# Event Desk

> Paste what your company does (the buyer, the deal size, the sales cycle, what a qualified meeting costs you from outbound or ads) and your notes on one event - a conference sponsorship, a trade show booth, a webinar, a speaking slot, a dinner or a meetup. The browser adds up the costs, counts your buyers in the room, works out the qualified meetings and the cost per qualified meeting, compares it with your other channels and flags the prospectus red flags. A paid decide lane says go, go with changes or pass, in which role and on what terms; a paid plan lane writes the before, the during and the after.

URL: https://event-desk.skillsafe.ai/
API tutorial: https://event-desk.skillsafe.ai/api.html

## Who it is for

Founders and marketers at B2B companies deciding whether one specific event is worth the money and the time, and then running it so it produces meetings and pipeline rather than badge scans. They want to know whether to sponsor, attend, speak or host, what to negotiate, and what to do before and after.

## The free math (in the browser, no model)

- The event notes as labelled lines ("Booth package: $28,000", "Attendees: 6,000", "ICP attendees: 350", "Budget: $45,000", "Meetings target: 20") or prose; the date in ISO or written form.
- Every cost line with its amount and the total; the attendee count and how many are your buyers (a count or a share); your deal size, sales cycle and what a meeting costs you elsewhere.
- Qualified meetings: your stated target, or for a booth the skill's directional 10-20% of your buyers in the room. Cost per qualified meeting, cost per buyer in the room, the ratio to your other channels, the days to go.
- Flags EV-01..EV-12: a meeting costs 3x your other channels or more, attendee counts without composition, no cost, no benchmark, no primary outcome, vanity metrics (badge scans, impressions, foot traffic), no follow-up plan, 5,000+ attendees, no date, soft sponsorship pricing inside 8 weeks, over budget, a date already passed.
- Exports: the math as Markdown and the costs as CSV.

## The paid lanes (one system prompt, routed on `task`)

- `decide`: go, go with changes or pass (pass is required at the 3x line with no strategic multiplier); the role to take; whether in person is needed; the three strategic multipliers (exact buyers concentrated nowhere else, a category-defining event, access you cannot buy elsewhere) quoted from the input; the four prospectus red flags; the counterfactual; for a sponsorship, the negotiation asks in the skill's order (speaking or panel slot, side-event rights, attendee list reality check, placement and timing, price); an alternative play such as a side-event dinner; conditions and what to find out.
- `plan`: the goal metric (qualified conversations, meetings booked, opportunities created, pipeline influenced, cost per qualified meeting, cost per opportunity, closed-won influenced) with its target and owner; who must be in the room and where to build the list; a 4-8 step timeline ordered before, during, after; a meeting-booking email of at most 100 words asking for 15 minutes; the 30-second qualifying question; what to capture after each conversation; follow-up in three tiers (hot, warm, scan-only) within 24-48 hours; content to capture; 3-6 metrics from the real and decisive tiers.

Every reply is checked against the input: quotes word for word, the verdict rules, the multipliers and asks in order, the follow-up tiers, no vanity metric, and no invented cost, count, rate, date or duration.

## Source

Derived from the agent skill @coreyhaines31/events (https://skillsafe.ai/skill/@coreyhaines31/events), by Corey Haines, in coreyhaines31/marketingskills (MIT): https://github.com/coreyhaines31/marketingskills/tree/main/skills/events - its four roles, one primary outcome, portfolio-first selection and in-person test, the 20% event / 80% before-and-after arc, the content arc and the measurement tiers; and its sponsorship reference (references/sponsorship-roi.md): the should-we-sponsor evaluation, the prospectus red flags, the negotiation order, the side-event play, working the booth and the tiered follow-up.

Runs on gpt-terra. Part of SkillSafe (https://skillsafe.ai/).