I have always respected The Wall Street Journal. It is the paper you open when you want the number, not the vibe. When the Journal asks a money question, you expect it to come back with money.

So when CMO Today ran a piece on October 5 under a dek promising to explain why marketers like HBO and Spotify think dance parties offer good returns on their investment, I read it the way you read the Journal. Looking for the return.

There isn't one. Not a cost. Not an attendance figure. Not a reach number. Not a lift study. Not a single stream, subscriber, sign-up or viewer. What you get is a party report. And the party report has a sponsor.

Let me be clear about where I'm sitting, because this is not a grudge.

I have had relationships with every advertising editor the Journal has had, right up until the current one. I don't know her personally. We've traded emails, and she has been nothing but cordial and generous, to the point of offering to help with my investigation into sexual assault in this industry. That is not a small offer, and I don't take it lightly.

Julia Angwin, who is a hero of mine, even wrote about me once.

So I came to this piece expecting the Journal I know. Research. Numbers. A reporter who makes the brand prove it. What I got reads like a sponsored recap and an RSVP to the next one.

Megan Graham's piece walks through three activations. LinkedIn and Spotify threw an office-themed rave for DJ John Summit, the former accountant, complete with cubicles, vintage computers and Post-its, timed to an album that dropped around tax season. Summit even updated his old LinkedIn profile with an "Open to Work" badge.

HBO hired Boiler Room to throw a Brooklyn warehouse party for "Industry," its finance drama about hard-partying young Londoners, with timed caviar bumps, Jell-O shots and The Dare going on around midnight. The cast rang the opening bell at the New York Stock Exchange the same day. HBO did another party in June for "House of the Dragon."

Red Bull shows up as the elder statesman, with its old Music Academy and a seven-city Midsummer festival sponsorship.

Now count the evidence offered for ROI.

A content creator says the room was "electric." Another creator says the event was so focused on the music that it might produce "maybe more organic posts." Spotify's Ashley Graver says brands "have to create real experiences." HBO's Mark Doumet calls the events "content generation factories."

That's it. That is the return. Two creators and two executives describing a good night. In the Journal.

Here's the part that made me put the coffee down. The lead case study is a LinkedIn event. And tucked into a parenthetical: LinkedIn is the sponsor of CMO Today.

To be fair to the Journal, it disclosed. That's the rule and they followed it. I'm not saying anyone was paid for this story.

I'm saying it reads exactly like one. When the sponsor's activation leads the column, and the proof of return is a partner executive plus two attendees, the reader can't tell the journalism from the activation. And that matters more at CMO Today than almost anywhere. Its readers are the people who sign the checks for these parties. They came for the math. They got a mood board.

A disclosure tells you who's paying. It doesn't excuse you from asking the sponsor the same hard question you'd ask anyone else. If anything, it raises the bar.

The Journal didn't run this as "brands are throwing raves." That would have been a perfectly fine trend piece. It ran it as an ROI story. Return on investment has two halves, and the article supplies neither.

There is no investment. Nobody says what a Boiler Room warehouse night costs, what John Summit's fee was, or what it takes to stock an office set with vintage computers in lower Manhattan.

There is no return. Nobody says whether "Industry" viewing moved, whether Summit's album streams rose beyond the people already in the room, or whether a single person signed up for LinkedIn because a DJ was "Open to Work."

What's left is sentiment, and sentiment from the most self-selected sample imaginable: invited guests, many of them creators whose business is posting about events they were invited to.

I produced the official party program at MediaPost's OMMA for years and the official dinners at ad:tech. I built the venues, the guest lists and the seating charts myself.

And I should say this plainly: I have invested in nightclubs. One of my investments, Pink Elephant, brought the bottle service system to the United States. I still own parts of several club companies based in Las Vegas. My investments haven’t made me money lately.

So let me be the first to say it. This trend is good for me. Every brand that decides a rave is a marketing strategy is money flowing toward rooms like the ones I have a stake in. If I were writing for my wallet, this column would tell you brands should throw more raves, book bigger DJs and buy out the whole bar. Tables for everybody.

That's exactly why I'm not writing that column. I know two things about rooms like this. I know exactly how good they feel at midnight. And I know that feeling is not a metric.

Here's what the club business taught me. The house always knows its number. Covers, bottles, tables, the night's take, down to the dollar, before the lights come up. The brand paying for the night is usually the only party at the party without a number.

A guest list is a decision. Once you've made it, the energy in the room mostly tells you how well you made it. It tells you nothing about the people who weren't invited, which is where every brand's growth actually lives.

The sharpest voice in the piece belongs to Toshiki Ohta, a DJ, broadcaster and consultant. He explains why nobody flinches at a Red Bull logo near the booth: Red Bull spent years making real financial and cultural contributions to the scene. Brands that skip that step, he warns, risk looking vampiric.

Read that again. The one credible theory of return in the article is the opposite of a one-night party. It's sustained investment over years. Which is precisely what a single LinkedIn rave and a single HBO warehouse night are not.

The Journal had the counterargument in its own copy and let it sit there as a vibe check instead of a verdict.

So what's the measurable outcome of this story? As far as I can tell, the next guest list just got one name longer.

The most important sentence in the Journal's brand-rave piece goes right by without a second look. Spotify built the invite list by identifying John Summit's top listeners in the New York area.

That one line does two things. It quietly destroys every ROI claim in the story. And it raises a data question that sits squarely on this beat, which nobody at the Journal asked.

Start with the ROI problem, because it's the simplest.

If you fill a room with people who already stream the artist most, the room will be electric. They'll post. They'll tell a reporter it worked. That proves John Summit can draw John Summit fans. It doesn't prove Spotify gained a listener, kept a subscriber it was about to lose, or moved anyone who wasn't already in the bag.

Every honest measurement person in this industry knows the word for this. Selection. You chose the people most likely to love you, then reported that they loved you. It's the same trick retail media networks get called out for when they take credit for shoppers who were already in the aisle, and ADOTAT has spent a lot of ink on exactly that move.

The real question is always the counterfactual. What would these same people have done without the party? Streamed Summit anyway, almost certainly. That's why they got the invite.

Now the part the Journal skipped.

Spotify took listening history, the most granular behavioral record most people have with any media company, and used it to decide who got into a co-branded promotional event with LinkedIn. That may be entirely within Spotify's terms. It may be exactly the kind of fan reward listeners love. But it is a data use, and it deserves the questions any other data use would get.

What did those listeners agree to when their streaming patterns qualified them for a promo party? Did LinkedIn receive the attendee list, or any data derived from Spotify's targeting? Were attendees' LinkedIn profiles connected to their Spotify behavior in any way, given that the whole event was built around LinkedIn profiles and an "Open to Work" badge? Who keeps the list now?

I don't know the answers. That's the point. Neither do the Journal's readers, because nobody asked.

HBO's Mark Doumet gave the Journal the most honest sentence in the article. He calls these events "content generation factories" and describes two audiences: the people in the room and the internet writ large, reached through the feeds and the algorithms.

That's HBO telling you the deliverable is posts. The warehouse is the set. The guests are the cast. The return is whatever the algorithm decides to do with the footage.

There's nothing wrong with that as a strategy. Plenty of good marketing works that way. But a factory has output, and output gets counted. How many posts? What reach? How much of it came from creators who were invited precisely because they'd post? And how much of that reach was paid amplification wearing an organic costume?

Earned social impressions dressed up as outcomes is the oldest stand-in in this business. It's the same move the industry makes when it calls a viewability rate a sales result. The Journal would never let a public company slide that past it on an earnings call.

Here's the bigger story the Journal walked past.

Brands will fight for weeks over a few basis points of programmatic fees. They'll hire auditors to chase principal media markups. They'll argue about incrementality methodology for CTV until everyone in the room wants to quit. Then they'll spend a fortune on one night in a Brooklyn warehouse and grade it on vibes.

Ask anyone who owns a piece of a club, and I'm one of them. The venue knows to the dollar what the night made. The brand almost never does.

Experiential budgets live in a strange accounting no-man's-land. Sometimes they sit under brand. Sometimes under PR. Sometimes under the artist partnership or the content team. When a line item doesn't have one clear owner, it rarely gets one clear measurement. And when the vendor producing the event is also the one reporting how great it was, you're looking at the brand grading its own homework, just with a fog machine.

I built these rooms for years. The honest answer to "what did the party return" was always harder than throwing the party. That's exactly why a newspaper should ask it.

CMO Today is sponsored by LinkedIn. The Journal disclosed it, in parentheses, in the paragraph that introduced a LinkedIn event as its lead example.

This isn't unique to the Journal. Sponsored newsletters and sponsored desks are everywhere in trade media now, and the disclosure line has become the industry's universal permission slip. You say who paid, and then you write the piece you were going to write anyway.

But disclosure is the floor, not the job. When your sponsor shows up in your copy, the reader needs more skepticism, not less. The sponsor's event should have to clear a higher bar of evidence than anyone else's. In this piece it cleared the lowest one: two creators said it was fun.

ADOTAT doesn't sell placement in its editorial, and it is funded by its readers. That doesn't make us better people. It just means nobody gets a parenthetical instead of a question.

Here's what ADOTAT is putting to the companies, and what any reporter could have asked.

To Spotify: what the Summit event cost all-in; how many invitees came from listener targeting versus Summit's own invitations; whether any attendee or targeting data was shared with LinkedIn; and whether Spotify ran any brand-lift or streaming-lift measurement against listeners who weren't invited.

To LinkedIn: whether it received any attendee data; what it measured; and whether any sign-ups, profile activity or engagement can be attributed to the event.

To HBO: what the "Industry" and "House of the Dragon" parties cost; how many posts and how much reach the "content generation factories" produced, and how much of that was paid; and whether viewing moved among people who saw the content but never attended.

To Megan Graham and the Journal: whether any of the brands provided cost or measurement data that didn't make the piece, and why a story framed around ROI ran without a return figure.

If the brands decline, that's the story. Three of the biggest consumer brands in the country throw raves and won't say whether they work. That's a Journal headline. "Brands think raves offer good returns" is a press release with better lighting.