The Disclosure Economy Paper No. 01, The Gathering Architecture Papers Author: David Adler. Publisher: GatheringPoint.news. Free and ungated. Web version: https://gatheringarchitecture.com/papers/the-disclosure-economy This file is the complete plain-text copy of the paper, provided so text-only readers and AI agents can read the full work. ======================================================================== This paper is published without a form, without a download gate, and in plain web text rather than a PDF, on the argument it makes. If the thesis here is right, a document nothing can read is a document that does not exist. The decision to attend — and the decision to exhibit — is about to be made by agents as often as by people. An agent does not browse a site, watch a sizzle reel, or trust a promise. It reads what it can see, compares it against a brief it was given, and moves on when the data runs out. If your event is not visible to an agent, it is not analyzable by one. Some organizers are already building agent-only access to their event websites: machine-readable specifications of who will be in the building, what will happen there, and what it costs to be part of it. This paper is about why that shift is rational, what it changes for attendees and exhibitors, and what to publish before the buyer becomes software. The examples here are drawn from business events — conferences, trade shows, and corporate gatherings — but the same decision machine is forming across ticketed experiences, member programs, and brand activations. The categories are merging. The disclosure problem is not. The question is not only what you publish after the event is over. It is what you publish about the event that has not happened yet. An anticipated event has a manifest too: who is confirmed to be in the building, what will happen there, how the rooms are arranged, what it costs, and how much meeting inventory remains. If that manifest is not visible, the agent cannot analyze the trip against the brief. The organizer who treats the next edition as a specification to be read — not a promise to be sold — is the one the agent can recommend. 01 Two rooms, one question Two conversations, two weeks apart, both with people who have run this business longer than most of the audience has been alive. The first told us the agent thing is nonsense. Buyers do not delegate a five thousand dollar travel decision to software, procurement will never sign off on it, and this ends up in the same drawer as the metaverse trade show and the virtual booth. The second told us that anyone not building for it now is already late, that the purchase order will be written by a machine inside three years, and that the industry's entire selling apparatus is aimed at a customer who will no longer be doing the buying. They cannot both be right. This paper will not tell you which one is, because no one selling you a certainty about it knows either. What it can do is draw the map. This paper lays out how the decision to attend gets made if the skeptics are right, how it gets made if they are wrong, and what an organizer does in the meantime that does not require betting the company on either answer. That last part is the reason to keep reading. The recommendation here holds in both futures. Publishing what actually happened to the people who came last year, in plain terms, before anyone has to ask, is the strongest marketing available to you if the buyer is a human being with a budget and a boss. It is the only thing legible to the buyer if the buyer turns out to be software. Almost nothing else on offer right now has that property. Most of what the industry is being sold is a bet on one future and a write off in the other. The rest of this paper is about which parts of your event you publish, which parts you hold, and why the oldest habit in the business, selling anticipation by withholding, is the thing that breaks first under either scenario. 02 What we are actually talking about The word agent is doing too much work in this conversation, and half the disagreement in those two rooms was people arguing about different objects. Four terms, defined once, then used consistently for the rest of the paper. Chatbot A system that answers questions. You ask it something, it responds, and nothing in the world changes. Most of what the events industry has deployed since 2023 is a chatbot with a registration form attached. Agent A system that takes actions on someone's behalf. It does not answer a question about the flight. It searches, compares, decides within limits it was given, and books. The difference between a chatbot and an agent is not intelligence. It is authority. Someone gave it a budget and a goal and let it act. The brief The instruction the agent is working from. It looks less like a search query and more like a purchase order: attend up to four industry events this year, under twelve thousand dollars total, prioritize contact with these named accounts, avoid the week of the fiscal close. Every claim your event makes is evaluated against a brief you will never see. Payload and manifest Two categories of information about your event, defined properly in section five. The payload is the content. The manifest is the record of who is in the room and what happened to them. Confusing the two is the expensive mistake in this paper. One more thing to name before going further. Nothing in this paper requires a machine that can sit in a ballroom and judge a keynote. That does not exist and may not by the middle of the next decade. What already exists is the layer underneath: systems that hold a budget, compare options against a brief, and complete a transaction. That is enough to change how the decision to attend gets made, which is the only claim this paper needs. 03 If the skeptics are right Take the first room seriously, because its argument is stronger than the enthusiasts admit. Business travel is not a commodity purchase. It is discretionary, political, and personal. It is approved by a manager who is also deciding who gets the trip and who does not, which is a status transaction dressed as a budget line. Procurement does not want a machine's name on an expense it may have to defend. Legal has opinions about software that binds the company to a contract. And the relationship half of this business, the reason a supplier goes to the same show for eleven years running, does not reduce to a specification at all. All of that is true and none of it is going away quickly. But notice what survives even inside the skeptical case. The human buyer is already doing an abbreviated version of what an agent would do, with worse tools. She asks a model to summarize three conferences. She asks a colleague who went last year whether it was worth it. She looks for the attendee list, cannot find it, and guesses. Her decision is still being made on evidence, and the evidence is still thin, and she still resents the sizzle reel. Every recommendation in this paper is something a skeptical CFO would have asked you for in 1994. That is the sturdy part of the argument. Publishing what happened to last year's attendees is not an AI strategy. It is what you would do if a serious buyer with a spreadsheet asked you to justify the trip, which happens every day already and has for thirty years. The agent, if it arrives, simply reads the same document faster and shares it with a thousand other buyers at once. 04 If the skeptics are wrong The second room's case does not rest on a forecast. It rests on plumbing that has already been poured. Both major card networks shipped agentic commerce rails in 2025, which means the payment layer for a machine buying something on a cardholder's behalf is no longer hypothetical. Travel, which is the connective tissue of every event in this industry, went agent native faster than events did, with booking flows running across hundreds of airlines and millions of hotel rooms inside a chat window. None of that was built for us. All of it lands on us, because a conference is a travel decision with a badge attached. Here is what changes on the day the brief starts doing the shopping. The agent cannot be moved by a teaser. It is not curious. It has no fear of missing out, no relationship with your sales director, and no memory of the year the party ran until three. It will not click a countdown clock. Every device the industry has spent forty years perfecting depends on a nervous system the buyer no longer brings to the transaction. What it can do is compute. It will price the fully loaded cost of attendance, which is a number most organizers have never published and many have never calculated: registration, flight, three nights at the block rate or above it, ground, meals, and the salary cost of the days away. It will compare that against the alternatives on the calendar. It will look for the named accounts in the brief and try to determine whether they will be in the building. And it will estimate the cognitive cost of the program: how many sessions are packed into a day, whether the schedule leaves room to think, how far the rooms are from one another, whether the breaks are recovery or just transitions to the next thing. An agent buying on behalf of a busy executive treats attention as a budget line. Where it cannot find an answer, it will not assume the best. Silence does not read as mystery. Silence reads as absence. That is the reversal the industry has not absorbed. For four decades, withholding created value. The unannounced headliner, the surprise on the show floor, the reveal held for the general session. Against a buyer that computes, withholding creates a blank field, and a blank field scores zero. See Appendix AThere is no single buyer and no single brief. Nine distinct reasons to attend produce nine different evaluations, and each one asks your event for a different set of facts. The Nine Briefs sets out the actual language of each, what it reads, and what it breaks in your planning. 05 The split that makes this manageable The obvious objection arrives here, and it is the right one. If we publish everything, why would anyone come. Because there are two categories of information about your event and they behave in opposite directions. The first is payload. Session content, the keynote argument, the slides, the market data, the product walkthrough, the recording. Every piece of payload you publish is an argument that the human does not need to travel, and a machine will make that argument for you, correctly. Publish the payload and you have told the buyer that your event is a content delivery system with a dress code. The second is manifest. Who is confirmed to be in the building. Which of my accounts, competitors, and suppliers are on the list. How much meeting inventory remains and how it gets allocated. Capacity, adjacency, the walk from registration to the hall. What happened to the people who went last year and what they closed. None of that substitutes for attending. All of it is what a buyer needs in order to justify three days and four thousand dollars. You can read a seat map all afternoon and you still have not taken the flight. The manifest has a past tense and a future tense, and most organizers are only comfortable with one of them. The past tense is the box office report: who came, what they did, what it produced. The future tense is the anticipated event itself, published before it happens. Who is confirmed to attend, speak, exhibit, or sponsor. The floor plan and the walking distances it implies. The session architecture, the meeting inventory, the pricing, the capacity that is already committed and the capacity that is still open. One is evidence that the trip was worth it. The other is the specification that makes the trip analyzable. Both have to exist in plain text, at stable addresses, before the buyer decides. The plumbing for this is not theoretical. Vendelux already sells pre-event intelligence about who will be in the building. Exposure Analytics uses WiFi sensors to measure footfall, dwell time, and flow routes on the show floor. Explori standardizes post-event feedback and benchmarks programs against a global dataset. Each one collects a different column of the manifest. What none of them can do is publish it for you: the data still has to leave the vendor dashboard and become visible, at a stable address, in plain text, before the buyer decides. The test If a buyer could read it and reasonably decide not to travel, it is payload. Hold it, or release it only to the people who came. If reading it makes the trip easier to justify, it is manifest. Publish it early, publish it plainly, and do not put it behind a form. The timing follows from the split. Specifications go out early, because early is when the decision gets made. Content goes out late, or only to the room, because that is what makes being in the room the way to get it. The post event archive is not a giveaway when it is written as a record of outcomes rather than a transcript of sessions. That is next year's evidence, which is a different job entirely. There is a harder point underneath the split, and it is the one organizers will not enjoy. What you are able to publish is a confession about where your value actually lives. If the only asset worth putting in front of a buyer is your session content, then the buyer is right and the trip is not justified. Disclosure does not create that problem. It reveals it. 06 The trailer and the box office report There is a meeting that happens at every event company in the six weeks after the show, and it always goes the same way. Marketing wants the keynote recording posted, ungated, for reach. Someone on the content side says yes, our best material should be our best advertising. Nobody in the room asks the question that will matter in ten years, which is what that recording tells a buyer about whether to send anyone next April. It tells them the trip was optional. The film came to streaming three weeks after opening weekend, free, and the buyer noticed. The movie business solved this a long time ago by running two instruments at once and never confusing them. The first is the trailer, which is a sample of the product. It works because two minutes out of a hundred and twenty cannot substitute for the film. Our version is a ninety second cut of a keynote, and it is perfectly good marketing right up to the moment the full ninety minute session goes up for free, at which point the clip stops being a trailer and becomes a table of contents. The discipline is not complicated. Publish the clip, never publish the thing the clip is advertising, and if you intend to release the recording anyway, admit that the session was never the reason to travel and go find one that is. The second instrument is the box office report, and this is the one the events business has never built. Opening weekend is a number the studio does not get to write. It is published, comparable, and checked on Monday morning by everyone deciding what to make next. The equivalent here is evidence about what happened to the people who went last time. How many booked meetings that turned into pipeline inside ninety days. How many conversations the median attendee had that were not available anywhere else. How many came back. What it cost per outcome against the alternatives on the calendar. And what attendees actually wrote when they were asked, analyzed at scale: the complaints that repeat, the praise that clusters, the requests that show up in different words across hundreds of responses. A machine that reads your keynote transcript has a reason to keep everyone home. A machine that reads what happened to fourteen hundred people who look like its buyer has a reason to book the flight. The catch is that evidence has to be about the edition that already happened. Unmissable insights and unparalleled networking is a promise, and a promise about the future cannot be checked, which means it computes to nothing. Only the past is auditable. This industry has spent forty years selling anticipation and almost no time building a record, which is why the shift is uncomfortable. You are being asked to market with your receipts instead of your renderings. Platforms like Explori are already positioned to become the record keeper: standardized feedback, business impact metrics, and cross-event benchmarks. The question is whether the organizer treats that data as an internal report or as a public box office report. A report that stays in the dashboard is an operational tool. The same numbers, published in plain text, become next year's specification. Comments are the newest column of that report. Until recently, open-ended feedback was too noisy to publish. Now it can be summarized by theme, sentiment, and urgency faster than a human can read the full set. The analysis is not the raw comments — nobody wants their quote mined without permission. It is the pattern: the top five friction points, the top five unexpected wins, the gap between what was promised and what was repeated back. A buyer's agent will read that summary the way a product manager reads app store reviews. It is the closest thing to a public conversation about whether the event was worth it. Which returns to the meeting six weeks after the show. The recording was never the asset. What happened to the people in the room was the asset, and nobody wrote it down. 07 The floor plan is a specification too Every general contractor in this business has a version of the same phone call. A week out, an exhibitor asks to move. The island they bought sits in a dead corner of the north hall, they have seen the traffic pattern from last year, and they want out. The answer is almost always no, and not out of stubbornness. The rigging plot is filed, the power is dropped, the drayage is scheduled, and the floor plan went to print in March. That is the part of this business the buyer is about to notice. If the decision to attend is made by reading specifications, the floor plan is one of the specifications that gets read. Where the meeting rooms sit. How long the walk is from registration to the one demo worth the trip. Whether there is anywhere to hold a conversation that is not a hallway. How much hall is committed and how much is still open. Today none of that exists in a form anything can read. The floor plan is a PDF, and a PDF is a photograph of a decision made nine months ago. The floor plan is also a proxy for cognitive load. An agent can read the distance between sessions, the density of the agenda, the ratio of programmed time to unprogrammed time, and the availability of places to recover. These are not amenities. They are specifications about whether the event respects the attention of the person being sent. An agent buying for a burned-out team will score a packed schedule as a cost, not a feature. The behavioral layer is already being measured. Exposure Analytics and similar sensor systems capture footfall, dwell time, and flow routes across the hall. What they produce today is an internal operations report. Published as a machine-readable record of what actually happened, it becomes the evidence a buyer uses to price next year's booth before the next year's floor plan is even drawn. The measurement is also getting closer to the person. Zenus AI reads anonymous, aggregated facial signals for attention, demographics, and positive sentiment in the room. Bishop-McCann's JOY Index, launched in 2024, uses AI and neuroscience methods to score the emotional impact of a program. Dr. Paul Zak's Immersion platform reads biometric signal from wearables to produce a moment-by-moment record of what held a room and what did not. TAWNY and knw. offer comparable vision-based attention and emotion analytics built for halls and show floors. The output is still treated as an internal optimization tool, and in most cases the attendee who supplied the signal never sees the record. The disclosure question is the same: once you can measure what happened in the room, the buyer will eventually ask why the manifest does not include it. The version of the next decade worth planning for is not a hall that rearranges itself while the show is running. A forty by forty island with hung truss and a union crew does not relocate on Tuesday afternoon, and anyone selling that picture is selling a rendering. What becomes responsive is everything around the steel: wayfinding, signage, meeting allocation, session placement, traffic routing, the white space held back on purpose, and the plan for the next edition. The booth stays put. The room around the booth starts paying attention. None of it works unless the floor is published first. An algorithm cannot arrange a room it cannot read. Every dynamic floor plan is downstream of a disclosure decision: dimensions, capacities, dwell time, meeting inventory, sightlines, and an honest answer about how far the walk really is. Which turns this paper's argument back on itself. Publishing the specification is not a concession extracted by the buyer. It is the precondition for the floor getting smart at all. The organizer who publishes because they were forced to will do it late and badly. The organizer who publishes because it is the input to their own operating system will do it first and own the data that comes back. The resistance here is not technical. Drayage costs what it costs because freight on a show floor is a closed structure of contractors, jurisdiction, and exhibitors with nowhere else to go. A published, machine readable floor is a repriced floor, and the people who own the docks will understand that before anyone explains it to them. Expect the digital twin in the sales office for years before it is allowed near the loading dock. Which brings back the exhibitor who called in March. Ten years out, they do not call. Their buyer read the plan before they bought, priced the walk from registration, and either paid up for the corner or went somewhere else. The floor plan stops being the document you defend and becomes the product you sell. 08 What breaks on the operations side Everything above is about how the event is sold. This section is about how it is built, and it is the half nobody has been told. A general contractor, a venue operator, or a production lead will recognize themselves here faster than a marketer will. The planning calendar inverts Today the specification is the last thing to exist. Marketing goes out in the fall, the theme lands in December, the floor plan freezes in March, and the actual detail of who is coming and where they will sit resolves in the final six weeks. Under a disclosure regime that order flips. The specification has to exist first, because it is the thing being sold, and the creative gets built on top of a published frame rather than in front of an empty one. The design brief can no longer freeze Once meeting inventory, capacity, and adjacency are public and updating, a design that was locked in March against assumptions from the previous November is a liability. The brief becomes a living document with a version number, and somebody has to own it, which is a job title that does not currently exist at most event companies. The exhibitor contract becomes a warranty The moment you publish traffic figures, dwell time, and what happened to last year's exhibitors, you have made a representation. Buyers will hold you to it, and eventually the contract will too. That is not a reason to avoid publishing. It is a reason to measure carefully before you do, because the first organizer to publish a number they cannot support will spend a year regretting it. Meeting space becomes a product line If the manifest is what sells the event, then the ability to guarantee a conversation is the highest value inventory in the building. Right now most organizers give that away as an amenity or leave it to the hallway. Priced, allocated, and published, it is a line of business. White space gets published on purpose The unsold hall, the open hour, the unprogrammed afternoon. The instinct is to hide all of it and pretend the show is full. The opposite move is stronger: publish the open inventory the way an airline publishes remaining seats, because scarcity that can be verified is the only kind that still works on a buyer who computes. 09 What to do in the next twelve months Five things, each of which pays off whether or not a single agent ever books a badge. Calculate the fully loaded cost of attending your event and publish it. Registration, flight, three nights, ground, meals, and the salary cost of the days away. You are being compared on this number already. Right now the comparison is being made without you in the room. Build a record of last year instead of a recap of last year. Not photos and a highlight reel. Meetings held, introductions made, what attendees reported ninety days later, retention. And a structured summary of open-ended feedback: the repeating complaints, the unexpected wins, the gap between promise and perception. Start measuring now even if the first year's numbers are thin, because the second year is when it becomes an asset. Separate your payload from your manifest and write the policy down. One page, agreed by marketing and content together, naming what ships early and what stays in the room. The argument is much cheaper to have in January than in the six weeks after the show. Get your specifications for the anticipated event out of PDFs and into plain web text at stable addresses. Floor plan, capacities, session times, speaker list, meeting inventory, pricing. A document behind a form is a document nothing can find. The agent evaluating next April needs a specification, not a sizzle reel. Publish the cognitive architecture of the event. Session density, walk times, break quality, quiet space, and the ratio of programmed time to unprogrammed time. An agent evaluating on behalf of a burned-out team will score this, and so will the human buyer who has been to too many conferences in a row. Stop marketing with promises you cannot check. Every unfalsifiable adjective in your campaign is a zero to a buyer that computes, and it was already an eye roll to the human one. See Appendix AThe fastest audit of your own event is to read The Nine Briefs and count the fields you cannot fill. Nobody has to tell you what that means. 10 The case for the room Nothing here is an argument against gathering. It is an argument against the way gathering has been sold. The two people in those two rooms disagreed about the machine. They did not disagree about the room. Both of them have spent their careers on the same conviction, which is that something happens when people are in a building together that does not happen any other way, and that the industry has never been especially good at explaining what. We compensated with atmosphere. We sold the promise of a feeling because we could not produce the evidence of an outcome. A buyer that reads specifications takes that compensation away. What it gives back, if you are willing to do the work, is the thing the business has been missing for forty years: a reason to attend that can be checked. The organizers who come out of this decade in good shape will not be the ones who guessed right about agents. They will be the ones who finally wrote down what happens in their room, and discovered they had a case all along. Appendix A The Nine Briefs There is no such thing as the attendee. There are nine reasons anyone gets on a plane, and each one produces a different evaluation of your event. What follows is written twice over, on purpose. For a buyer, it is a toolkit. These are the briefs, in the language a serious evaluator would actually use, and running your own decision through the right one will tell you in ten minutes what a brochure will not tell you in ten weeks. For an organizer, it is an audit. Read the nine and count the fields you cannot fill. Each brief carries three things: the instruction, the facts it needs from you in order to reach an answer, and the part of your operation that breaks when those facts have to exist in public before the show rather than in a spreadsheet afterward. 01 Named account pipeline The brief Attend up to four industry events this year, twelve thousand dollars all in. Prioritize events where decision makers from these fourteen target accounts will be present. Give me expected meetings per event before I commit to any of them. What it reads Attendee composition by company and seniority. Meeting inventory, how it is allocated, and whether meetings can be locked before registration closes. Historical meetings held per attendee. What it breaks The attendee roster stops being a sponsorship upsell and becomes the primary published asset of the event. Meeting inventory becomes priced product rather than an amenity. And the sales team has to be able to describe the room long before the room is full, which pulls commitment deadlines forward for everyone. 02 Sourcing and supplier discovery The brief Find an event where I can evaluate at least six vendors in this category side by side inside two days, see working demonstrations rather than literature, and speak to people who can answer technical questions. What it reads Exhibitor list resolved by category rather than by booth number. Whether the category is clustered on the floor or scattered. Booth type, meaning demonstration capability versus display. Walking distance between the relevant stands. What it breaks The floor plan becomes a taxonomy problem rather than a real estate problem. Assigning space by tenure and spend, which is how nearly every show has done it since the beginning, collides directly with a buyer who needs the category in one place. Somebody has to decide which customer the floor is arranged for. 03 Skill or certification The brief Find training that delivers this specific competency with a credential my employer recognizes. Compare cost per hour of instruction against the online alternatives and tell me what the in person version adds. What it reads Session level learning outcomes stated as outcomes. Instructor credentials. Accreditation and contact hours. Class size. Whether the material is available on demand afterward, which is the question that decides the whole thing. What it breaks The session description written to sound interesting loses to the one written as a syllabus. Programming has to publish outcomes months before most organizations have settled on a theme, which inverts the content calendar. And the free recording policy stops being a marketing decision and becomes a pricing decision. 04 Recruiting and being recruited The brief Which events put me in a room with hiring managers in this function, and which ones have a genuine career component rather than a job board nobody visits. What it reads Attendee composition by function and level. Employer presence. Whether recruiting is a sanctioned activity on the floor or a discouraged one. What it breaks Most organizers quietly suppress recruiting because exhibitors complain about poaching, and the policy has never had to be written down. Under disclosure it becomes a published field, and declining to state it does not make the evaluation generous. The organizer has to pick a side and say so. 05 Launch and visibility The brief Where does this product get in front of the most qualified press, analysts, and buyers per dollar spent, and what did comparable launches at this event actually generate in coverage and pipeline. What it reads Credentialed press by outlet type rather than a raw media number. Analyst attendance by firm. On site broadcast and interview capacity. What happened to last year's launches. What it breaks The press list becomes an asset the organizer is expected to describe, which the press will not enjoy. More painfully, the organizer has to track outcomes from launches that were not theirs, which is a measurement function almost nobody staffs. 06 Defending relationships at renewal The brief Which of our top forty accounts attend this event, when do their contracts renew, and can I get scheduled face time with them before the renewal date. What it reads Overlap between the attendee list and a customer list. Event dates against fiscal and renewal calendars. Availability and pricing of private meeting space. What it breaks Date selection stops being a venue availability problem and becomes a customer calendar problem. Meeting space stops being overflow and becomes retention infrastructure, which means it should be designed, priced, and sold as such rather than given away to whoever asked first. 07 Market and competitive scan The brief Give me the single event per year with the highest concentration of category decision makers and the broadest supplier coverage. I am buying situational awareness, not meetings. What it reads Coverage expressed as a share of the defined category rather than raw attendance. How much of the industry is actually in the building. What it breaks Attendance figures stop being the headline metric, which removes the number the industry has been marketing on for a century. The replacement is penetration of a defined universe, and defining that universe honestly is work most organizers have never done, partly because the answer is often uncomfortable. 08 Peer community and belonging The brief Find gatherings where people at my level and stage will be, capped small enough that I will actually be known by the end of it, and tell me what share of last year's attendees came back. What it reads Cohort composition by level and stage. A stated size cap. Return rate. The ratio of participation to presentation in the format. What it breaks Growth stops being self evidently good. Publishing a cap and a return rate makes the small event legible for the first time and leaves the very large one explaining why its retention is what it is. This is the brief that most threatens the economics of scale. 09 Obligation The brief None. The board seat, the committee, the sponsorship agreement, or the boss already decided. What it reads Nothing. The decision preceded the evaluation, which is what makes this one the outlier and the one worth studying hardest. What it breaks Obligation attendance inflates retention and disguises it as loyalty. An organizer who does not know what share of the room is here by contract rather than by choice is misreading their own strongest number. It is also the most fragile segment in the building, because obligation is exactly what a finance team eliminates first when it finally runs the arithmetic. Measure it before someone else does. One closing note for organizers. Nine briefs, and only two of them are satisfied by your program. That ratio is the paper's whole argument in a single number, and it does not change whether or not a machine is the one doing the reading. 11 Crystal ball: what would have to exist before a machine buys the ticket Everything up to here is happening now. This section is guesswork, and it should be read that way. None of the services below exist in usable form, some of them never will, and all of them assume a buyer willing to let software spend money before a person has looked at the details. So far the agent in this paper only reads and recommends. It takes the manifest, checks it against the brief, and hands back a shortlist. A person still registers, picks the sessions, and calls the hotel. The harder version is the one where the machine does the paperwork: registers the attendee, holds the meetings, and sorts out the change of plans in March when the brief moves. For that to be normal, better data is not enough. Seven things would have to exist. Each one is a plain answer to a question an organizer already gets asked by phone. Proof of who is actually coming A badge proves very little today. Anyone can type a job title into a registration form. If software is being sent to find fourteen named accounts, it needs some way to check that the people on the list are the people they say they are and still work where they say they work. That is not a surveillance question. It is the same question a sponsor asks when they see the attendee count and quietly discount it by a third. The difference between registered and coming Every organizer knows the gap between the registration list and the room. A free badge, a tentative hold, a competitor keeping an eye on you. Nobody publishes that difference and nobody can currently check it. If someone made it checkable, by flight booked or calendar held or a simple confirmation, attendance would become a probability instead of a headline, and the waitlist would stop being treated as a full house. Cancellation terms written where they can be read Anything that can book has to be able to unbook. Right now cancellation and transfer are settled by a phone call to someone in the office six days out. That works between people and not at all otherwise. The rules would have to sit in plain language, before purchase: what transfers, until when, for how much back. A meeting slot that is actually a meeting No agent can promise its buyer a conversation with a named account on the strength of a scheduling app. It needs a slot that is claimed, agreed by both parties, and attached to a room, and that leaves inventory when it is taken. That is the difference between a matchmaking feature and a booking. Airlines worked this out decades ago. Somebody other than the organizer checking the numbers The box office report only works if the buyer believes it. Audited circulation exists in publishing for exactly this reason. Somebody neutral would have to confirm that the meetings happened, that the survey went to the whole room rather than the happy part of it, and that the pipeline figure means what it says. This is the item organizers will like least, which is usually a sign it is the one that matters. A floor plan you can buy from If space is being bought on adjacency and traffic, a PDF will not do. The plan would have to carry real dimensions, who is confirmed in which stand, and a stated rule about what can move and when. Same for meeting rooms and suites. A plan that can be read is a plan that can be sold against. Shared words for what kind of event this is Summit, conference, forum, retreat, congress. The words are used interchangeably and everyone in the business knows it. People cope with that; software does not. Comparing three events fairly requires agreeing on a few basics, starting with how much of the time is people talking at the room versus the room talking to itself. What all seven have in common is that they shift trust away from the organizer's name and onto something a stranger can verify. That is the uncomfortable part, because reputation has been this industry's answer to hard questions for a very long time. And one last caveat, because it is the honest one. Even with all of it built, the machine is working for a person who wrote the brief and will live with the trip. Attending is too expensive and too political to hand over completely. The realistic risk is smaller and closer: the human stops doing the comparison by hand, and an event nothing can read simply never makes the shortlist. Back to the series