7 sources · four cities · None of it about us
Why Offsite
Your people are already in the same city and already apart. We book one thing a month for every 6 of them and pay for it. Five figures below: what the research finds, what it cannot show, and what a seat costs against the budget it comes out of.
Free to set up. You see your clusters and your price before anything is owed. Then 6 seats minimum, 3 months minimum.
The hour, and who it lands on
EMANUEL ET AL., SCIENCE, 2026 · OVER 500,000 AMERICANS
The effect is real and it is narrow. It sat entirely with people who live alone — a group you cannot identify, and should not ask about.
Twelve, not one
The annual offsite is one week of twelve months. We book the other eleven — 6 people from one city who mostly do not work together, a venue held and paid for, once a month.
What a seat is, as a share of the budget
THE TRACK IS MEDIAN HR SPEND PER FULL-TIME EMPLOYEE PER YEAR, $2,479 · SHRM CHRO BENCHMARKING 2025, N = 2,371
Somebody in your building will do that division. Better that it was us. The whole price list works through the two caveats: a seat is not an employee, and this is benefit money rather than software.
Everything it argues from
Finds ·Remote workers spend one more hour a day alone.
Does not show ·Measures a population. Tests no remedy.
Finds ·Collaboration networks became more static and siloed.
Does not show ·One employer, in 2020, under a forced shift.
- 03Peer-reviewedRajkumar et al., Science (via MIT IDE), 202220 MILLION PEOPLE, AROUND 2 BILLION CONNECTIONS
Finds ·Moderately weak ties drive the most job mobility.
Does not show ·The outcome measured is people leaving.
Finds ·About 2 in 10 US employees have a best friend at work.
Does not show ·Published by a firm that sells engagement surveys.
Finds ·Median HR spend is $2,479 per employee a year.
Does not show ·A median across every size and sector.
Finds ·About $850 a year per employee into lifestyle accounts. 43% are blocked by budget.
Does not show ·Averages programmes of very different designs.
Finds ·Puts the lifestyle-account median at $1,200 a year.
Does not show ·Sells the software it is benchmarking.
What none of it shows
That Offsite changes any of it. Nothing above was run on us.
That an evening returns the hour. The hour is measured; the remedy is not.
That your company resembles an average of 500,000 people.
That a weak tie keeps somebody. The measured effect of a weak tie is that it moves them.
Attendance, and how many people came back. Both of them whichever way they come out, on this page, with the method next to them. Until then the honest version is the one above: other people’s research, about something adjacent, with its limits printed beside it.
Offsite runs in New York, Seattle, Chicago and Washington.
Free to set up. Nothing is owed until you approve a cluster.
Over 25 seats we send an order form and an invoice, not a card.