Industries / Media, arts and recreation / Media and entertainment / Concert promotion
Booking and tours
This work signs artists to play, agrees what they will be paid and plans the order of cities on a tour.
5
Milestones achieved, from 1850 to 2026.
3
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1850
A star on a guarantee
On 9 January 1850 Barnum's agent, John Hall Wilton, signed Jenny Lind to sing 150 concerts in America within about 18 months, for $1,000 a night plus all her expenses, and a fifth of the profits once Barnum had recovered his outlay and cleared £15,000 after 75 concerts. Barnum carried the risk before a ticket was sold. The tour ended by agreement after the 93rd concert on 9 June 1851, and Barnum's accounts record $208,675.09 paid to Lind.
LessonA guaranteed fee moves the risk of empty seats from the artist to the promoter.
Credit
P. T. Barnum; Jenny Lind
1964
Agents back local promoters
Frank Barsalona, who had worked on the Beatles' first US dates at the General Artists Corporation, started Premier Talent, an agency for rock acts. Where older agencies, by his account, got what they could out of young promoters and moved on when they failed, he promised local promoters who helped break a new act, and made competitive offers, that he would keep bringing them that act. He signed British acts including The Who, Jimi Hendrix and Led Zeppelin, and Pollstar wrote in 1988 that he was "universally credited" with building the network of major local promoters.
Credit
Frank Barsalona and Barbara Skydel, Premier Talent
1996
Promoters bought up
In 1996, the year the Telecommunications Act deregulated US radio, the radio owner Robert Sillerman's SFX bought the New York promoter Delsener/Slater. Over the next three years it bought regional promoters including Bill Graham Presents ($65 million), Pace Concerts ($130 million), Don Law ($71 million) and Cellar Door ($106 million); by 1999 it had spent more than $2 billion and owned or operated 120 venues. In February 2000 it agreed to sell itself to the radio group Clear Channel, whose chief executive cited "the natural relationship between radio and live music events", and the business became Live Nation. Pollstar gives the price as $4.4 billion; Playbill, on the day, valued the stock swap at $2.7 billion to $4 billion.
Credit
Robert F. X. Sillerman (SFX); Lowry Mays (Clear Channel)
2010
Ticketing joins promotion
The US Justice Department let Ticketmaster merge with Live Nation, the largest promoter, on conditions: Ticketmaster sold its Paciolan ticketing business to Comcast-Spectacor, licensed its Ticketmaster Host software to the rival promoter AEG, and accepted rules against retaliating against venues; 17 state attorneys general joined. In 2019 the department said Live Nation had "repeatedly and over the course of several years" broken those rules, including by retaliating against venues for using another ticketing company, and in January 2020 the decree was extended by five and a half years with an independent monitor.
LessonRules on conduct attached to a merger proved hard to police, by the Justice Department's own account.
Credit
US Department of Justice, Antitrust Division; Live Nation; Ticketmaster
2026
A jury finds monopolies
On 15 April 2026 a federal jury in Manhattan found unanimously that Ticketmaster held a monopoly over live-event ticketing and that Live Nation monopolized large amphitheaters and tied their use to its promotion services. The states said Live Nation owns, operates or exclusively books 78% of the 87 large US amphitheaters, and jurors found fans had been overcharged $1.72 a ticket. The Justice Department had settled five days into the trial, and 34 states and the District of Columbia took the case to verdict on their own. Live Nation has recognized $450 million against the verdict and the states that settled, and has asked the judge to overturn it.
LessonA federal settlement did not end the case, because the states could carry it on alone.
Credit
the plaintiff states, represented at trial by Jeffrey Kessler of Winston & Strawn; Judge Arun Subramanian
Watch for
A settlement awaiting a judge
The Justice Department and Live Nation agreed terms on 9 March 2026: a $280 million fund for the states' damages and penalty claims, and, as Politico reported via TheWrap, opening Ticketmaster to rival sellers, cutting its long exclusive venue contracts to four years and letting venues give a share of tickets to competing platforms. The judge called the late notice of the deal "entirely unacceptable", and the US National Independent Venue Association said it "does not appear to include any specific and explicit protections for fans, artists or independent venues and festivals".
Achieved whenThe court approves it under the Tunney Act, the law that requires a judge to find an antitrust settlement in the public interest, and a published compliance report shows the terms in use.
Claimed by
US Department of Justice and Live Nation, March 2026; the Proposed Final Judgment was filed on 12 June 2026
A breakup sought
The states that won at trial are seeking remedies that could include a sale of Ticketmaster, Live Nation selling amphitheaters, limits on exclusive contracts, damages and civil penalties. On 3 June 2026 the judge paused discovery on remedies, the exchange of evidence between the sides, until he rules on Live Nation's motions to overturn the verdict, which were argued in July 2026 and are still pending.
Achieved whenThe court orders remedies and they survive appeal to the US Court of Appeals for the Second Circuit.
Claimed by
34 states and the District of Columbia, May 2026
AI reads deal memos
Quantizr, started by Jeremy Salken, drummer and business manager of the band Big Gigantic, with Joshua Litwack and Tim Nielsen, says it automates "the extraction of critical financial and deal terms from deal memos, contracts, and settlement statements" for artists, managers and agents. Its launch gives no accuracy figures.
Achieved whenAn outside test publishes how often it extracts terms correctly from real or realistic contracts.
Claimed by
Quantizr, November 2025
Open
01
Route slack, measured
A public table comparing how far real tours drive with the shortest route through the same cities. Take each tour's dates and cities from setlist.fm, a crowd-sourced database of concerts, and compute the road distance driven. Then compute the shortest route through the same cities in the same span of days, holding fixed dates such as festivals in place, and report the gap in kilometres and in tonnes of CO2 from the trucks. The gap is an upper limit, since real routes also depend on which venues are free on which nights.
The briefCredit · unclaimed02
Deal memos, tested
A public test of whether AI models read concert offers correctly. Write 100 sample offers in the forms agents and promoters use, from email text to PDF offer sheets and scans. Cover flat guarantees, "versus" deals that pay the artist the greater of a guarantee or a share of the net box office, and door deals that pay a share of ticket money after costs. For each, work out by hand what the artist is owed at three attendance levels, with a second person checking, then ask AI models to extract the terms and compute each payout. Use no real contracts and no real money.
The briefCredit · unclaimed
Open 01 · the brief
Route slack, measured
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table comparing how far real tours drive with the shortest route through the same cities. Take each tour's dates and cities from setlist.fm, a crowd-sourced database of concerts, and compute the road distance driven. Then compute the shortest route through the same cities in the same span of days, holding fixed dates such as festivals in place, and report the gap in kilometres and in tonnes of CO2 from the trucks. The gap is an upper limit, since real routes also depend on which venues are free on which nights.
Why it is still open
MIT's Climate Machine report on live music, covering more than 80,000 US and UK events, finds that trucking is about 14% of US live-music emissions once fans' travel is left out, and names "routing efficiency" as a way to cut them, but its coverage gives no measure of how much distance real tours could save. This research found no published study comparing real tour routes with the shortest ones available.
Built with what exists
The setlist.fm API, free for non-commercial projects, which searches concerts by artist, city, country, date and year; OSRM, open-source route planning on OpenStreetMap road data, for distances; Google's open-source OR-Tools for the route search; the UK government's published greenhouse gas conversion factors for heavy goods vehicles.
First step
Pull five 2025 tours from the setlist.fm API and compute their road distance with OSRM.
Done looks like
50 tours from 2025 with at least 20 shows each, in a public table giving each tour's driven distance, the shortest distance under the same fixed dates, and the gap in kilometres and tonnes of CO2, with the code published so anyone can rerun it.
Open 02 · the brief
Deal memos, tested
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public test of whether AI models read concert offers correctly. Write 100 sample offers in the forms agents and promoters use, from email text to PDF offer sheets and scans. Cover flat guarantees, "versus" deals that pay the artist the greater of a guarantee or a share of the net box office, and door deals that pay a share of ticket money after costs. For each, work out by hand what the artist is owed at three attendance levels, with a second person checking, then ask AI models to extract the terms and compute each payout. Use no real contracts and no real money.
Why it is still open
Quantizr says it automates the extraction of deal terms but publishes no accuracy figures, and booking software such as Gigwell's free settlement-sheet maker does the arithmetic only once a person has entered the figures. This research found no public test of AI models on concert deal terms.
Built with what exists
pdftotext, and a vision-capable AI model to read scans; AI models built for repeated extraction at low cost, such as Claude Haiku 5.5 or Meta's Muse Ember (see the Note "The jobs agents repeat"); each offer run 20 times per model and scored on whether it is right every time (see the Note "All twenty times"); a spreadsheet of hand-worked answers.
First step
Write ten offers of each deal type and work out their 90 payouts by hand.
Done looks like
100 offers, at least three AI models and 20 runs each, with a public table of the share of offers whose payout is right at all three attendance levels on all 20 runs, broken down by deal type.