Now that HomeExchange is truly the behemoth house swap network, perhaps it’s less important that I represent their listing counts accurately. We can say with confidence that this network is at least 20x bigger than the next largest network. Still, I think this is a worth assessing, both for HomeExchange, but also for the methods I should use with other networks.
First a bit of back story: I put a lot of work into maintaining an up to date spreadsheet of all the home exchange networks. Some networks have purposely published misleading membership claims. So I don’t want to just trust the claims on their websites. In addition, I think it’s not fair to compare free networks with paid networks if members in the free networks are inactive but are counted in the totals. I’ve tried to come up with fair ways to count truly active members in both free and paid networks.
HomeExchange claims 400,000 listings. But it’s free to join, so there needs to be a way to assess how many members are actually active. It turns out, HomeExchange themselves has put a lot of thought into this question. They are particularly concerned about only presenting active members in search results. I got some very useful data from the folks at HomeExchange. And I was surprised and impressed with how much work they have done around this issue. Though a bit less impressed with their communication of this information with members. Communication does seem to be the HE achilles heal these days.
I’m super appreciative with how open HomeExchange is in sharing this information with me and letting me publish it. If I were them I’d be sharing this info with their members. I guess tangentially they are, through me 🙂
How to count active members
If membership must be paid each year, I think it’s fair to count those folks as active. Sure, some percentage of them are going to stop using the network before it’s time to renew. Or sign up and then never complete their profile. But they’ve paid money for the service and have to do so each year. The challenge comes with free networks.
Some free networks (i.e. SwitcHome) limit the active lifespan of memberships. So you join for one year, and you must renew at the end of that year. It’s still free, but you have to actively opt in each time. This is a good way to ensure members didn’t join and then decide not to use the network but are still being counted in membership numbers. I think most free networks don’t not require any activity to keep your account. I can see some advantages to this decision, maybe someone will come back in a few years and get going again. But it means we need another way to count active membership.
Fortunately HomeExchange has already accounted for this problem. Accounts that have not been interacted with in the past year (at a minimum, the user logs in) are removed entirely from all search results and not counted in the listing totals. This means the 400,000 homes that HomeExchange claims are actually all from people who have used their account in the past year. That’s not a perfect metric, but it’s certainly more compelling than the networks that count old inactive free accounts years after the last time the member logged in.
Still, because it’s free to create an account I’m not sure logging in is sufficient for me to consider a listing active. I’m sure some people get started creating a profile and never finish. They lose interest, or decide swapping isn’t for them. And many of those profiles aren’t completed. Everyone who uses the HomeExchange network is familiar with the many listings lacking pictures. I never contact these folks. I just assume they aren’t actually interested if they can’t be bothered to load pictures of their home. Some of them are probably new members still building their profiles, but I don’t think these folks ready to make a swap agreement either.
Some number of those 400,000 people may have just started to create a profile and then abandoned it. There are currently 255,000 members who have completed 60% of their profile. At 60% of a profile completed I don’t think someone is ready to do exchanges. So for the purposes of actually finding a swap partner, some number less than 255,000 are active in the HomeExchange swap community.
HomeExchange currently has two filters for use in searches that might be proxies for active members. These are: >80% Response Rate, and Homes with Pictures. I wrote about an attempt to calculate percent of active members in Budapest using these filters. I concluded about 32% of the listings in that city were active. This was before the integration of HomeExchange.com members into GuesttoGuest. Based on my recent discussions with HE folks about active members, and the changes they have made to the website, my previous calculation is no longer relevant and it can’t be replicated. (See below on the way they now display search results).
Here are things that I’d say are a reasonably proxy for active membership:
- Payment for membership in a network
- Actively opting in to membership in a network in the past year (i.e. the SwitcHome model)
- Responding to a swap request or sending out a request within the past year
- Creating a new, complete profile within the year (including pictures, not including calendar availability)
To me, options 2, 3 and 4 seem comparable to paying for membership. And so this is a fair way to measure active membership of free networks comparably to the paid networks. This will still capture some folks who aren’t really active, but those folks will exist on paid networks too.
Odd decisions around search results
On the new HomeExchange network searches only display 19% of members by default. This is the percentage of members who have filled in at least some calendar availability. If your search returns fewer than 20 results it will start to show members without anything on their calendars, so these people aren’t entirely invisible. But in big cities they are basically excluded from the search results.
I think this is overly restrictive and I’m not really sure why HomeExchange made this decision. I definitely think more than 19% of HomeExchange members are active. And I know lots of people are confused about why their home doesn’t turn up in searches. If you’re open to offers any time, perhaps it doesn’t make sense to have calendar availability listed.
An interesting finding with these 19% results displayed (currently just over 76,000 listings). A quick test with the Homes with Pictures filter shows that 9% of these folks don’t have pictures. So they’ve filled in calendar availability but haven’t added pictures to their profile. I agree with HomeExchange that filling in the calendar strongly suggests someone wants to be an active house swapper. So I’d guess these are mostly new folks who are still working on adding photos.
Number of exchanges is the real metric
I know house swap networks don’t publish data on how many exchanges are completed each year. But I think this is a far more interesting metric than number of listings. And in fact I’d also want to know the number of nights exchanged. Some networks have folks tending towards the longer term swaps. While those with points-based exchanges are probably going to have more short swaps of just a few days. Maybe these should be comparable, but for folks looking for long term exchanges, this distinction really matters.
HomeExchange was forthcoming with their numbers without me even asking for this data. They have more than 250,000 exchange overnights finalized each month. That’s a count of nights being exchanged. If you add that up over the year it means about 3 million nights of home exchange are arranged through HomeExchange each year. That’s an impressive number!
In the end I still don’t know what is the right number to use for active members in the HomeExchange network. The 400k who have logged in to their account in the past year, the 255k who have completed at least 60% of their profile, or the 76k that has filled out their calendar availability. I have decided that for now I will display 255k in my spreadsheet but I think with some more data I could come up with a better answer.
The folks at HomeExchange have agreed to chat with me again in six months so we can dig into this whole question more deeply after the dust settles from the HomeExchange.com integration.
Keep in mind that I’m applying more rigorous criteria to HomeExchange than I do to smaller free networks. I know some of those have lots of listings that have been inactive for a long time. And some have lots of incomplete profiles. But I don’t have a good way to calculate this unless those networks share the numbers with me.
I invite all networks to start publishing information about number of exchanges and exchange nights completed in the past year. And I encourage the free networks to be aggressive about removing inactive listings from counts and search results.