May 2017 Note: You probably want to go here instead. This page is largely archive now, because each map required some manual work. It will remain so unless I find an automated mapping solution. However, I now maintain a more regularly updated and much more extensive collection of data for download at this other page, and you can do your own mapping from there.
Sep 2016 Note: The comments section has become unwieldy: if you want me to collect data for a city not listed here, please do so here. Also note: I do know that Sri Lanka and some other areas are not “cities”, but I retain the name because that’s how it started.
Here are surveys of Airbnb cities converted to Google Fusion Tables. Fusion Tables include a map and a downloadable spreadsheet (Choose File > Download), and the ability to filter the data. For many maps the “neighbourhood” column holds the most useful spatial partitioning.
Data is obtained to the best of my ability, and for most surveys I believe it is statistically good enough for public policy discussions, but I can make no guarantees as to accuracy. Methodology details are here, and the source code is available here. From time to time, changes in Airbnb’s web site or operational problems in my code can make individual surveys inaccurate, so you should treat individual trends and detailed results with care.
Other Airbnb mapping pages can be found for Amsterdam here, for New York and Portland (and others) at the excellent Inside Airbnb, and for San Francisco at the Anti-Eviction Mapping Project.
Each link opens a new tab with a Google Fusion Table page, where you can download the data as well as filter the data in various ways. Please note that (for those spreadsheets reporting it) “income” is a very rough estimate obtained by multiplying the price by the number of reviews — it far underestimates actual income but should be proportionately reasonable over the city as a whole. The listings are colour-coded by price (red is over $300US, yellow is $200 – $300, green is $100 – $200, blue is $50 – $100, purple is under $50US).
The list here is not a complete set of the data sets I have. Contact me (tom at tomslee dot net) with requests for more cities or repeated surveys for a single city. Suggestions for improved data/presentation are always welcome.
I have put comments in [square brackets] after some of these. Those with [GIS] are broken into neighbourhoods using a Shapefile of the city boundaries and are more useful than those without, as searches can return listings well outside a given city and as the “neighbourhood” column for those without a Shapefile are not useful. If you have access to a Shapefile of city boundaries, it would be helpful if you could provide it with the request.
Asia
- Beijing (August 2015)
- Busan (October 2016) [GIS]
- Hong Kong (December 2015)
- Seoul (May 2016) [GIS]
- Shanghai (May 2016)
- Singapore: June 2015, June 2016 [ GIS], October 2016 [GIS]
- Sri Lanka (April 2016) [GIS]
- Tokyo (August 2015)
Australasia
- Auckland (August 2016) [GIS]
- Brisbane (January 2016)
- Christchurch (August 2016) [GIS]
- Melbourne (February 2015)
- Sydney (August 2014) Sydney (January 2016)
- Wellington (August 2016) [GIS]
Europe
- Aarhus (October 2016) [GIS]
- Alpes Maritime (September 2016) [GIS]
- Amsterdam (May 2014), Amsterdam (December 2015) [GIS]
- Barcelona (Sept 2014), Barcelona (November 2015) [GIS]
- Bergen (February 2016)
- Berlin (May 2014), Berlin (November 2015) [GIS]
- Bologna (February 2016)
- Bordeaux (March 2016) [GIS]
- Bremen (January 2016)
- Brno (October 2016) [GIS]
- Brussels (March 2015)
- Budapest (March 2016)
- Copenhagen: February 2015, June 2016 [GIS]
- Dublin (March 2016), Dublin (August 2016) [GIS]
- Edinburgh (October 2015)
- Estonia (July 2016)
- Florence: October 2015, September 2016 [GIS]
- Groningen (March 2016) [incomplete]
- Hamburg (January 2016)
- Helsinki (May 2016) [GIS]
- Iceland: July 2016 [GIS]
- Istanbul (February 2015), Istanbul (January 2016)
- Lisbon (March 2015), Lisbon (March 2016), Lisbon (June 2016) [GIS]
- London (May 2014), London (January 2016) [GIS]
- Lucca (August 2015)
- Lyon (May 2016) [GIS]
- Madrid (February 2015)
- Mallorca (March 2016) [incomplete]
- Menorca (April 2016) [GIS]
- Marseille (February 2016) [GIS]
- Milano (Nov 2014), Milano (Jan 2016) [GIS]
- Moscow (Oct 2016) [GIS]
- Nice (June 2016) [GIS]
- Oslo (February 2016)
- Palermo (Sept 2016) [GIS]
- Paris (May 2014), Paris (Dec 2015) [GIS]
- Porto (October 2016) [GIS]
- Prague (March 2016) [GIS]
- Reykjavik (April 2016) [GIS], June 2016 [GIS], Sept 2016 [GIS]
- Riga (December 2015)
- Rome (May 2014), Rome (Nov 2014), Rome (August 2015)
- Seville (July 2015)
- Spoleto (March 2016)
- Stockholm (January 2016)
- Switzerland (August 2016) [GIS]
- Tallin (July 2016) [GIS]
- Tuscany (August 2016) [GIS]
- Valencia (January 2016)
- Veneto (November 2016) [GIS]
- Venice (November 2016) [GIS]
- Vienna (August 2015)
- Zurich (January 2016)
North America
- Albuquerque (December 2015)
- Anchorage, Alaska (May 2016) [GIS]
- Asheville (September 2016) [GIS]
- Austin (March 2015)
- Boston: Sept 2014, Sept 2016 [GIS]
- Cambridge MA (Sept 2016) [GIS]
- Charlottesville (May 2016) [GIS]
- Chattanooga (July 2015)
- Dallas / Fort Worth (February 2015)
- Denver (August 2015)
- Jersey City (May 2016) [GIS]
- Joshua Tree, CA
- Kitchener-Waterloo (May 2016) [GIS]
- Lincoln County, OR (May 2016) [GIS]
- Los Angeles (August 2014) and August 2016 [GIS]
- Miami (August 2015) [GIS], Miami (January 2016) [GIS]
- Montreal (May 2014), Montreal (October 2015) [GIS]
- Nashville (May 2016) [GIS]
- New Orleans (January 2015) [GIS], New Orleans (March 2016) [GIS]
- New York (May 2014)
- Newport, OR (April 2016) [GIS]
- Ottawa (May 2016) [GIS]
- Peel Region (Sept 2016) [GIS]
- Philadelphia (February 2016) [GIS]
- Portland OR (October 2014)
- Portland ME (July 2016) [GIS]
- Saint Lucia (September 2016) [GIS]
- San Diego (May 2014)
- Santa Fe (December 2015)
- San Francisco (August 2014)
- Savannah (May 2016) [GIS]
- Toronto (June 2015), Toronto (March 2016) [GIS]
- Vancouver (April 2015), Vancouver (March 2016) [GIS]
- West Indies (September 2016) [GIS]
- Yellowknife (September 2016) [GIS]
Hi Tom,
That is pretty impressive work. Would you be able to do one for Sydney for me?
Thanks
Robert
Added at the end, and I’ve sent you a bit more by email.
I’m really interested in how you acquired this data. Did you directly query the database or did you use a scraper? Could you share your method/code? Very cool stuff 🙂
Thanks! I used a scraper — my code is posted at https://github.com/tomslee/airbnb. I’m no expert so would appreciate tips/improvements.
Hi Tom, this is amazing! I have been trying to figure how to grab the code (assuming that is OK?). Would you mind doing Brisbane (Aus)?
Hi Again Tom, you work has attracted a lot of interest and some questions. Further to my previous request is it possible to do all of Australia and add into the data the host profile number?
Hi Michael,
You are welcome to grab the code, although I did have to make a couple of fixes to get it work for Brisbane (and other places that don’t have “neighborhoods” listed on the search page) and I haven’t uploaded that yet.
But here is a map and KML file as a Google Fusion Table. The host_id is in the data but not in the caption above each icon. You are free to take it and change it however you want, and I can send you a spreadsheet with a few more columns if you want — just send me an email with your address (tslee at web dot ca).
I think Brisbane looks pretty accurate. My map has 17 shared rooms, 407 entire homes, and 600 private rooms. Looking at the Airbnb search page for Brisbane claims 6 shared rooms, 397 entire homes, 605 private rooms. There’s probably a little fuzziness in which listings show up so it looks like I have almost all.
Finally – I thought it would not work for the whole of Australia, but as you can do a search on the whole country with the same filters, maybe it will work. I’m running it now — it takes a couple of days — and I’ll post an update when it’s done.
Thanks Tom, this is much appreciated. Your work has generated a lot of interest locally.
Happy to be of use Michael. My script won’t manage Australia as a whole, unfortunately. I hope to do a round of enhancements and I’ll try to get that in, but it may be the new year before I can get to it.
Wow this is pretty incredible stuff! I would love to see a map like this for the city of Portland, OR. I wish I had the know-how to follow your code and whip something up but alas, I come to you for help.
Thanks!!
Sorry to take so long to approve your comment. here is Portland, done over the weekend. I could send you a spreadsheet with more detail if you’re interested. Send me an email.
Thanks again Tom, have looked at Brisbane and some of the fuzziness is likely around the geography. the map is capturing some surrounding areas which are not really Brisbane but not impacting to any significant degree.
If you are inclined to do all of Australia, great, and naturally whenever you are ready. If you could extend the fields to include the states that would be very useful.
I am not sure what your interest or motivation in this is but for mine the inclusion of the host details is very useful and would be terrific is we could do the same for Sydney’s if you are inclined to do so.
Similarly if you wish to have a look at Melbourne and the other state capitals there are lot of individuals who would be interested in the results.
My own skills are not sufficient to do this.
HI Michael,
I only just stumbled across this data! I’m interested to know if you got anywhere, so it would be good to hear from you!
This is impressive.
Could you do Berlin, Germany? I heard it is one of the most popular airbnb cities?
Thank you!
Sorry for the delay Sebastian. I’ve added Berlin above, and the map is here.
Hi Tom –
Great work, can you confirm the currency used for the prices, USD across all cities?
Thanks,
Yann
The prices are actually Canadian Dollars; Airbnb seems to detect my location and return in that currency.
Hi Tom, i’m an italian student of Tourism at the Univerity Bicocca of Milan. Could you explain me the meaning of the points colors? The different points colors are random or have a specific meaning?
Thank you and good job.
Ester
Hi Ester,
The colours show the price of the listing per night. I see the prices in $Canadian: Red is over $300, Yellow is $200-300, Green is $100-200, Blue is $50-100, and Purple is $0-50.
Hi Tom,
Your site is amazing, very valuable information! I am trying to size the Airbnb market share, but as you know much better than me, they are not very open about their data… Especially I am interested in guests per city per year. Do you know of any data on this?
Thank you!
Sven
Hi Tom,
your maps are just AMAZING. Really shoking.
I don’t know whether it is easy to you to do the maps but it would be really interesting to have the one from Barcelona since it is becoming a crucial political issue.
The fact is that Airbnb are telling they have major presence in the peripheral areas but the dataset I have made at the neighbourhood points to the concentration to the Old City Area (the most overcrowded in the city).
Just saying that it would e very useful for the political discussion if we could visualize the Airbnb listings for all the city.
Thanks a lot.
Regards!
Albert Arias
Thanks Albert!
I have added a link to a map for Barcelona, which I retrieved in September 2014. I can run a new one soon if you want.
I have seen this kind of thing before: in Montreal they claim to have listings all over the place, but half of all visits are in just two neighbourhoods, which are ones with a big tourist presence.
If you are interested I can send you a spreadsheet with more attributes (number of reviews, etc). Just let me know by email (tslee at web dot ca)
The detailed counts per neighbourhood are not always reliable, but if you have a way of getting a file of Barcelona district boundaries then I could get fairly accurate counts. An ESRI shapefile perhaps from an open data site? I tried going through the Barcelona Open Data portal but my complete lack of Spanish stopped me 🙁
Thanks a lot Tom!
I write you a mail today and we keep on talking about shapefiles and other things related.
Cheers!
a
Hi Tom,
I’ve been aware of and admired your airbnb data crunching for while now (in print) and I just found that you have this blog!
I’ve also been following new york’s battle with airbnb (i used to live there). We have our own nightmare at ground zero in San Francisco, however most of our politicians are bought and sold thanks to curry favor with the disruptors.
We have been doing similar analysis in my little collective. Yours is more thorough though. May I suggest you try the map product CartoDB for prettier maps? You can also use your coding skills to animate and stuff if you are so inclined. (No I dont work for them. )It’s free and here is a sample: https://www.antievictionmappingproject.net/airbnbmap.html
I’ve been looking for a better mapping interface. I’ll definitely give it a go. The anti-eviction work looks great: congratulations on it. Happy to collaborate if there are ways to do so.
It would be really great to collaborate. I will email you soon.
Are you assisting the folks in NY’s Share Better campaign? Cuz I’m in Share Better SF and I hear we share information with each other.
I am looking at your tables for this map and wondering what “income” means exactly? Did you calculate that based on the number of bookings in the future? Is it an estimate?
Also what does rank mean? If I missed the explanation somewhere, I apologize…
Yes, income is an estimate. The absolute value does not have any meaning, but relative values might. It’s just the nightly price times the number of reviews, so if there is a pretty constant average visit length and a pretty constant ratio of reviews to actual visits then it would give a useful relative idea of which properties are being rented. Ranking is just the order of the incomes.
Hi Tom
I’m thinking of joining the herd and trying out AirBnB mostly because of I’m nervous about the inflexibility of having fulltime tenants, especially with the potential legal issues that come with 30-day tenancy in Ca. I have some questions for you : )
How did you come up with the total income for these individual listings? Where does one find that?
Will you be doing an extension of the SF map to include Marin soon? And did you do/will you please do one for Palm Beach Island FL?
Finally, the stories on AirBnb Confessions almost made me gag. Do you think they’re really true? Or do you think most hosts and guests participating in this are more decent that that?
Hi Tom,
The maps and studies are amazing. Could also preapare for Istanbul? Thank you in advance.
Selim
Hi Selim,
You’ll see the map listed above now. I was surprised how many listings Istanbul has — over 7,000. A lot of multiple-listing owners too. I can send you a spreadsheet with more fields if you are interested.
Hi Tom,
The stuff you have shared here is awesome! I dont understand the “scraper” or how to code but might do a little learning and give it a shot. Any chance you could help me out with Dallas? I am working on a project and it would be a great help!
Best, Dave
Hi Dave,
You’ll see the map listed above now. Dallas has surprisingly few listings: only 539. I can send you a spreadsheet with more fields if you are interested.
Hi Tom!
The maps look amazing again. I was wondering if you are willing to share spreadsheets of data of European cities that I want to use for research. It would really help me out! How many years of observations have you been collecting?
Kind Regards,
Thomas Mutsaerts
I’ve given Istanbul its own page here, accessible from the front page of the blog on the “Airbnb resources” menu. I am hoping to post a bunch of city pages with the same kind of basic information. I’d be happy to add to the reports if there are requests.
Hi Tom,
Amazing stuff you’ve done here, well done!
I’m trying to put together more of a time series look at airbnb data for San Francisco. I was wondering if you had happened to save more detailed output from your August 2014 scrape? Also, were you able to de-duplicate multiple listings? (I’ve noticed that some hosts will put the same pictures/descriptions up with different listing IDs).
Best,
Alex
Hi Alex, and thank you.
I have three surveys of San Francisco details, from November 2013, May 2014, and August 2014. I’d be happy to send you spreadsheets of each if you want.
I haven’t made any effort to de-duplicate. I do see that some people offer a “private room” and then also offer the “entire home or apartment” as a separate listing, but for some purposes it makes sense to keep them separate. The spreadsheet does have latitude and longitude.
I’ll send you a separate email over the weekend.
Hi Alex,
I have a colleague in SF, Guss, who was also scraping the website over the last couple years. If you contact him, he might have more snapshots in time, as well as Tom’s to fill out your picture. Email him at his blog: https://darkanddifficult.com/wordpress/?p=5004
Also please send me an email of your finished product if you want it publicized. ([email protected]) I’m part of the ShareBetter SF coalition. We love data viz!
Guss gollects more detail than I do as well, so definitely worth reaching out to him.
Jenn and Tom, you all are lifesavers, thanks again. I’ll certainly follow up with whatever I end up with (this is part of a capstone project for a masters degree) hopefully has some cool data vis stuff as well but is mostly focused on critiquing the regulation.
Hi Tom,
As many readers already mentioned, congratulations for a truly remarkable work!
I am interested myself in understanding the footprint of Airbnb, and was trying to gather information online when I landed on your Github page, which is very cool thanks for sharing.
In your opinion, would there be a way to have a global overview of the geographical footprint of Airbnb? that is getting a high-level view of how many listings per country (they mention ~200 countries), and largest cities in these countries (without necessarily getting the details of each listing). I’m trying to assess the size of Airbnb in countries other than the ones they usually talk about (France/paris, UK/london, ..), and also the penetration in each country beyond Capital cities
Do you think your code can be twisted to do that?
Many thanks in advance for your input!
Best
Hi Bruno,
Good idea. It turns out I can, but it will take a little while. The room_ids are integers with a maximum value of about 6 million, so I can just pick random numbers from that range, go to the site, and if the room exists save the information about it.
It will take a little while – I figure I can get about 150 or so listings an hour while staying within Airbnb’s query limits (like many sites, they shut you down if you throw too much traffic at them, which is fair enough). So I should be able to get about 3,000 a day. It may take the best part of a month to get a really good data set, but maybe by the weekend I’ll have something to show. Watch this space and thanks for the suggestion.
Hello again Tom,
I’ve been following on this thread’s activity for the past few weeks and it seems there is a lot of traction going around your work, congrats!
I was wondering whether your attempt to “map the entire Airbnb” supply had been successful, or it was a bit too ambitious after all?
Thanks and good luck on your book (if you find time to write it!)
Cheers
Hi Tom!
Thank you for the great work, I really appreciate it!
Is there any chance to have insights about the number of AirBnB listings in each of the approx. 300 biggest cities in the world?
I figure Bruno has already asked kind of the same question.. do you already have results?
Thank you in advance!
Hi Tom,
Really nice work I must say! I’m currently writing my bachelor thesis on the sharing economy and how AirBnB affects the hotel market in Copenhagen, is it possible to extract the data from that city as well?
Regards,
Viktor
Hi Viktor,
No need for the compliments, but Copenhagen is now listed above. I’d definitely be interested in any findings you make.
Congratulations, your work is awesome! I am a student in Data Journalism and since I have been living in Brussels for some time I did notice how the Airbnb phenomenon is most probably bound up with the ongoing gentrification is some areas of the city. It would be great to actually double check this feeling with actual data…would be possible for you to get data for Brussels as well?
Thank you in advance for your time
bests
Antonio
Hi Antonio,
Thanks for the compliment. Brussels is now on the list above. I’d be really interested in anything you find…
Hello Tom,
Congratulations, this is amazing. I am currently a student and working on a research project about Airbnb in the city of Vancouver. Do you mind to help me with that info?
Thank you in advance!!
Monyka
Tom, I want to clarify that I am talking about Vancouver, British Columbia.
Thanks,
Monyka
Coming soon: next week, I hope. Meanwhile, you may be interested in this Airbnb timeline by your fellow Vancouverite Karen Sawatzky: https://karensawatzky.ca/2014/10/29/a-timeline-of-housing-related-airbnb-events/
Thanks!!!
Hi Monyka,
Vancouver is now added. Best of luck with your research.
Many many thanks Tom for the Brussels dataset,wonderful!
Tom, Thank you for sharing your knowledge and helping everybody. This is a real example of the Sharing Economy!!! 8o)
Hi Tom, great job on putting together this scraper! I’m definitely going to give a whirl on some more rural areas in Canada, might yield some interesting data there.
I was wondering if this scraper grabs bookings data as well? Like x/365 days are booked on their calendar? And if so, is there any chance that data for Vancouver could be posted or sent over email? It would be very much appreciated.
I’m very interested in mapping just booking data and perhaps using an interpolated heatmap to show which areas have airbnb businesses that are more packed. Ideally, I’d like to pinpoint areas with an overflow of bookings in certain times of the year.
Anyway, great job! Keep it up. Thanks for taking the time to read this too. Cheers!
Tom! This is truly incredible, I’m a data scientist myself but haven’t mastered scraping websites the way you have!. I’m very interested in researching Barcelona (SPAIN) prices by neighborhoods. Does your code extract the neighborhood associated with each property? I got hold of Barcelona’s neighborhood boundaries (KML), which I guess I could use to isolate the coordinates on the spreadsheet you provided. But I wonder if your code is able to do that automatically!
Your work is amazing!
Thanks.
Rom
Hi Rom,
I also work with Airbnb maps and data (https://insideairbnb.com). I’m working on a map for Barcelona, and need a neighbourhood Shape File for the city – are you able to share yours (or let me know where you got it)? In exchange, I could help you use the shape file with Tom’s data, or provide you with some alternate data coded to the neighbourhoods.
Feel free to e-mail me directly: murray [at] murraycox [dot] com
Hi Rom and Murray,
I have a shapefile of Barcelona neighborhoods that I got from Albert (search his name in the comments above). I have uploaded it for convenience: I hope this link works. I do have a breakdown by neighbourhood, so if you are interested in comparing notes, email me and we can compare queries/results. tom at tomslee dot net.
Rom: you may want to contact Albert – his interests were similar. It’s always interesting to find out what people who actually know a city see, so please keep in touch.
Thanks Tom, much appreciated.
I also tried trawling through some Barcelona and Catalonia government Open Data web-sites, but could only find shape files of districts, not Barrios – so thank you.
Is there a source on the shape file so I know who to credit when I use it?
Hi Tom,
I recently learned of your work through an Airbnb study published by LAANE in Los Angeles. Your work is truly remarkable, and I appreciate the accessibility of this information.
My organization is currently working on a similar study about Airbnb in Oakland, CA. Would it be possible for you to get the data for Oakland, as well?
Thank you in advance!!
Mia
Hi Tom,
I have been using the data from Vancouver, BC. I am trying to limit my analysis of Airbnb to certain areas of the city. I would like to know if you could share with me neighbourhoods and any other field in which I can have more information about hosts and reviews.
Regards,
Monyka
Hello Tom,
I´m impressed by your great job. I´m very interested in LISBON, MADRID, SEVILLE.
Are these ones in your plans?
thanks
cesar
Hi Cesar,
Lisbon and Madrid now up. Seville coming soon I hope.
Can’t find anyone local to help with data right now – legalization of STRs proposal comes to Berkeley Council on June 9 but then will have a few weeks / months of Planning Commissionn etc.
Any chance you can help with stats on whole units, absentee owners, multiple whole units listed, here in Berkeley?
Hi Katherine,
I can’t get that kind of stuff. You may want to reach out to the people at Anti-eviction mapping (https://www.antievictionmap.squarespace.com/). From the little I have had to do with them, they are good people.
Hi Tom,
Thanks for the shoutout to the mapping project, but I’m afraid I sent Katherine to you. Our Guss is on vacation right now… I know that his scrapes are different than yours; are yours just lat and lon? Are can you do “entire unit” counts and lists by the same host ID?
Hi Tom,
Amazing stuff you are doing, keep up the great work.
Can you do one for Singapore?
Singapore added. Let me know if you find anything interesting.
Hello, impressive work!
Are you planning to get another spanish cities? Madrid?
Madrid is there now. Seville coming soon.
Seville now available (not too many listings)
So the praise is nice and appreciated, but from here on I will delete it, so don’t bother. Keep on asking for cities though.
Hi Tom,
Would love to see some data on Oakland, CA. The city recently entered negotiations with Airbnb to collect transient occupancy taxes. Information about the amount of “entire” units listed, impacted neighborhoods, and generated revenue would greatly help with a campaign to get some of those funds for affordable housing development in Oakland.
Wonderfull, thanks so much.
Cesar
Hi Tom,
I pulled the data for Boston (July 2015), and I’m showing that listings doubled since your pull in September, 2014. Did the geographies change? If not, this is huge. I wonder how much they’ll increase, and if you’ve seen huge increases in any other cities.
Thanks
I did not see that volume of changes. If you used my code you may need to include “and deleted = 0” in any query you do.
Hey I would love to see Toronto listing as well!
Done.
Hello Tom,
I’m a guy from Mallorca, Spain. Here tourism it’s very important. For these reason I would like to analyse a turistic resort called PLAYA DE PALMA. Could you do this work? I will use in a college project. Thank you verey much. I admire your maps.
I’m from Chattanooga, Tennessee and we’ve been seeing a lot more of these go up in the city. I’d love to see a map of where in the city they are all happening! Thanks!
I don’t see very many. They are listed now.
Dear Tom,
as all the previous commenters, I wanted to thank you for this great work. I am currently working on a paper on the impacts of AirBNB on the Viennese housing market. I was wondering, if you could add Vienna in Austria. I would be very interested in the data!!
Best wishes,
Nina.
Any data on Austin Texas ?
We could use your help
Sorry for the slow response on these requests. I hope to post a few in the next couple of days.
Can someone do Austin Texas? 78702 and other zip codes?
Sorry for the slow response. I’ve posted one of Austin from March. Hope that is recent enough.
Hi Tom – not sure if you’re aware but it seems like your “airbnb” project has disappeared off Github?
Only the superceded “airbnbpy” remains
Yes. There are a few reasons. One is that I’ve been moving to PostgreSQL running on Amazon Web Services RDS, and so there will be some changes soon. Along with that I posted a version that had a password in, which I have now changed but it gave me the creeps. But mainly, Airbnb is getting more aggressive about blocking scrapes — my home IP has been blocked from the site for weeks now. So posting things on GitHub seemed maybe too friendly. I am happy to share what I have and may well post it up on GitHub again once it is in a clean state. Email me if it’s a problem for you.
Aha – yes I noticed I’ve been getting repeated 503 errors too after a while, assuming these are Airbnb blocking the scrapers?
I was using a free basic AWS instance and figured they would probably block anything coming from there pretty quickly.
I’ve kept a working copy too and have been modifying it to for instance randomise User Agent headers and building in more delays. Not sure if it’s helping much… Time to look at incorporating an IP randomising service maybe.
Hi. Thanks for your work. I also noticed the disappearance of the airbnb folder but I kept a working copy of the main script which I am modifying to work with psycopg and Postgres. Could it be possible to get the doc to use this script (the use of -asa, -asv and the right order) please ?
Sending a separate email.
Tom,
This is fantastic. Would you be willing to help me out with a Denver, CO map? We have some issues coming to light in the local city council and that would be helpful.
Thanks!
It is taking some time now to accumulate listings. I hope to have this by the end of the week.
Denver is now in the list. Only 1500 listings, which surprised me. I wonder if something in the search led to me missing some? Please do let me know if you come across obvious discrepancies (like many listings you find on the site that are not included in the data set here).
Hey Tom! I really don’t wanna be annoying, I just wanted to ask whether you think it would be possible to get the data for Vienna, Austria in the next time. Otherwise I guess I have to find another way,
Really appreciate your work! Thank you! Best,
Nina.
Hi Nina, sorry it’s taken so long. I do have a dataset for Vienna completed, and will post it this evening.
Vienna is finally posted. Sorry again for the delay. The format is slightly changed – there should be more columns if you download – but otherwise it should be the same.
Dear Tom,
Thank you very much!!
I was also looking at the insideairbnb website and the available data there – and noticed a difference in the number of listings. Could it be that in the file above also includes some of Vienna’s sourroundings? I guess, I should just check with a shape-file of the city.
One more question, I would like to ask: It is not possible, to travel back in time and to get the data from a year ago, is it?
Thanks again, the data will be so very helpful for my analysis.
Dear Nina,
It’s interesting that they differ, and not easy to tell what the reasons are. I would say that a 10% difference is not too bad. Both Murray (who runs insideairbnb) and myself have seen that a lot of listings appear and disappear in a short time — often listings with very few visits. So there is an inherent noisiness to the data, I think. (Is a host with no guests really a host?)
Unfortunately it’s not possible to look back in time. At least, I have not thought of a way to do that.
Hi Tom,
i am really impressed by your work! may i can ask you the data of Lucca (Italy) ?
Thank you very much !
Hi Acchile: Lucca is now listed.
Dear Tom,
Your Airbnb analysis has helped me very much and saved tons of time. I found a website call elance to try hiring an Indian programmer but I found your blog instead and half of the problem is solved!
I know nothing about programming and IT but you are a highly skilled / most importantly passionate IT professional. I want to make a small token of appreciation for your work if you dont mind. It would be great if us readers can donate to keep this blog running. My internet skill is extremely poor and couldn’t find your email address. Hope you can contact me via email.
Best Regards
Ken
Dear Tom,
this is very impressive work.
Is it possible to work on map in any China cities, like Hong Kong, Shanghai, Beijing?
thanks.
Wallace
Beijing is now listed. I’ll do Hong Kong and Shanghai over the next couple of weeks.
Hi Tom! Your work is absolutly amazing, thanks for the extensive analisis, i was wondering if you could do Buenos Aires.
Saludos!
Now done.
Hi Tom, sorry for asking for one more city. Could you please list Moscow? I would appreciate it so much. Thanks for the nice job.
Dear Tom,
this work is remarkable!
Please can you add a map of Florence, Italy?
Here the city center is almost empty of inhabitants who were substituted by touristic rentals (much more lucrative than normal rentals).
I read an article in which it is said that Italy represents quite the 10% of the total AirBnB offer.
Thank you very much.
Silvia
Wow – Florence is certainly very full of listings for a small city. Thanks for suggesting it. The 10% figure sounds quite possible. You can find the spreadsheet and map above.
Tom
Hello
I’ll add my voice to all those impressed by your service. I am very keen to understand how the impact of holiday/short let business on the city – particularly the loss of accommodation for regular residents who might want to rent longer term. Am particularly interested in Scottish cities – especially Edinburgh. Would be willing to make a contribution for your time/effort.
Thanks
Dan
Edinburgh added.
Hi Tom, really nice work! I’m working on a project about Antwerpen, could you help me make on for this city?
With kind regards
Demian
A bit of a problem at the moment: a search in Antwerp shows up a lot of listings in Brussels. May take another week or so to fix…
Dear Tom,
Your work is great and really useful!
I am working on Rome, Italy.
I see that you updated the data for 2014… would it be possible for you to add some more data for 2015 as well?
Thanks a lot for your great great and valuable contribution.
Venere
Hi Venere,
I’ve added an update of Rome. You’ll see it includes some listings a long way away (northern Italy). I can fix that, but need a shapefile of Rome boundaries. If you can find one, I’ll update. I’m at tom at tomslee dot net.
Thanks TOM!
I will provide you with a shapefile 🙂 thankssss you are a star!
Finally updated the August data set to have only those within the boundary of Rome. Thanks for the shapefile!
This info is great, it is very helpful for a school project I am working on. Would you be able to do a map for Santa Fe or Albuquerque?
Thanks!
Probably too late now, but these are done now.
I have a few software problems at the moment – hope to get to the most recent requests in the next week.
Hi Tom,
Sorry to add more requests but would you have any data for Whistler, BC, Canada? Love your work on this 🙂
Tom, try this:
https://www.istat.it/it/archivio/124086
You want the Comune geography – there should be one labelled “Roma”…
—
Murray
Thanks Murray.
Hi Tom,
I follow all your posts and was wondering if you could add Buenos Aires to your data set too.
Saludos!
Buenos Aires is now added.
If you’re still working on this, I had a peek at the airbnb site where it allows you to select your currency.
It seems to be set with a cookie – here are the request ‘Cookie:’ headers for EUR and CAD:
_user_attributes=%7B%22curr%22%3A%22EUR%22%2C%22guest_exchange%22%3A0.91012%2C%22device_profiling_session_id%22%3A%221450051038–729f4a3864519714b2cb7ed6%22%7D;
_user_attributes=%7B%22curr%22%3A%22CAD%22%2C%22guest_exchange%22%3A1.37582%2C%22device_profiling_session_id%22%3A%221450051038–729f4a3864519714b2cb7ed6%22%7D;
This looks like JSON but URLencoded:
_user_attributes={“curr”:”EUR”,”guest_exchange”:0.91012,”device_profiling_session_id”:”1450051038–729f4a3864519714b2cb7ed6″};
_user_attributes={“curr”:”CAD”,”guest_exchange”:1.37582,”device_profiling_session_id”:”1450051038–729f4a3864519714b2cb7ed6″};
The “guest_exchange” value is presumably a conversion rate. I was looking at St Petersburg, FL
Changing the currency to USD gives this value:
_user_attributes={“curr”:”USD”,”guest_exchange”:1.0,”device_profiling_session_id”:”1450051038–729f4a3864519714b2cb7ed6″};
Hello Tom,
first of all i like to join the numerous compliments for your great work ! I am doing research for my thesis about air BNB in Germany right now.
Could you help me out with Hamburg and/or Bremen?
Thanks in advance
Done. If you feel like telling me more about your research I’d be interested.
Thank you very much Tom, just saw your work on Germany!
The thesis is about the disruptive effect off airbnb on the accomodation sector and the strategic implications for traditional hotels etc., yet i am in the research and conception stage but i am happy to tell you more, if you are interested 🙂 just let me know
Thanks for all your work Tom. Would you mind having a go at Valencia for me? Muchly appreciated.
Done
Wow! Great stuff and thanks a lot Tom.
Hi Tom,
Very nice work!! Would it be possible to have the data on Zurich please?
Best!
Finally done.
Hi Tom,
I’d like to access the data from London you posted, however, when clicking on the link there’s only the map for London but no option to open the fusion table. Could you provide me with the data in table format please?
Thanks!
Done by separate channel.
Hi Tom
WOW. Thank you for all this amazing work. I was hoping for two things 1) Brisbane to be updated as it is 18 months old now and I believe has changed significantly in that time and 2) Do you have any data for Melbourne?
Thanks so much again for all your work.
Liz
I will run them in a few days time. Meanwhile, https://insideairbnb.com has Melbourne data.
Hi Tom,
This is amazing work!!!
I’d like to access the data from London you posted for my architectural research , however, when clicking on the link there’s only the map for London but no option to open the fusion table. Could you provide me with the data in table format please?
Many Thanks!
The London data is finally downloadable. Sorry about that and best of luck with the research.
Hi Tom,
it is a great an impressive work. I’m an Italian guy and i’d really like if you could help me with the data from Bologna – Italy
Bests
Walter
Done.
Super!! Thx 😀
Hello and great work, Is it possible to get a link to the London data table please?
Now available. Sorry for the delay.
Hi Tom, very impressive and helpful work. When you have time, would it be possible to add Marseille city in France?
Thanks,
Olivier
In progress…
Thanks.
Marseille finally added.
Thanks Tom, really appreciated. Will work on it and share the results of my analysis.
Hi Tom,
Have started to look at the data and see that many rooms are missing in some areas. Can send you exemples but if you look at “L’Estaque” part north of Marseille, your file displays 2 rooms_id when Airbnb has 4 pages with a total of 61 rooms_id.
Have selected some randomly and they are indeed missing so not a question of mapping.
Trying to understand where the difference could come from, was wondering if you extract only rooms that are not booked the day of the extract? (But probably will not be enough to explain the difference).
If you think that altogether, Marseille has 7635 hotel and residence bedrooms and your file gives 5581 bedrooms from Airbnb, we can understood the level of impact it might have on the industry.
Another peace of analysis is that we have clearly hotels having closed as such and switch to an Airbnb bedrooms model.
Will continue to dig into it and provide feed-back.
Hi Olivier,
I think L’Estaque is too far from the city to have been caught by my search. The search goes through the neighbourhoods listed on the Airbnb web site for Marseille, and L’Estaque is not included. That said, occasionally listings far from the city do get mixed in with other Airbnb search results.
I have two kinds of map here (and should point this out at the heading of the post): one is simply “listings that show up when you search through Airbnb for listings in a given city, no matter where they actually are” and the second (better) type is listings combined with a GIS map of the city, so that neighbourhoods are better represented. I will try to mark the maps.
If you want to go beyond Marseille proper, we will have to look for ways to configure the search so that L’Estaque is included.
New maps for Bergen, Oslo, Bologna, and Melbourne
Curious question… What about Rural Areas? I am located in the US (Vermont) and am curious if there is a measurable impact on rural housing? Rent in the Montpelier area is high by my standards ($500/room or $1200/house) and many people live in sub-standard mobile home trailers or other decrepit conditions.
I’m not asking for a survey of a rural area yet, just trying to spark a conversation about AirBnB impacts beyond the cities.
One other piece of context, a friend recently rented another house for the winter to try to pay the heat and cover part of the mortgage while it languishes on our sluggish housing market. Instead he had to spend $25,000 fixing damage done to the basement foundation by the tenants (insulation that led to failure of a wall). Had he had Air BnB as the broker it would’ve presumably covered it.
It is a platform, can it be used for good too?
The nearest I have seen to someone asking these questions is a very recent hospitality industry report (https://cbrepkfcprod.blob.core.windows.net/downloads/store/12Samples/An_Analysis_of_Airbnb_in_the_United_States.pdf). It suggests that Airbnb makes more of its money from big centres than other areas of hospitality company, which would suggest that rural areas will not be very important for the company. Maybe that is different in rural areas with big tourist attractions though?
Hi Tom,
Great work, very impressive.:) Could you please add Prague (Czech Republic) and also Budapest (Hungary)?
Thanks very much.
All the best,
Dora
A long pause, but Prague finally added. Budapest soon, I hope.
Hi Tom, very impressive work! When you have time, would it be possible to add Victoria, BC?
Thanks,
Andrew
Wow….. good work Tom!
What about Munich, Germany? I would be interested now and later of the year (in Oktober is the Oktoberfest – prices increase up to 500% in some accommodations).
Also a new file for Berlin, Germany would be interesting e.g. in next month as airbnb deleted several multihosts in February 2016 (they not visible anymore after 02 March 2016 as far as I know).
Thanks very much.
All the best,
Ge DE
The most recent data I have is from November 2015, which I have now posted. I will get to the others ASAP, including a new survey of Berlin.
Hi Tom,
This is impressive! I’m doing a study about this, is it possible if you can do Bali, Indonesia for me? I would really appreciate it.
Thanks!
ZS
Hi Tom, really impressive. If you’re still working on this, i cannot find the code on github, I’m doing a research on Torino,italy about gentrification, house price rising and tourism. it would be amazing if you could add the data for me. Thank you very match, tommaso
New maps for Hong Kong, Spoleto (Italy), and Mallorca
Thank you very much Tom! I am really impressed, do you think you could try to download more info about the listings in Mallorca?
Cheers
Hi Veronica,
It looks to me like the data for Mallorca is incomplete. You may wish to look at https://insideairbnb.com/mallorca/, which reports many more listings.
Hi Tom thank you very much for your prompt answer!
Yes, from the dataset I can see there is only few available variable compared to the other european cities.
Sorry for asking you again but I would be really interested in having data about Rímini (Italy). Do you thing is it posible to have for Rímini a complete dataset as for, for example, Barcelona?
Thanks again for the all the work you have been doing.
Cheers
V
Tom,
Understood the logic.
l’Estaque is within the border of Marseille (16th quarter), not a subburn or a neighborhood but true that the city is quite large.
Would you have a listing with all the Marseille city available as when you search Marseille on Airbnb and I can adjust the geographical impact to filter and send back the result?
Thank you so much, Tom; i just received the data I needed for Spoleto. Let me know how to help your project.
Hey Tom impressive work. Would you consider adding Orlando Florida to your site? I think we really have two markets here. One is in the downtown area (32801) while another clusters around the Disney attractions to the southwest of the city.
Thanks for your consideration!
great work, I would appreciate a map of Reykjavík, Iceland
thank you
-uni
Hi Tom,
Great work! Could you please add Bergamo (Italy)?
Thanks very much.
All the best,
Isaia
Hi Tom,
Really awesome work and how many people you´ve helped 🙂
I would Like to ask you if you could add Prague (Czech Republic)
Currently I am writing my bachelor thesis on rising shadow economy and its impact on Czech Tourism by screening the Airbnb
I have growth listings from May(2014) to February(2015) so 10months from (https://home-for-prague.com/) and here is the question if it is possible to find the total number of listings Before May(2014)? like Jan,Feb,Ap (2014) it would really helped me to have more roboust econometric model.
Thank you for your work Quang.
Hi Quang,
I have added a data set for Prague, but I have no way to get the data set as it was at a past date. You may want to ask Airbnb 🙂
Even though thank you for your reply and work really appreciate.
New maps and data sets for Lisbon, Bordeaux, Miami, Stockholm, and Philadelphia.
Also Budapest.
Hello Tom!
Your work is great and very helpful. I come from Athens, Greece so I want to ask you to publish dataset and map(if possible).
I used to work with GIS so I know there are very interesting things to do with the data.
A note to anyone subscribed to this page. I have found that for cities with few (or no) neighbourhoods listed in Airbnb filters, I am regularly missing listings (see Groningen and Mallorca above for examples). This affects some requests like Phoenix, Torino, Frankfurt and Orlando. I would like to implement a different search algorithm than “loop over neighborhoods” but it will probably take a few weeks. Thanks for your patience.