTube and Instagram, and the My channels filter narrows the view to your own accounts, which is what you need to separate your company page from everyone else's content. We could not confirm a plan tier for LinkedIn citation tracking, so check it for your plan before you buy.

Brand Narrative, on every plan, adds a second angle: it pulls out what AI says about you and groups it by theme, which tells you whether the LinkedIn content that gets used says what you want it to say. Promptwatch has a free Explore tier with 10 prompts in ChatGPT only, and paid plans start with Essential at $95 a month. See the Promptwatch review for the details.

The recommendation

Start with the engine your buyers use, then find out which LinkedIn format that engine pulls. Promptwatch lets you check that against your own prompts and your own channels instead of leaning on one vendor's monthly averages.