LinkedIn Ads data differs from Google and Meta in one significant way for B2B advertisers: professional demographic breakdowns tied to job title, company, and seniority. This note covers what those fields provide, their constraints, and how they affect pipeline and modeling decisions.
What No Other Platform Has
LinkedIn’s adAnalytics endpoint provides the standard metrics you’d expect — impressions, clicks, spend, video completions, conversions. What it also provides, uniquely, are professional demographic breakdowns:
MEMBER_COMPANY— which companies your ads are reachingMEMBER_JOB_TITLE— job titles of viewersMEMBER_JOB_FUNCTION— function (engineering, finance, marketing, etc.)MEMBER_SENIORITY— individual contributor, manager, director, VP, C-suiteMEMBER_INDUSTRY— industry verticalMEMBER_COMPANY_SIZE— company size band
No other ad platform has this. Google and Meta can tell you about someone’s browsing behavior, purchase history, demographic signals, and inferred interests. They cannot tell you that your ads are reaching mid-market IT directors in the financial services industry. LinkedIn can, because it’s built on professional identity data that members actively maintain.
For B2B companies doing account-based marketing (ABM), where the question isn’t “how many clicks did we get?” but “which companies are we getting in front of, and are they the right ones?” — this data is qualitatively different from anything available elsewhere.
The Data Comes With Caveats
The demographic pivots are real, but they come with constraints that matter for how you interpret and model the data.
Privacy noise. LinkedIn applies approximation to protect individual privacy. The values are accurate at a population level but imprecise at the individual level. Treat demographic counts as directional signals, not exact measurements.
Top 100 only. For any given creative on any given day, LinkedIn returns the top 100 companies, job titles, or seniority values — not the complete distribution. If you’re running ads that reach 500 distinct companies in a day, you get data on the top 100 by impressions. The tail is invisible.
Minimum threshold. A demographic value doesn’t appear at all if there are fewer than 3 events tied to it. Small accounts running narrow targeting may get less demographic data than they expect.
Retention limits. Standard performance data (impressions, clicks, spend) is retained for 10 years. Professional demographic data is retained for 2 years. Daily granularity for any data is only available for 6 months — after that, LinkedIn auto-rounds to monthly aggregation. Build your pipeline to capture daily demographic data within that window; you can’t backfill beyond it.
These constraints mean demographic analysis requires explicit documentation of what the numbers represent. A “company impressions” count isn’t “every company who saw the ad” — it’s “the top 100 companies by impressions who saw the ad, for days where at least 3 members from that company saw it, approximated for privacy.” Stakeholders who don’t know this will draw incorrect conclusions.
The CTR Paradox in B2B
One finding that changes how you should think about LinkedIn optimization: click-through rate has a negative correlation with pipeline generation in B2B contexts.
This sounds counterintuitive. CTR is a standard proxy for ad engagement across every platform. But the mechanism makes sense when you think about who clicks on LinkedIn ads. Decision-makers at target accounts — the people who could actually become customers — often don’t click. They see the ad, recognize the brand, maybe bookmark it mentally, and continue scrolling. Junior employees doing research click more. Competitors click. People outside your ICP click.
A high CTR on LinkedIn often signals you’re reaching an audience that’s interested but not in a position to purchase. A lower CTR on ads targeted at senior decision-makers at specific account types can drive more pipeline even though it looks worse in a standard performance dashboard.
The implication for your analytics: measuring LinkedIn performance the same way you measure Google Search (CTR, CPC, conversion rate) will mislead you. The metrics that matter most are company-level reach within your target account list, share of voice among a specific seniority band, and brand awareness growth at accounts you care about. These are harder to measure but more predictive of B2B pipeline outcomes.
Company Intelligence and the Third-Party Gap
LinkedIn’s Company Intelligence API, launched November 2024, promises to connect paid impressions to company-level CRM pipeline data — giving you a direct line between “we showed ads to companies at this stage” and “those companies became opportunities.” This is the ABM analytics dream.
The catch: it’s only accessible through certified partners (Channel99, Dreamdata, Factors.ai, Fibbler, Octane11). There’s no direct API access. If you want company-level attribution linking LinkedIn impressions to your CRM, you’re routing through one of these tools regardless of what your LinkedIn API access looks like.
For most teams, this means the warehouse-centric analytics you build from the adAnalytics endpoint gives you demographic reach data, and the company-level attribution story lives in a third-party tool that you might or might not bring into the warehouse via a separate data feed.
The 90-Day Attribution Window
LinkedIn’s default conversion attribution window is 90 days — significantly longer than Meta’s 7-day post-iOS 14 default and Google’s 30-day default for search. A LinkedIn ad click can claim attribution credit for a purchase that happens three months later.
This long window reflects how B2B buying actually works. Enterprise software deals don’t close in a week. A decision-maker sees your LinkedIn ad, discusses it with their team, puts it in a budget cycle, and might not complete a purchase for a quarter. The 90-day window tries to capture that reality.
The pipeline engineering consequence: conversion data is mutable within the 90-day window. A conversion that appears to have zero attributed conversions today may show attributed conversions 60 days from now as clicks from earlier in the window result in purchases. Your extraction pipeline needs a lookback mechanism that re-pulls recent data to capture these late-arriving attributions. See LinkedIn Ads Analytics Endpoint and Late-Arriving Data and the Lookback Window Pattern for the implementation details — a dbt incremental model with insert_overwrite on a 90-day lookback window is the standard pattern.
What Goes Into Your Cross-Platform Model (and What Doesn’t)
When you’re building a cross-platform reporting model that includes LinkedIn alongside Google and Meta, only a subset of LinkedIn’s data belongs in the unified layer.
In the unified layer: clicks, impressions, spend, external_website_conversions, one_click_leads (combined into a single conversions field). These are the five metrics comparable across platforms, with documented divergences in definitions.
LinkedIn-specific models only: social actions (likes, shares, comments, follows), viral metrics (organic sharing of paid content), video engagement metrics (starts through completions), and professional demographic pivot data. Forcing these into a cross-platform model where they’d sit as NULLs for Google and Meta rows adds noise without value.
The demographic data especially belongs in LinkedIn-specific mart models that can be queried directly for ABM analysis — which companies are we reaching? which seniority levels are engaging? — rather than bolted onto a unified model that other platforms can’t match.
LinkedIn’s value in a cross-platform analytics stack is in professional reach data (unavailable on other platforms) and in the 90-day attribution window, which captures longer B2B sales cycles. Standard cross-platform CPM comparisons do not reflect these differences.