Most efforts to make an event more open happen at the wrong end of the process. Organisers put together a programme, notice that everybody speaking looks broadly similar, and try to fix it with invitations in the final few weeks. By then the decisions that determined who could come were made months earlier, in the ticket price, the date, the venue, the way proposals were requested, and whether there was any travel budget at all.
Who ends up in the room is mostly a question of logistics and money. The visible result is the programme. The causes sit almost entirely upstream of it. The same is true of much of what makes volunteer groups work.
The ticket price is only part of the cost
The price on the ticket is the visible cost. The real cost is that plus travel, plus somewhere to stay, plus time away from paid work, plus, for a substantial number of people, arranging for somebody else to be looked after.
An event can be free and still be impossible to attend. The intervention that works is money set aside specifically to cover those other costs, announced early and easy to request. Announcing it early matters more than organisers expect, because somebody deciding in March whether they can manage September needs to know before they commit to anything else.
Asking people to prove hardship defeats the purpose
Two details determine whether a support scheme actually reaches anyone. The first is that applicants should not have to document their circumstances in detail. A demanding process filters out precisely the people it was created for, because explaining why you cannot afford something to strangers is its own cost.
The second is that it has to be announced alongside the call for speakers. People decide whether to put themselves forward based on whether they could attend if accepted, so a support scheme revealed after the deadline has already missed most of its effect.
Every date excludes somebody, so choose deliberately
A weekday event selects for people whose employer pays for their time. A weekend event selects for people without weekend caring responsibilities. An evening event selects for people who do not do the school run.
There is no neutral option, only an explicit one. The useful exercise is to name who a given date excludes and decide whether that is acceptable. The common alternative is to pick whatever suits the organisers and assume it works for everyone else.
Accessibility means to the front, not just through the door
Venue decisions carry more weight than almost anything in the programme. Step-free access throughout the building, not only at the entrance. Accessible toilets that are not on another floor. A room where somebody using a wheelchair can reach the front and speak, not only sit in the audience.
The one most often forgotten is the social part afterwards, which frequently happens somewhere that excludes people the main event had successfully included. An event is only as accessible as its least accessible component, and the pub is part of the event.
How you ask for speakers decides who offers
The format of a request for proposals is itself a filter. Asking for a long written abstract in polished English selects for confident writers and against people working in a second language or putting themselves forward for the first time, regardless of whether their ideas are any good.
What demonstrably helps is publishing examples of proposals that were accepted, offering to look at drafts before the deadline, saying explicitly that first-time speakers are wanted, and separating assessment of the proposal from assessment of the person who submitted it.
Paying speakers afterwards excludes the people who most need paying
There is an unexamined gap between inviting somebody to speak and making it possible for them to speak. The practical question is whether costs are covered in advance or reimbursed later.
Reimbursement afterwards excludes anybody who cannot float several hundred pounds for two months, which describes a great many early-career and independent participants. This is rarely deliberate. It is usually just how the finance process works, and changing it is one of the highest-impact things an organiser can do.
Let speakers say what they need
Some organisers now publish a list of things speakers can ask for, inverting the usual direction. The organisers set out what is available and invite people to say what applies, which removes the need for anybody to raise a requirement they feel awkward about.
This works because the barrier is usually embarrassment, not ignorance. Somebody who would never write to ask whether there is a quiet room will happily tick a box saying they would use one.
A code of conduct without anyone to act on it is a liability
An event code of conduct is a commitment that somebody will respond appropriately if something goes wrong. That commitment is only credible if the capacity behind it exists.
The questions that establish whether it does are narrow. Is there a named, reachable person during the event, identifiable on sight, with a number that has been published. Is there a process agreed in advance, before somebody has to invent one at eleven at night. Are the people receiving reports distinct from the people who might be reported about, which at a small event run by a close-knit group frequently fails. Is there any route to appeal. An event that adopts the document without answering these has adopted a document.
Recording the talks reaches more people than any travel fund
The largest single expansion of who can take part comes from making travel unnecessary.
Talks recorded and published promptly, with captions and slides available, reach people for whom attending was never realistic: those in the wrong time zone, without a budget, unable to get a visa, or managing a health condition that makes travel impractical. For a global audience it is the only intervention that reaches most of them at all.
Hybrid events usually serve one audience and claim to serve both
It is worth being honest that running an event for people in the room and people at home simultaneously is genuinely difficult. Most attempts deliver a poor experience remotely while consuming organiser attention that would otherwise have improved the in-person event.
Nobody has convincingly solved this. Events that claim to have done so are generally prioritising one audience and describing it as both, and the more useful approach is to decide openly which is primary and then do the other one properly.
Publishing who came is more useful than saying it went well
Claims about how open an event was are usually self-assessed and impossible to check. Two things change that.
The first is collecting basic information about who attended and who spoke, on a voluntary self-identified basis, and publishing the totals. Ideally before the event, so it can still influence decisions. The data will be imperfect and small events will have numbers too small to interpret, but it is still better than an assertion.
The second is submitting to assessment against criteria somebody else wrote. The difference between a standard met and a claim made by the organisers is the difference between a measurement and an opinion.
Small events should report what they did
An event with forty people has categories counted in ones and twos. Reporting proportions gives an impression of precision the numbers cannot support, and invites arguments about statistics.
The honest alternative is to report what was done. Was financial support offered and advertised with the call for speakers. Were costs paid in advance. Was the venue checked to the front of the room. Were named contacts published. Were the recordings put up. Those are answerable, verifiable, and considerably more useful to somebody deciding whether to attend next time.
The order these decisions get made in
Budget first, because financial support and paying speakers in advance both have to exist before anything is announced. Then the date, chosen with its exclusions named. Then the venue, checked properly and in person. Then the call for proposals, with examples and an offer to review drafts. Then the arrangements for conduct, with real people named. Then, before anybody arrives, the decision about what will be recorded and when it will be published.
Almost everything that determines who can attend is settled by the end of that list, and none of it involves the programme.