On this page12
- Why is hosting during conference week so hard to get right?
- The five-gate test
- What are the signs that a hosted event is ego spend?
- What budget band fits each format?
- How do you protect guest list quality?
- When does co-hosting beat hosting alone?
- When should you not host at all?
- FAQ
- How far ahead should we decide whether to host?
- Is it better to host before, during or after the main conference days?
- Can a small team host and still take its own meetings?
- Should we invite journalists to our hosted event?
Host your own event at a conference only if you can name most of the guests you need before you book the venue, those guests are hard to reach through ordinary meetings, and you have the people to run the follow-up afterwards. If you can't meet those three conditions, co-host with your ecosystem or skip hosting and spend the effort on pre-booked meetings. Hosting pays when it buys you a room you couldn't otherwise get. It's ego spend when it mainly buys your logo on a party invite.
I use a five-gate test with founders before they put down a venue deposit. It takes ten minutes, and it's ended more than a few plans that would have been expensive regrets.
Why is hosting during conference week so hard to get right?
Supply. Big crypto weeks now come with hundreds of parallel events, and every one of them competes for the same evenings. In 2024 at TOKEN2049 I counted the problem from the guest side:
events too fragmented, more than 800+ events. I found myself lost thinking which events to attend:/ Shilika Jain on X, 24 Sep 2024
The guests you most want are drowning in invitations. Many side events also gate entry through application forms, so your guests are doing admin just to get through the door, often for several events a night.
So a new event has to beat hundreds of alternatives for each guest's attention, with a tired audience doing paperwork to attend. That's the real bar. The test isn't whether you can throw a nice event. It's whether the specific people you need will pick yours.
The five-gate test
Score each gate 0, 1 or 2. Be harsh.
| Gate | What you're checking | 0 | 1 | 2 |
|---|---|---|---|---|
| 1. Named guests | The guests who'd make it worthwhile, by name | No names, just "builders" | A category with a few names | 15 or more specific names, most already in contact |
| 2. Reach gap | How hard those people are to meet any other way | They'd take a call anyway | Some would, some wouldn't | Most are unreachable without a reason to gather |
| 3. Reason to come | A reason beyond free food and drink | None | A theme | A real draw: a closed discussion, an announcement, a host they want to meet, a format nobody else runs |
| 4. Capacity | Running it without losing your own meeting schedule | No | With strain | Yes, or a co-host or vendor handles logistics |
| 5. Follow-up owner | One named person following up with every guest within 48 hours | Nobody | Shared, vaguely | One named owner with time blocked |
0 to 4: don't host. Attend, book meetings, and accept invitations to good events instead. 5 to 7: co-host. You have part of the case but not enough to carry the risk alone. 8 to 10: host, in the smallest format that serves your goal.
Gate five is the one teams skip. A great event with no follow-up is a cost with no return. My 48-hour follow-up system is the minimum I'd want in place before saying yes.
What are the signs that a hosted event is ego spend?
Watch for these in your own planning conversations.
- The first decision made was the venue, not the guest list.
- Success is described as "a packed room" or "the event everyone's talking about".
- The guest list is mostly whoever applies.
- Nobody can say what happens with the RSVP list after the night.
- The budget conversation started with "what are competitors doing?"
- The main deliverables are photos and a recap thread.
None of these are crimes. Brand events can be a legitimate choice. They just need to be budgeted and judged as brand spend, not as pipeline. If you want to compare formats before choosing, I've written a separate breakdown of side events versus booths versus sponsorship.
What budget band fits each format?
I don't quote figures because venue minimums, catering rates and AV costs differ wildly by city and by how close to the week you book. What changes between formats is which cost categories appear. Use this to see how fast the list grows, then price each line yourself. The conference budget template has the full line items.
| Band | Format | Cost categories that appear | Who it suits |
|---|---|---|---|
| Band 1 | Closed dinner or breakfast, under 20 guests | Venue minimum spend or private room, food and drink, invitations, one person managing RSVPs | Teams with a specific list of partners, investors or press |
| Band 2 | Co-hosted ecosystem session or roundtable | Your share of venue, food and AV, joint promotion, guest approval time, shared coordination | Teams in an active ecosystem with natural co-hosts |
| Band 3 | Solo mid-size event, a workshop or reception | Venue hire, AV, food and bar, guest platform, promotion, photographer, staff on the night | Teams with a strong reason to gather and a team to run it |
| Band 4 | Solo large party | Everything in Band 3 plus security, entertainment, branding, merch, transport, and often a premium for peak-week dates | Brands with recognition to protect, treated as brand spend |
Cost rises with each band, but the quality of conversation usually doesn't rise with it. The comparison between the smallest and largest formats is covered in depth in small dinners versus big parties. If budget is the tightest constraint, conference marketing on a budget has lower-cost options.
How do you protect guest list quality?
The guest list is the product. Everything else is packaging. A few rules I'd hold to.
- Write the list before the invite. At least half the target headcount should be named people, invited personally.
- Approve every RSVP. Open registration fills seats with whoever has the free evening.
- Track the named-guest ratio. That's the share of confirmed attendees who are on your list, or who you'd happily have added to it. Check it three days out. If it's low, cut capacity rather than open the doors.
- Decide plus-one policy in advance. Plus-ones are fine at a dinner for senior guests. At a mid-size event they dilute the room fast.
- Seat or introduce on purpose. A host's job on the night is matching people, not giving a speech.
Format can do some of the selection for you. An unusual format draws curious, senior people who skip the standard party. My favourite half-serious suggestion is still this one:
i don't understand why founders in the west aren't vibing on mango parties. this could be a crazy networking sess + experience for your community Shilika Jain on X, 15 May 2026
It doesn't have to be mangoes. It has to be something guests remember choosing.
When does co-hosting beat hosting alone?
Most of the time, if you scored 5 to 7. My TOKEN2049 notes from 2024 had a line on this:
Ecosystem events were impactful than siloed ones✨ Shilika Jain on X, 24 Sep 2024
Co-hosting with teams in the same ecosystem gives guests a reason to come (several relevant teams in one room), gives you pooled guest lists, and splits costs. It also borrows credibility. During that same TOKEN2049 week I attended a client event co-partnered by Google:
during the T2049, I got to visit Google’s SEA headquarters for a client event co-partnered by Google Shilika Jain on X, 2 Oct 2024
A known co-host changes who says yes. Co-hosted formats can also travel beyond the room. In my work with Fluence, DePIN Day Hong Kong, run with Titan Network, was covered in Chinese media by Blocktempo, and DePIN Day Singapore was picked up by CryptoTimes JP. A themed, multi-team event gives reporters a category story rather than one company's party.
Before committing, agree these points in writing with every co-host:
- Who approves guests, and whether each host has a veto
- How costs split, including overruns
- Whose branding goes where
- Who gets the attendee list, and in what form
- Who owns follow-up with each guest, so nobody gets three identical emails
When should you not host at all?
Even with a decent score, skip hosting if any of these apply.
- You're launching or announcing during the same week, and the team can't do both well.
- It's your first time at this conference. Attend first, learn the room, host next year.
- Your target guests are mostly reachable through pre-booked meetings. Don't pay for a room to have conversations you could have had over coffee.
- Your budget only stretches to a format that would be outclassed by the dozens of events on the same night.
There's a cost to skipping, too. You give up some visibility, and you're relying on others' events for chance encounters. For an early-stage team that's usually the right trade. If you do host, plan the press and content around it as part of a wider Web3 PR campaign, so the event feeds the story rather than standing alone.
FAQ
How far ahead should we decide whether to host?
Six to eight weeks out is the latest I'd decide, because venues near major conferences book up and prices tend to climb as the week approaches. Deciding earlier also gives you time to approach co-hosts, who are planning on the same timeline.
Is it better to host before, during or after the main conference days?
Before or on the first evening usually beats the last night, when people are tired and leaving. Breakfasts and lunches compete with fewer events than evenings. Pick the slot your named guests are most likely to have free.
Can a small team host and still take its own meetings?
Only if the event is small or someone else carries the logistics. A dinner for a dozen people is manageable. Anything larger needs a co-host, a vendor or a dedicated team member who skips meetings that day.
Should we invite journalists to our hosted event?
Invite them only if there's something for them there: a closed discussion on a topic they cover, a founder worth meeting, or news. Don't expect coverage of the event itself. The value is the relationship and the briefing that follows.
Not sure whether your conference week needs your own event or just a better calendar? Send me the plan and I'll tell you honestly which gate it fails.
