LinkedIn published an event-advertising framework on 30 July 2026 stating that "there are 31 times more viewers for company-hosted events promoted by event ads compared with company-hosted events without event ads promotion", given as an average. In November 2025 LinkedIn had reported that "event ads drive 31% more viewership of events on average". PPC Land placed the two claims side by side on 4 August 2026 and found that LinkedIn has published no reconciliation between them.
The honest conclusion is narrower than the headline suggests. PPC Land does not assert that the 31x figure is wrong. Its finding is that the two numbers, in its words, "share a numeral and describe the same mechanic, but they differ by two orders of magnitude in effect size", and that "the document offers no reconciliation, and neither construction is documented in enough detail for an advertiser to reproduce the calculation". That makes the multiple unusable as a planning input. It does not make it false, and nobody outside LinkedIn is in a position to say which construction is the sound one.
What LinkedIn published on 30 July 2026
The framework was published on 30 July 2026 by Jae O., Head of Ads, Formats, Placements, Measurement and Audiences at LinkedIn. Alongside the 31x claim it introduces a second figure: members are "28% more likely to engage with subsequent marketing campaigns after seeing an event ad". PPC Land notes that this describes a carryover effect rather than in-campaign results, and that "no methodology accompanies it". Neither the 31x nor the 28% arrives with a sample size, a measurement window or a stated geographic scope.
The November 2025 figure described the same mechanic at a fraction of the size
LinkedIn's November 2025 wording was that event ads drive 31% more viewership of events on average. Eight months later the same mechanic, promoted events against unpromoted events, is presented as a 31-fold difference. A 31% lift and a 31x lift are not variations on a theme. One says promotion improves turnout at the margin; the other says promotion is responsible for almost all of it. Media plans built on those two premises look nothing alike.
The scopes are not identical, which is the strongest argument for LinkedIn
PPC Land is explicit that the scopes differ. The November 2025 figure covered events generally. The 30 July 2026 figure is scoped to company-hosted events, comparing promoted against unpromoted. A narrower denominator can legitimately produce a much larger multiple, and PPC Land allows for exactly that. What it does not accept is that either figure, as published, contains enough detail for an advertiser to reproduce the calculation. So the scope difference is a plausible explanation rather than a documented one, and LinkedIn has not offered it as an explanation at all.
Every figure here is LinkedIn measuring LinkedIn
The conflict is worth stating plainly. The 31%, the 31x and the 28% are all LinkedIn's own numbers, published by LinkedIn in material that promotes LinkedIn ad formats. There is no third-party verification and no disclosed methodology for any of them. Vendor performance claims are normal in ad tech, but a vendor claim that shifts by two orders of magnitude in eight months, with no correction and no explanation, has no place in a forecast a client will be held to.
The product behaviour is documented, and it is the part worth planning around
Underneath the numbers, the 30 July 2026 framework documents something concrete and verifiable inside the product. Event Ads and off-platform Event Ads "automatically shift from driving registration to driving live viewership" on event day, without the advertiser rebuilding the campaign. That is a real change in how an events buy is structured. Historically the transition from a registration push to an attendance push meant a second campaign, a second set of creative and a manual switch at the right hour, which in practice meant somebody watching a clock. The automatic shift removes that handover, along with the risk of a registration ad still running while the session is live.
The feature set dates to 28 April 2026
LinkedIn dates off-platform Event Ads, in-ad lead generation forms and event clipping to an overhaul on 28 April 2026. Those three are checkable against a campaign build in a way a lift multiple is not. Off-platform delivery extends event promotion beyond the LinkedIn feed, in-ad lead generation captures a registration without sending the member to a landing page, and event clipping turns a session into reusable segments. An advertiser can confirm each of them exists, test what it costs and measure what it returns, which is more than can currently be said for the 31x.
The three claims, side by side
This is the whole argument in one table. The right-hand column is the same in every row, and that is the point.
| Claim | When LinkedIn published it | Methodology disclosed |
|---|---|---|
| Event ads drive 31% more viewership of events on average | November 2025 | None |
| 31 times more viewers for company-hosted events promoted by event ads versus company-hosted events without event ads promotion | 30 July 2026 | None |
| Members are 28% more likely to engage with subsequent marketing campaigns after seeing an event ad | 30 July 2026 | None, and PPC Land states no methodology accompanies it |
| Event Ads and off-platform Event Ads shift automatically from registration to live viewership on event day | 30 July 2026 | Product behaviour, not a performance claim |
What this means for Thai marketers
PPC Land does not say whether Thailand or the wider APAC region sits in the sample behind either figure, so nothing below is sourced to LinkedIn; it is reasoning about how the disclosure gap plays out locally. Thai B2B teams running webinars and industry events on LinkedIn are usually working with small absolute audiences, where a headline multiple drawn from an undisclosed global base is the least reliable kind of input. The safer approach is to plan the buy around the mechanics that are documented, the event-day auto-switch and the April 2026 formats, and to establish a lift figure from a holdout in your own account rather than importing one.
That means deciding in advance what a promoted event has to beat. Run a comparable unpromoted event, or hold back a segment, and read the difference in your own numbers. If event registrations are being handed to sales, the measurement problem sits downstream of the ad platform anyway, which is a case for clean analytics and conversion tracking before the next campaign rather than after it. The same discipline applies to the rest of a paid social programme, and to the content that has to fill the event itself once people register.
Frequently asked questions
Which figure is correct, 31% or 31x?
The source does not state which is correct. PPC Land reports that LinkedIn has offered no reconciliation and that neither figure is documented in enough detail for an advertiser to reproduce the calculation. It stops short of saying either number is wrong.
Does this mean LinkedIn event ads do not work?
No, and PPC Land makes no such claim. The reporting is about disclosure, not about whether the format performs. Event ads may well lift attendance; the point is that the published figures cannot tell you by how much, so the multiple should not carry a forecast.
Has LinkedIn corrected or withdrawn the November 2025 figure?
The source does not state that it has. PPC Land reports both figures as published claims and notes the absence of any reconciliation between them.
Is Thailand or APAC included in the sample behind these numbers?
The source does not state it. No geographic scope is disclosed for the 31%, the 31x or the 28%, so there is no basis for assuming Thai or APAC events are represented, and no basis for assuming they are not.
Do I need to change anything in my existing event campaigns?
Nothing is being forced on advertisers by this reporting. The one behaviour worth knowing about is the documented automatic shift from registration to live viewership on event day, which changes how a campaign should be structured and briefed even though the advertiser does not trigger it.
The reporting comes from PPC Land, published 4 August 2026. The reasonable position for a media planner is neither to repeat the 31x nor to treat LinkedIn event ads as discredited. Use the documented product behaviour to build the campaign, use your own holdout to size the lift, and keep vendor multiples out of the forecast until a methodology arrives with them. If you want a second read on how your LinkedIn events buy is structured and measured, that is a conversation worth having before the next event date, not after it.
