The arithmetic nobody does
Suppose each person in your group is free on any given evening about 60% of the time - a generous assumption for adults with jobs, partners or children. The chance that all of them are free on one particular evening is that 60% multiplied by itself once per person, and it collapses fast.
| Group size | All free on one given evening | All free on at least one of three options |
|---|---|---|
| 4 people | 13% | 34% |
| 6 people | 5% | 13% |
| 8 people | 2% | 5% |
| 10 people | 0.6% | 2% |
So aim for a quorum, not for everyone
The useful move is to stop treating full attendance as the goal. Ask instead how many people make the evening work - and notice how much easier the problem becomes the moment you do.
| Target, from a group of 8 | On one given evening | Across three options |
|---|---|---|
| All 8 can make it | 2% | 5% |
| At least 7 can | 11% | 29% |
| At least 6 can | 32% | 68% |
| At least 5 can | 59% | 93% |
What that means in practice
Dropping the target from “everyone” to “six of the eight” takes the same three date options from roughly a 1-in-20 chance to better than 2-in-3. That is the entire difference between a plan that happens and a thread that dies in a fortnight.
It also reframes what you are apologising for. You are not failing the two people who cannot come; you are running an event that most of the group can attend, and catching up with the others another time. Say that out loud when you announce the date - it lands much better than a fortnight of trying to solve an impossible constraint.
Name the people who genuinely have to be there
Before you open a poll, decide who is non-negotiable. Usually it is a very short list: the guest of honour, the host, anyone whose house or car the plan depends on. Check those diaries first, privately, and only offer dates that already work for them.
Everyone else is optimisation. This one step removes most tie-break arguments, because the awkward outcome - the winning date being the one the birthday person cannot do - is now impossible by construction.
Designing a poll that closes
- Offer three options. Five is the absolute maximum, and only when the group is large.
- Make each option complete: “Saturday 12 September, 7pm”, not “week of the 12th”.
- Remove any date the essential people cannot do before you post it.
- Say when voting closes - two to four days is usually right.
- Say how ties will be broken, before anyone votes.
- Say what happens after: that you will confirm the date and then ask for actual RSVPs.
- Close it on time and announce the result, even if turnout was poor.
Messages you can copy
Trying to get everyone together before the end of the month. Three options - vote for every one you could genuinely do: • Saturday 12th, 7pm • Friday 18th, 8pm • Saturday 26th, 7pm Voting closes Sunday night. I’ll take whichever works for the most people, and if it’s a tie I’ll go with the earlier date.
Five of you still to vote - takes about ten seconds and I’m closing it tonight. If none of the three work, just say so and I’ll count you out rather than waiting.
Closed - we’re going with Saturday the 12th at 7pm. Six of us can make it, which is enough to be a proper evening. Sorry to Priya and Tom, who genuinely couldn’t do any of the three - let’s do something separately soon. Now that it’s fixed: can everyone confirm you’re actually coming? Voting that you were free isn’t the same as being in.
Voting free is not the same as coming
Someone can be genuinely free on all three dates and only ever intend to come to one. If you treat poll votes as a headcount you will over-book the table and over-buy the food, and then feel let down by people who never actually said yes.
Once the date is fixed, ask the separate question - are you coming? Status Social keeps the two steps joined up: guests vote, the host picks the date and the same event then collects a real going, maybe or can’t-make-it against the confirmed plan.