Marketing Attribution for Live Events: Why Meta, Google and TikTok Each Claim the Same Ticket

Marketing attribution for live events has a problem that most promoters have felt and few have named. Run a show on Meta, Google and TikTok, then open the three dashboards on the Monday after the on-sale. Meta reports 1,900 purchases. Google reports 1,400. TikTok reports 900. Your ticketing system says 1,400 tickets were sold. Add up the platform reports and you get 4,200: the show sold 300 percent of its tickets.

None of the three platforms is lying. All three are measuring exactly what they were built to measure. The problem is that what they measure is not what you need to know. This article explains how platform attribution works, why it overstates results more in live entertainment than in almost any other category, and how promoters can get to the one number that cannot be argued with.

What marketing attribution actually means

Attribution is the process of deciding which marketing touchpoint gets credit for a sale. Every ad platform runs its own attribution model, and every model answers a slightly different question.

Meta answers: did someone who saw or clicked our ad buy a ticket within the attribution window? The default window is seven days after a click and one day after a view. Google answers a similar question with its own windows and, increasingly, with modeled conversions that fill in the gaps where tracking is blocked. TikTok does the same with its own windows and its own model.

Each platform only sees its own touchpoints. Meta does not know that the buyer also searched the artist's name on Google. Google does not know that the buyer watched three TikToks about the show the week before. So when the same person touches all three and buys one ticket, all three count that ticket. One sale, three reports, 300 percent.

Why live entertainment gets hit harder than e-commerce

An online shop selling sneakers has the same overlap problem, but the effect is smaller. In live entertainment, four structural factors turn a rounding error into a systematic distortion.

Demand arrives in spikes. A tour announcement, a line-up drop or a presale creates a wave of buyers who were going to buy anyway. Any ad that runs during that wave gets credited for sales it did not cause. The artist's own brand does most of the selling, and the platform report quietly claims it.

The buyer touches everything in a short window. The path from announcement to purchase is often two to five days. In that window, a fan sees the artist's post, a Meta ad, a TikTok, a Google search result and a newsletter. Every touchpoint sits inside every platform's attribution window at the same time.

View-through credit is generous. A fan who scrolls past a video ad without stopping and buys two days later because a friend sent the link counts as a view-through conversion. In a category where fans are already primed to buy, view-through inflates every report.

Tracking on the ticketing side is incomplete. Many promoters sell through third-party ticketing shops where they cannot place their own pixel, or where cookie consent and browser restrictions block a large share of the signal. Platforms then fill the gap with modeled conversions, which are estimates, not tickets.

The result is a set of reports that are internally consistent, individually plausible and collectively impossible.

The cost of believing the reports

The distortion would be harmless if it stayed in the dashboard. It does not. Budget decisions follow the reports.

If Meta shows a return of 12 and TikTok shows a return of 4, the natural move is to shift budget toward Meta. But if Meta's number is inflated by demand that the announcement created, and TikTok's number is understated because its buyers convert later through search, the shift moves money away from the channel that was actually creating new demand. The show still sells, the report still looks good, and the promoter never learns that a different split would have sold more tickets for the same money.

Over a season of forty shows, this is not a rounding error. It is a structural misallocation that repeats on every show night, in every market.

The one number that cannot be argued with

There is exactly one figure in the whole chain that nobody can dispute: the number of tickets sold in the ticketing system. Not purchases reported by a platform, not modeled conversions, not clicks. Tickets, with a timestamp, a price and a seat.

Everything else is an opinion about that number. This is the foundation of ticket attribution as opposed to platform attribution: start from the box office and work backwards, instead of starting from the ad platform and hoping the sum makes sense.

Working backwards from the box office changes three things:

Overlap disappears by construction. A ticket can only be counted once, because it only exists once.
The baseline becomes visible. When ticket sales are read as a curve over time, the demand that the artist, the announcement and the presale generate on their own shows up as a baseline. Campaigns can then be judged on what happens above that baseline, not on the whole curve.
Channels can be compared on equal terms. If every channel is measured against the same source of truth, the comparison is fair. Meta versus TikTok is no longer Meta's model versus TikTok's model.

How ticket-level attribution works in practice

Ticket-level attribution connects the ticketing system directly to the campaign data and matches sales at the level of individual tickets rather than aggregate platform reports. In practice this involves four steps.

Connect the ticketing system. Sales data flows in live from Ticketmaster, Eventim, See Tickets, DICE, vivenu, Eventbrite or whichever system the promoter runs. The ticket record, not the platform report, becomes the primary source.

Establish the demand curve. Every show has a sales curve from announcement to doors. Comparing it to thousands of comparable shows (same genre, venue size, market, price level, time of year) makes the expected shape visible. This is the baseline.

Match campaigns to movement in the curve. Campaign activity by channel, audience and creative is laid over the sales curve. Movement above the baseline while a campaign runs, in the markets where it runs, is the signal. Platform-reported purchases become a secondary, diagnostic input rather than the headline.

Report verified sales only. The number that reaches the promoter is tickets sold, verifiable at their own box office. Any figure the promoter cannot check against their own ticketing system is not reported as a result.

This is how NYBA measures every campaign it runs. Platform reports are used internally to steer creative and audience decisions. What the promoter sees is real ticket sales from their own system, at ticket level, and nothing else.

What a promoter can do this week

Even without new infrastructure, three habits reduce the damage of the 300 percent problem immediately.

Put the ticketing number on the same page as the platform numbers. If the platforms together claim more tickets than the box office sold, the reports are wrong by at least that ratio. Write that ratio down for every show. It is usually between two and four.

Read the sales curve before the campaign report. Ask when the tickets were sold, not which ad claimed them. A spike on announcement day belongs to the artist. Movement in week four while a campaign ran belongs, at least in part, to the campaign.

Stop comparing channels on their own reports. Compare them on cost per verified ticket over the same period in the same market. This is the only comparison that survives contact with the box office.

Frequently asked questions

Why do Meta and Google both claim the same conversion?
Each platform can only see its own touchpoints. If a buyer clicked a Meta ad and later clicked a Google ad, both platforms see a click followed by a purchase inside their attribution window, and both count it. Neither platform has access to the other's data, so neither can deduplicate.

What is the difference between platform attribution and ticket attribution?
Platform attribution starts from the ad and asks whether a purchase followed. Ticket attribution starts from the ticket sold in the ticketing system and asks what moved the sales curve. The first can overcount without limit; the second can only count each ticket once.

Is it possible to deduplicate the platform reports manually?
Only roughly. Shortening attribution windows and switching off view-through credit reduces the overlap, but it does not remove the demand the artist and the announcement create. Only a baseline from real ticket sales does that.

Does this mean ads do not work for live events?
No. It means the platforms cannot tell you how well they work. Measured at the box office, campaigns for concerts, festivals and shows routinely move tickets above the baseline. The point is to measure that movement with a number the promoter can verify, not with a number the platform reports about itself.

Which ticketing systems can be connected for ticket-level attribution?
NYBA connects directly to Ticketmaster, Eventim, See Tickets, DICE, vivenu, Eventbrite and more than 20 further systems, as well as promoters' own shops.

Measured at the box office, not in the dashboard

Attribution is not an academic question for a promoter. It decides where the next euro goes on the next show. The platforms will keep reporting 300 percent because that is what their models produce. The only defense is a number they do not control: tickets sold in your own ticketing system.

If you want to see how your last season looks when it is measured at the box office instead of in three dashboards, book a demo. One event, clean numbers, then you decide.