Designing a conference programme

A programme built to fill slots looks different from a programme built around what attendees need to learn, and the difference shows up in the running order before a single speaker has been booked. The slot-filling version starts with who is available: a sponsor's product manager, a colleague who owes someone a favour, three strong applications from the call for papers (CFP). The audience-first version starts with a list of what attendees should be able to do, decide or argue by the end of the day, and only then asks who can teach that.

One test separates the two quickly. For any session on the running order, ask who it is for and what they leave with. If the plain answer is "whoever turns up" or "awareness of the product", the slot has been filled.

Working backward from outcomes

Before drafting a speaker list, draft an outcomes list: three to six things attendees should be able to do, decide or argue once the day is over. A programme for finance managers might aim for attendees who can explain a new reporting standard to their own team, or who can name two suppliers worth a follow-up call. Each session then earns its slot by mapping to one of those outcomes. A session that does not map to anything on the list is a favour to whoever is speaking.

This reordering also changes who gets invited to speak. Working from outcomes first makes it obvious when a strong CFP submission does not actually fit anywhere, and obvious when a gap in the outcomes list has no speaker attached yet. Guidance on running a CFP alongside invited speakers without losing this discipline is covered in Finding, Choosing and Looking After Speakers.

Where supply-side pressure creeps back in

Sponsorship money and speaker favours do not disappear because the outcomes list exists. They show up as pressure to add a slot after the outcomes have already been set, which is exactly where a pay-to-speak programme starts, a paid sponsor slot dressed as a talk with no outcome attached to it. Holding the line means being able to say, to a sponsor or a colleague, which outcome a proposed session serves, and turning down the ones that cannot answer that.

The same audience-first question applies to who can attend the sessions once they are built. A session format that assumes everyone can stand for two hours, or a panel with no captioning (CART) for a deaf attendee, has quietly narrowed who the programme is actually for. Where the accessibility decision belongs in programme design, not bolted on afterwards, is set out in Accessibility and Codes of Conduct. A worked example of an outcomes-first running order is in Worked Examples and Templates.

A single-speaker session holds a room for about 45 minutes before attention starts to drop, and after that the speaker is talking to people who have already checked out. A workshop can run longer, because the audience is doing something, and a panel discussion tends to sag past 40 minutes unless the moderator is actively steering it. Building a programme around a single 60-minute slot length for every format is the single most common pacing mistake, because it treats a keynote and a hands-on workshop as if they place the same demand on the audience.

What each session format can sustain

• Keynote: 30 to 45 minutes plus questions. Longer than that and you are relying on the speaker's charisma to carry the room.
• Panel: 35 to 45 minutes. Four or more panellists rarely fit inside this without each one being reduced to a soundbite.
• Workshop: 60 to 90 minutes, because participants need time to actually do the exercise.
• Lightning talks: 5 to 10 minutes each, run back to back, because the format's value is variety and pace.

Why the break is a programme decision

A break is not the absence of programme; it is the part of the programme where people process what they just heard and decide who they want to talk to about it. Attendees consistently name the corridor conversation as the thing they remember from a conference, and that conversation only happens if the break is long enough for someone to queue for coffee, find the person they wanted to speak to, and actually finish the sentence. A 15-minute break that includes a room change and a queue for the toilets is not a 15-minute break. It is a five-minute break with a sprint attached, and it produces a room that files in still talking.

A break under 20 minutes between sessions in different rooms is usually too short once you account for room-to-room walking time on a large venue floor, and it is too short for a wheelchair user relying on a lift, or for someone who needs to find a quiet room to manage sensory load. The accessibility guide covers what reasonable adjustment looks like under the Equality Act 2010; the pacing decision that supports it is made here, in the schedule.

What a tight schedule actually does

An over-packed day does not fail gracefully. Sessions run over because a speaker was told 30 minutes and used 40, and every subsequent slot shifts by the same margin, which means the last sessions of the day lose their audience to catch-up conversations, missed trains and simple fatigue. Break time is the first thing organisers cut when the day is running late, which is exactly backwards: cutting the break protects the schedule on paper while removing the part of the day attendees actually valued. A programme that has no slack built in for a session running long has no way to recover except by shortening the thing people came for.

Building in five minutes of slack per hour, and treating the lunch break and one mid-afternoon break as fixed points that do not move regardless of what else has slipped, keeps the rest of the day recoverable. A sample timeline showing where that slack sits across a full single-track day is in the worked examples and templates; the venue and catering constraints that interact with break timing, such as when a room can physically be turned around for lunch, are covered in venues, catering and the costs nobody budgets for.

The corridor track is what organisers call everything that happens outside the scheduled sessions: the chat in the doorway after a talk, the queue for coffee, the group that ends up round one table at lunch and doesn't leave. None of it is on the agenda, but ask an attendee three months later what they got out of the day and it's often the thing they name first. A programme that treats this time as a gap between the sessions that matter is treating it as an afterthought it can't afford to be.

Give the corridor somewhere to happen

Conversations need a place to stand that isn't blocking a fire exit or squeezed against a wall where the AV cabling runs. A room booked purely for its seated capacity often leaves no usable space once the chairs are in, so check what the layout looks like with the breaks in mind. A registration area or a foyer with somewhere to put a cup down does more for the networking that people came for than an extra keynote slot. Venue capacity and layout are covered in more detail in Venues, Catering and the Costs Nobody Budgets For.

The lunch queue sets the networking window

A single buffet line for two hundred people can eat most of the lunch break before anyone has sat down to talk to anyone. Staggering lunch by ticket colour or table number, or running two or three serving points, buys back the time the queue would otherwise take. That decision belongs with catering and layout planning, because it's set weeks before the event.

The evening event is part of the programme

A drinks reception or dinner is still your event, and the same questions apply as to any session: who can attend, what it costs to opt out of, and what happens if something goes wrong. An evening built entirely around a bar leaves out anyone who doesn't drink, anyone with caring responsibilities who needs to leave by a set time, and anyone who finds a loud room hard going. A code of conduct that only covers the sessions misses the point that most reported incidents at events happen in exactly this unstructured time. Where the code applies, and who's responsible for enforcing it after the formal programme ends, is worth deciding before the day. Accessibility and Codes of Conduct covers what a workable policy needs to include.

Who the unstructured time leaves out

A quiet room away from the main space, sign-posted and genuinely quiet, costs little more than a spare room and a sign, and it matters to attendees managing sensory load, anxiety, or simply a long day of talking to strangers. An alcohol-free option at the evening event that isn't just water is a small addition that changes who can join in. These are decisions made when the programme is drafted.

A call for papers (CFP) is an open invitation for anyone to submit a talk proposal, reviewed against published criteria. Most programmes mix this with a handful of invited speakers, people the organisers approach directly because they hold a position, a piece of research or a story the programme needs. The two can sit together without conflict, but only if the selection process for each is kept separate and visible, because the moment a sponsor's product pitch slides into a session chosen the same way as an open submission, the audience notices and the programme's credibility goes with it.

What pay-to-speak looks like, and why it costs you trust

Pay-to-speak is a speaking slot granted in exchange for money, sponsorship or an in-kind deal. It rarely announces itself. It shows up as a sponsor's account manager filling a track with product demonstrations, or as a session that was clearly built to satisfy a contract. Attendees can tell the difference between a talk chosen because it earns its place and one that was bought, and once they've spotted a pattern of the second kind, they stop trusting the rest of the schedule too.

Separate the sponsor slots and label them

The fix is structural, not editorial goodwill. Decide in advance what proportion of total session slots sponsors can buy, commonly kept to a small minority of the programme, and hold that line even when a bigger sponsor wants more. Sponsored sessions should be marked as sponsored on the printed and online agenda, sit outside the CFP review process entirely, and never occupy a keynote or headline slot unless the audience is told why. A sponsor paying for visibility is a legitimate part of the budget; a sponsor paying for the appearance of editorial endorsement is a different transaction, and the two should never look identical on the page.

Run the open call against published criteria

Submissions to the CFP should be scored by a review panel against criteria set out before the call opens: relevance to the stated theme, originality, and whether the proposal is pitched at the level the audience actually needs. Reviewing abstracts with speaker names removed reduces the pull of who is well known versus what the talk actually offers. Publishing the criteria alongside the call also does some of the filtering for you, because it discourages proposals that were never going to fit. Finding, choosing and looking after speakers covers how to run that review and what to do once speakers are selected.

Deciding the split between open call and invited speakers

Fix the proportion between open-call and invited slots before the CFP opens, so the decision isn't made under pressure from whichever camp is most persuasive. A programme built mostly from invited experts gives you certainty over quality but narrows the range of voices and topics; a programme built mostly from the open call widens that range but adds review workload and some risk on delivery. Speaking at a conference sets out what a strong submission looks like from the speaker's side of the same process.