event-attendee-experience

Event Attendee Experience

You design the day a general attendee actually walks through: getting in the door, finding the room, where to go when the day gets loud, eating, asking, and what they remember. Inputs arrive decided - a signed floor plan, a published grid, a register, an accessibility bar - and the output is a design somebody can execute and audit by walking the building.

You add only the attendee-specific layer on top of the population-agnostic fundamentals in samber/dev-event-organizer-skills@event-hospitality, which decides only the catering service style, what a scheduled break contains, the alcohol posture, and whether an evening programme exists. It routes signage, wayfinding and facilities back here, and the welcome register and staff warmth to samber/dev-event-organizer-skills@event-cultural-identity. Consume its brief; do not restate its decisions here.

Seven siblings already own part of this day

Each states its own side of the boundary. Consume them; do not re-derive them.

Sibling	The line
event-cultural-identity	Sets the register you carry; you decide where the desk stands and how the queue moves. Ask which register it set, never redesign it.
event-code-of-conduct	Borrows your quiet room as an intake space - the room must serve decompression and confidential intake at once (§ Facility provision depth).
event-accessibility-inclusion	Decides whether a provision exists, how many, dietary categories, the accessible route; you own where it sits and how someone reaches it.
event-run-of-show	Staffs and executes what you design. Its named organizer-on-duty is this skill's escalation at check-in.
event-speaker-experience	A separate population with zero overlap: speakers get their own host, tech check, green room.
event-venue-sourcing	Secures the envelope, already naming hallway-track space and a quiet room as requirements. Work inside the signed floor plan; never reopen square footage.
event-schedule-design	Owns the grid. Design what happens inside a scheduled gap, never move or lengthen a break.

Five further siblings now ship: samber/dev-event-organizer-skills@event-hospitality, event-booth-experience, event-volunteer-experience, event-attendee-email-sequences and event-no-show-management. Every boundary against them below is stated from this side only - check theirs before relying on it.

event-booth-experience and event-volunteer-experience are genuine two-sided design problems: booth traffic runs through the same corridor your wayfinding serves, in the opposite direction, and volunteers use the meal line and quiet room you size without owning their design. event-attendee-email-sequences owns cadence and copy; you are the source of truth for what those emails say about the on-site day.

Every ranking below is a default, not a law: it shifts with context and who executes it. Re-rank all four menus after the interview. Each of these overturns a default rung: a venue with one entrance and one corridor, a registration system that exports a sorted list, a signage set you already own from last edition, a volunteer who has run a desk before, a sponsorship agreement that already promised lead capture.

Interview

Ask one question at a time, multiple-choice where possible. Questions 7-9 exist because the menus below diverge sharply on time-to-effect, durability and effort - those defaults cannot be picked for the user.

1. How many attendees, across how many days, in how many buildings or floors?
2. Is there a registration gate at all - paid ticket, free-with-registration, open door? Decides whether half this skill applies.
3. What register did the culture work set - warm and chatty, or efficient and quiet?
4. What does the signed floor plan actually give you: rooms, corridors, food, sponsor tables, entrance?
5. What dietary categories and accessible-route bar were set for you? Ask; never invent the list.
6. Does any sponsorship agreement already promise badge scanning or lead capture? If yes, the posture was decided upstream.
7. For each artifact, what's the last date it can still change - badge print, signage print, catering headcount, room allocation?
8. One-off edition, or recurring, where this year's walkthrough becomes next year's starting map?
9. Effort ceiling: volunteer-hours on the desk, print budget, and what you already own (last edition's signage, a floor plan, a list export)?
10. Which distinct ticket categories exist, and do any of them need their own check-in lane?
11. Does a meaningful share of the audience attend with caring responsibilities (children, dependents)?
12. Does the code-of-conduct pipeline already name an on-duty responder for the day?

Community-run vs vendor-run, and the axis that actually reshapes the menus

What changes the design is who owns the event, not who paid for the ticket:

• Community-run - volunteer labour, thin print budget, care bought with attention: pre-sorted badges, a named person at the desk, a room signposted by hand.
• Vendor or corporate-run - paid staff, printed signage systems, a lead-capture expectation written into sponsorship contracts. Bigger machinery hides an attendee's confusion more easily, so the walkthrough audit matters more here, not less.

The axis that genuinely reshapes the menus is different: whether the event has a registration gate at all. With no gate - a legitimate posture at 8,000+ attendees (FOSDEM), not an omission - there is no badge, no list, no dietary match, no check-in queue, and the whole design collapses onto wayfinding and the help path. Ask Q2 before ranking anything.

Workflow

1. Write down what you may not move - floor plan, grid, register, accessibility bar. Contradictions route to that sibling.
2. Walk the journey on paper first. One row per touchpoint from arrival to departure: references/journey-audit-and-first-timers.md. Design against the rows that come back bad.
3. Design check-in. Place the desk clear of the sponsor queue, pre-sort badges by last then first name, give walk-ins their own slower lane. Full mechanics: references/checkin-badges-and-swag.md.
4. Set the badge and scanning posture, then execute it in the artifact itself - where the code is printed and how the lanyard hangs are part of the posture.
5. Design the wayfinding from the floor plan's decision points, not the building's rooms: references/wayfinding-facilities-and-food.md.
6. Locate the facilities, designing the quiet room for both purposes in one pass - decompression and confidential intake.
7. Execute the dietary bar via one of two fulfilment models this skill deliberately refuses to rank (see reference) - your catering contract picks one.
8. Build the help path, and make whoever staffs it visibly identifiable.
9. Design the first-timer's first twenty minutes and the one moment worth remembering - both self-set.
10. Walk the finished design in the actual building before doors open, using the same rows from step 2, and record what you changed.

Show the user the floor-plan markup, the badge design and the signage list before anything is printed. A printed sign with the wrong room number is the one mistake you cannot fix at 09:00.

If your harness has persistent memory, record per edition:

• The longest check-in queue and the hour it happened.
• Which questions the help point actually received.
• Whether the quiet room was used, and for which of its two purposes.
• How many dietary-restricted meals went unclaimed or went to the wrong person.
• Every sign you added by hand on the day.

The hand-added signs are the wayfinding gaps your design missed.

Check-in model

How an attendee gets from the door to being inside with the right badge. Ranking (default, not a law - Q1, Q2 and Q9 re-rank it):

• effort: multi-lane staffed desk > pre-sorted pickup with a walk-in lane > single-table name-check > no registration at all
• value: multi-lane staffed desk > pre-sorted pickup with a walk-in lane > single-table name-check > no registration at all
• compliance cost: multi-lane staffed desk == pre-sorted pickup with a walk-in lane == single-table name-check > no registration at all
• efficiency: pre-sorted pickup with a walk-in lane > single-table name-check > no registration at all > multi-lane staffed desk

Value and effort tie exactly, so no pair strictly dominates: whichever rung leads on value costs strictly more effort. The three staffed rungs carry equal compliance cost - same attendee list and data, differing only in headcount who holds it; no registration is the only honest zero. Full dominance-check reasoning is in references/checkin-badges-and-swag.md.

• Pre-sorted pickup with a walk-in lane - the default: badges sorted alphabetically before doors open, a separate slower lane for walk-ins who still complete the full form. Highest value per hour since the sort happens the night before. Move up one rung once the door-open rush exceeds one queue, or a real ticket-category split exists.
• Single-table name-check - one volunteer, a printed or searchable list, a mark against each name. Buys a headcount, a name badge and a dietary match for one person's time.
• Multi-lane staffed desk - the starved option: tops effort and value together, so efficiency never picks it. Promotion condition, keyed to Q10: at scale, for a multi-day badge window, or wherever sponsor and regular tickets differ.
• No registration at all - a legitimate posture, not a gap, at 8,000+ attendees (FOSDEM). Removes the queue, the list and the data entirely, and costs you the headcount, badge, dietary match and identification the code of conduct leans on. Choose it deliberately.
• Deleted, not demoted: one table doing check-in, swag, walk-in registration and questions at once. It fails at exactly the busiest ten minutes of the day. Split the walk-in lane, move questions to the help point, or accept a queue out the door.

Full desk mechanics, including the ticket-category confusion the sort has to tolerate, are in references/checkin-badges-and-swag.md.

Badge and scanning posture

What is printed on the badge and what a machine can read off it. One rule sits above the menu: attendee contact lists are never given or sold to sponsors. Ranking (default, not a law - Q6 re-ranks it outright):

• effort: QR on the badge's back with consent, opt-out and sponsor terms > UUID-only code > name badge with no machine-readable code
• value: QR on the badge's back with consent, opt-out and sponsor terms > UUID-only code > name badge with no machine-readable code
• compliance cost: UUID-only code > QR on the badge's back with consent, opt-out and sponsor terms > name badge with no machine-readable code
• efficiency: name badge with no machine-readable code > QR on the badge's back with consent, opt-out and sponsor terms > UUID-only code

Value and effort run in the same order, so effort always outpaces value's lead. Compliance re-orders the menu but rescues no pair: UUID-only tops compliance cost despite protecting the badge itself, because a scan log handed to a sponsor re-identifies everyone in it, the exact act the rule above forbids. Full reasoning is in references/checkin-badges-and-swag.md.

• Name badge with no machine-readable code - the default and the published stance: scannable badges are actively discouraged. The name alone identifies who belongs on site and doubles as an icebreaker. Zero exposure, zero build.
• QR on the badge's back with consent, opt-out and sponsor terms - the starved option: tops effort and value together, so efficiency never picks it first. Promote it the moment Q6 confirms a signed sponsorship agreement already promised lead capture.
• UUID-only code - an opaque identifier resolved only by the organizer. Protects the badge but moves the entire risk into what you do with the scan log afterwards.
• Deleted, not demoted: attendee data encoded directly in the QR. Every scanner in the building reads it, not only an authorized sponsor's - the same rung with the consent removed.

Full badge mechanics, both technical fallbacks and all four required mitigations, are in references/checkin-badges-and-swag.md.

Facility provision depth

How much of the venue you take out of circulation for people rather than programme. No compliance-cost axis: providing a room triggers no review - what it is used for does, and that sits in the code-of-conduct runbook. Ranking (default, not a law - Q1, Q5 and Q9 re-rank it):

• effort: full wellness catalog > quiet room plus inclusion wayfinding on the map > designated quiet room with stated hours > venue's own facilities only
• value: full wellness catalog > quiet room plus inclusion wayfinding on the map > designated quiet room with stated hours > venue's own facilities only
• efficiency: designated quiet room with stated hours > quiet room plus inclusion wayfinding on the map > full wellness catalog > venue's own facilities only

Value and effort are the same list, so every higher-value rung is strictly the costlier one; the ordering is argued, not verified. Full reasoning is in references/wayfinding-facilities-and-food.md.

• Designated quiet room with stated hours - the default: one named room, located on the map, with published hours. Wins efficiency by a wide margin: it is the only rung the incident runbook depends on, serving as both decompression space and the private space a responder needs. Design it for both at once - a door rather than a curtain, findable without crossing the main thoroughfare, signed so ordinary use never looks like an incident.
• Quiet room plus inclusion wayfinding on the map - add colour-coded gender-neutral restrooms and mapped first-aid points. Cheap once the map is being drawn anyway.
• Full wellness catalog - the starved option: quiet room, nursing/wellness suite placed near registration, multiple m

...