Between 3 and 7 August 2026, six Spanish festivals published their closing figures in the same place: the Portal de la Música en Vivo, the news outlet of the Asociación de Promotores Musicales (APM), the promoters' trade association. None of the six releases explains how those people were counted, and in none of them is there a person with a name and a job title standing behind the number.
This is not a reproach aimed at anyone in particular. It is the circuit itself: the promoter issues the figure and the trade association publishes it under its own editorial byline. In none of the six releases does the word audit appear, nor licensed capacity, nor access control, nor ticketing report. What follows sets the six side by side, measured by the same yardstick, and asks what holds them up.
One thing needs saying before going any further: we published one of those six ourselves. FAR València and its 45,000 attendees appeared in this magazine, and what interested us at the time was not the headline but the 47.5% of buyers from outside the city. FAR's figure is self-declared too, and we did not verify it either. This article is not an exception to that practice. It is a way of looking that practice in the face, starting with our own.
The week's six figures
| Festival | Edition | Declared figure | What else the release says |
|---|---|---|---|
| Noches del Botánico (Madrid) | 10th | more than 190,000 | 52 concerts, from 4 June to 31 July 2026 |
| Cádiz Music Stadium | 2nd | more than 76,000 | no event dates |
| Low Festival (Torrevieja) | 1st in the city | 50,000 | 3 days, 4 stages; no dates, no promoter named |
| FAR València | 2nd | 45,000 | 17 concerts |
| SonRías Baixas (Bueu) | 23rd | 25,000 | three days; promoter, PlayPlan; no event dates |
| Santvi Fest (Sant Vicenç de Montalt) | 3rd | 6,000 | four days at the Parc dels Germans Gabrielistes; no event dates |
Three things repeat across all six.
They are signed by the association, not the promoter. All six releases carry the byline "Departamento de Contenidos de APM", the association's content department. None is attributed to the company that organised the festival.
Four of the six do not say when the festival took place. Those are Cádiz Music Stadium, Low Festival, SonRías Baixas and Santvi Fest. The Santvi Fest deserves a caveat: the gap is in the release, not the festival. The dates, from 31 July to 3 August 2026, are published on the festival's own website, and they are the ones this article uses. Noches del Botánico is the clearest exception: it gives the full window, from 4 June to 31 July. Low's release also fails to name the promoter; SonRías Baixas does, PlayPlan.
None says where the number comes from. The word audit appears in none of the six. Nor do licensed capacity, access control or ticketing report. And no one is identified as answerable for the figure: the closest any of them comes to an attribution is "the organisation".
What stands behind an attendance figure
On the basis of what has been published, nothing. And it is worth saying precisely what that means, because it does not mean the figures are false. There is no evidence that they are, and that is not what is being argued here. What is being argued is that, as published, they cannot be checked: a reader, a sponsor, a public authority or a competitor has nothing to test them against.
For an attendance figure to be checkable, three things would be needed, and all three already exist in most productions:
- 1The venue's licensed capacity. This is the ceiling. Without it, no number has any scale: 6,000 people could be a historic sell-out or a third of the space, and from the outside there is no way to tell.
- 2The access-control count. Turnstiles, scanners or a manual count at the gate produce a number. That number exists by the following Monday.
- 3The ticketing provider's report. Tickets issued, tickets validated, invitations, accreditations. It is the document a promoter uses to settle up with their own team.
A fourth thing is missing, and it is not a document but a clarification: the unit. When SonRías Baixas declares 25,000 attendees over three days, is that 25,000 distinct people or 25,000 attendances added together? When Noches del Botánico declares more than 190,000 across 52 concerts, how many of those attendees went to two concerts? None of the six releases clears this up. Two festivals with exactly the same audience can publish figures that differ by a factor equal to the number of days, depending on which of the two conventions they use. There is no common convention in the sector, so the six figures in the table above are not strictly comparable with one another. They have been placed in the same column, and that is the most that can be done with them.
There is a quick way to test this: add them up. Stacked one beneath the other, the six figures come to more than 392,000 attendees. That number means nothing. Noches del Botánico accumulates 52 concerts spread between 4 June and 31 July; SonRías Baixas counts three days and the Santvi Fest four; and none of the six says whether it is counting distinct people or tickets. Adding them up is what everyone does with these headlines, and it is exactly what cannot be done.
Here there is a nuance that usually gets overlooked: at festivals with paid entry, the number exists and it is exact. The Santvi Fest sold general admission at 8 euros in advance and 10 on the door, according to its official website. That means there is a sales system and an entry count behind it. The problem is not that the figure cannot be known. It is that the result is published without the method.
And who is asked for the method? None of the six releases points to a third party that has checked the number, nor refers to any document. Every year the APM publishes the Anuario de la Música en Vivo, its yearbook with the sector's turnover, which is the market's reference document, but end-of-edition releases go by another route and do not pass through it.
The Santvi Fest case
Of the six, one shows the whole mechanism in its cleanest form.
The APM release of 7 August 2026 says: *"The third edition of the Santvi Fest brought together 6,000 attendees over four days at the Parc dels Germans Gabrielistes in Sant Vicenç de Montalt."* The figure is stated flatly. It is attributed to no one.
It is not the first time that number has appeared. Nor the second. All three editions of the festival have closed with the same published figure, and all three times in almost the same words.
On 5 August 2024, after the first edition, the festival's website published: *"The festival brings together more than 6,000 attendees over four days"*.
On 4 August 2025, after the second: *"The event brings together more than 6,000 attendees over four days"*.
On 7 August 2026, in the APM release: *"6,000 attendees over four days"*.
What repeats is not just the figure. It is the entire construction: six thousand attendees, four days. Three years running, in the first week of August, changing only the subject of the sentence: "the festival", "the event", "the third edition". That is not an impression or a suspicion: these are three published, linkable texts, and they can be read one after the other.
There is also a matter of wording that is not minor. Both times, the festival wrote "més de 6.000", more than 6,000, which is a floor: it means six thousand or more. This year's APM release writes "6.000" on its own, which reads as an exact figure. The same quantity changes its nature depending on the sentence wrapped around it, and none of the three texts says which of the two readings is the right one.
In 2025, though, there was a second source. The Ajuntament de Sant Vicenç de Montalt, the town council, put out its own round-up on 6 August, under the headline *"Santvi Fest closes a successful second edition with more than 6,000 attendees"*, with a quote from the mayor: *"We are very satisfied with the public's response..."*. Same figure, same "more than", a different website and a person with a name behind it.
It is worth being exact about what this proves and what it does not. It does not prove the figure is false. A venue with a stable capacity and an identical format — four days, same park, same ticket model — can repeat a figure without any mystery, and it would be odd if it did not. What is documented is something else: that the number and its packaging repeat identically three years running, and that none of the three texts says where it comes from. Those are two different claims, and only the second is being made here.
Nor is there any way to give it scale. The capacity of the Parc dels Germans Gabrielistes is not published in any of the sources consulted, so there is not even a way to know whether 6,000 is a sell-out, half the venue or a figure for attendances accumulated over four days.
What has changed this year is who is saying it. As of 14 August 2026, the APM release is the only place the third edition's figure has been published. The festival has not issued a closing release: the latest item in its news section is dated 18 June 2026. Nor has the town council: its municipal news from 2 to 13 August 2026 is all about the town's Festa Major celebrations, and none of it mentions the festival.
That silence carries weight because there was none last year, and because the town council is not a neutral third party that happened to be passing by: it declares itself co-organiser, in its own words, alongside "the Barcelona promoter TheProject". In 2025, the public co-organiser confirmed the figure with a round-up of its own and a quote from the mayor. In 2026 it has said nothing. We do not know why, and we are not going to guess. That is the fact.
The rest of the edition is documented, and by the festival itself. It took place from 31 July to 3 August 2026. General admission cost 8 euros in advance (plus fees) and 10 on the door, with reduced prices for children and families. The line-up featured Gemma Humet, Gessamí Boada, Irene Cortés and Palmira Yelo in the afternoons, and Arrels de Gràcia, Tonia Richii, Las Karamba and Clarence Bekker Band in the evenings. It is organised by the Barcelona promoter The Project — the footer of the festival's website reads "© 2026 Paral·lelos Musicales" — and the town council, the Generalitat de Catalunya and the ICEC are listed as collaborators.
Put together, those pieces describe the model well: a local festival with an 8-euro ticket, a town council as co-organiser, a promoter that belongs to the APM and an attendance figure that this year appears only in that association's outlet. None of those things is objectionable on its own. Together they describe rather well how the story of a summer takes shape.
Who signs
The APM brings together more than 100 members and claims to represent "more than 80% of private promoters", according to its own self-description.
It is also a media outlet. The Portal de la Música en Vivo publishes news, and that is where the six August releases appeared, signed by its content department. The portal also has a separate category, "Comunicados", for press releases, and hosts author profiles for member promoters. The Santvi Fest release did not go that route: it was published as news, under an editorial byline.
The Project, the Santvi Fest's promoter, is an APM member and has its own author profile on the portal, with a corporate biography and articles of its own published there. It therefore had a channel for signing its own texts. Its festival's closing release did not go out that way: it went out as news, signed by the editorial team of the association it belongs to.
Set out in order, the circuit looks like this: a member promoter organises a festival, the association it belongs to publishes that festival's attendance figure as its own editorial content, the release points to no third party that has checked it, and the public co-organiser does not confirm it. Every link is legitimate and none hides anything: the author profiles are public, the press-release category exists and the APM's self-description is on its website. The problem is not one of conduct. It is structural. When the issuer, the distributor and the beneficiary of a figure all belong to the same circuit, there is no point along that route where the figure is put to the test.
One note, without reading more into it than there is: publishing a figure without a method is not unique to this portal either. Brunch Electronik Festival's 65,000 attendees were published the same way, as the organisation's figure and with no explanation of where it came from. That is a one-off, not a pattern: one case proves no general rule, and we are not going to pretend otherwise. What is particular about the Portal de la Música en Vivo is that it concentrates many of these releases in one place, and that makes it possible to see them together. Which is exactly what we have done here.
What to do with this
If you are a promoter, this is not about criticising anyone. It is about making sure that the next time you publish a figure, it is one of the few this summer that someone can actually check. It is a cheap competitive advantage: none of the six releases in this batch does it.
- Publish the unit, not just the number. "18,000 attendances over four days" and "18,000 distinct people" are different sentences, and only one of them describes your event. Pick a convention, write it into the release and do not change it from one year to the next.
- Publish where the number comes from. One line is enough: access-control count, tickets validated by the ticketing provider, the venue's licensed capacity. If your figure includes invitations, accreditations or staff entries, say so.
- Put a name and a job title behind it. "According to the organisation" is not an attribution. It is a way of not attributing. Have the figure signed by the festival director or the head of production, by name.
- Put the dates in the release. Four of the six releases in this batch leave them out. It is the easiest piece of data in the world, and without it your release can be neither filed nor compared.
- Keep the exported report, not just the headline. The day a sponsor, a public authority or an investor asks for the backing, the difference between having that July's validation file and not having it is several weeks of work.
- If you repeat last year's figure, explain it. A stable capacity with an identical format produces identical figures, and that is a perfectly good answer. Giving it in the release itself saves someone else from asking the question. And if you also repeat the entire sentence three years running, someone will end up lining the three texts up in a column, as here.
- If you have a public co-organiser, sign off the round-up with them. A council that publishes its own round-up turns your figure into a data point with two sources and a named person behind it. That is what happened in Sant Vicenç de Montalt in 2025 and what has not happened in 2026. It costs nothing.
- Before publishing, cross-check your figure against your own sales. Tickets issued against tickets validated, advance against on the door, buyers' postcodes. It is the cheapest analysis a promoter can do, and it usually changes the headline you were about to run.
Six festivals published their figures in five days and none of them published how it had obtained them. Until the method is published, an attendance number has whatever credibility each reader chooses to grant it. The first promoter to add that line will have something none of the six releases has, and it will have cost them one sentence.