At the Embassy SF, we operate a system whereby residents in our communal home can make their room available to guests when they travel. After a few iterations, our current system addresses what we observed as fairness and incentive concerns in a more naive strawman system of direct payouts. It essentially smoothes the guest income in any given month, across all rooms as a function of the price of the room and number of nights available.

The effect is mutualizing: by smoothing out our guest income, we also smooth out risk and increase the incentive to share a valuable resource (rooms). Since our goal is to enable fluid mobility, collaboration and community, this system prioritizes those features over one of competition and potential windfalls.


  • One of our goals is to cultivate communities that are supportive of people who travel, the “modern nomad” if you will.
  • We want to encourage and support fluid mobility
  • We also want space to be used effectively - we want humans in community, not empty rooms.
  • We have many different possible rooms that guests can select from, and the selection of one room over another is often somewhat arbitrary, influenced by fragmentation of availability, a burst of color in the picture, random associations, etc. this can mean that full time guest rooms, or different resident rooms, can end up competing with each other rather than working together.

Backing a room

“Backing” a room means being the backstop to ensure that the cost of a room is covered.

It also intentionally implies that although you are likely a resident, this is not the only way to cover the cost of the room.(of course, what we consider the “cost” of a room is a different conversation, and a subjective decision to be made by the house ). By default, a room is “backed” by the house (which is itself backed in different ways by the community who operates it, and the people who sign the lease and take on financial responsibility etc.).

Experiment 1 - 50-50 split

To the initial way of doing things was, if your room makes money, it goes to you. pretty simple, but not necessarily fair. What if your room isn’t chosen? How do we account for the cost and labor from the community to oversee guests, to provide them with linens and orientation and cleaning, etc. Also, culturally, that opened up a possibility where residents could actually be profiting from their rooms, outside the community.

We wanted to create a system where the house and the room backer were “in it together” - that meant if one wins, the other wins. We also wanted a system that incentivized residents to make their room available for guests when they were traveling.

For a year or so we had a model where it was 50-50 split between the resident and the house for all guest income to their room. This aligned incentives, and covered the cost to the house of managing the guests for those rooms, but actually ended up giving too much to the house. This was a disincentive for the residents because they weren’t recouping enough costs, and it over-compensated the house for what we ultimately decided was more of a fixed cost per booking. It also often felt that it was very random which room was chosen (and therefore made income) versus others. Sometimes it felt like the rooms were in competition with each other, not collaboration, which was also a distraction from our goal of maximizing people in the house.

Experiment 2 - Proportional Distribution

We thought - shouldn’t it be more of a unified front, where we pool the resources we have, and treat them as equivalent?

If we as a house taken on these guest rooms, then we are presumably indicating a desire for and value of those spaces as a unified whole. So if there are two rooms available tonight, and one receives a booking, then actually, it would be more fair if those rooms split that income, right?

There are some caveats - some rooms are different prices. So the split should actually be proportional to the cost of the room. We can do that.

Also, there should be a fixed fee that goes to guest ops. Based on several years of occupancy data and guest ops experence, we set this at $30/booking.

Finally, the split should be a function of the availability - so of course if your room is available for 4 nights, and another room for 30, then your “share” should be proportional to availability as well.

All this can be done with the occupancy data we have for the rooms. Currently we’re using a spreadsheet to make these calculations, and room backing is in the progress of being built directly into modernomad.

Policy questions and observations

  • Oversupply - how do we make decisions about when or if there’s too much guest room supply for the demand we have as a community. is it fair to limit the volume of guest spaces? how do we do that fairly?
  • Advance notice - in the extreme, if someone makes their room available the same day, there is almost no opportunity for it to be booked, and this will always translate to a burden, not an earning opportunity, for this system. we don’t have a good way to police this right now systematically, but can cetainly build a way to track this. in the meantime, it’s an honor system.
  • Availability - our broad assessment was that we took on the house as a whole, with all the variances in the room features and qualities it entailed. so this presumes we agree to take on equal risk for all rooms (NOT that they are equally desireable, just that we want to equally support them. for example someone might have taken a worse room because it was good for the community, not because they are arguing it is better). but there ARE scenarios where a room not being rented might be a genuine sign of a problem with it (messy, unloved/unmaintained, etc.). we should have a way to handle that - at the moment it’s on the human layer - eg. we have a conversation about it.
  • Guest ops fee - this is subjective, and requires some tweaking.
  • Where does the money go? to offset next month’s rent? (currently this is what we do at Embassy SF). alternatively it could go into cobudget. or be paid out directly.
  • Since the house is the backer of many rooms, what does it mean if those rooms don’t cover their cost? this presumes the house has some kind of reserves or can float the ups and downs due to variation in guest income, which is more likely if a house has been operating for some time. if not, then “house” backed may fall back to the people who are backing the lease/ownership itself.

Try it out

Room Backing Calculator

I made a little room backing calculator that shows how the number of nights you are home vs. away would translate into savings in this proportional backing system. It shows for a given month, as a function the number of nights your (hypothetical) room was available, what you would end up paying, and how that compares to what you would have paid as a regular full time resident, and as a guest.