Sorry, No Hotels Available. The Cities With The Fewest Hotels

When a big event rolls into town, many cities can struggle to cope with the demand for hotel rooms.

Airbnb and other holiday rental sites have alleviated such problems, although in many cases prices can be significantly higher than normal for guests.

Which cities have the fewest number of hotels for visiting tourists?

Methodology

I used Euromonitor’s Top 100 City Destinations Ranking (tourist arrivals) to identify the top 100 cities for tourism in 2015.

Using this list of 100, I then turned to Tripadvisor to identify number of accommodation options in each city. Tripadvisor splits accommodation options into 4 types:

  • Hotels
  • B&B and Inns
  • Speciality Lodging
  • Holiday Rentals

Note, the “holiday rental” category does not include rentals from popular sites like Airbnb and other equivalents which will result in a lower number reported for this category than is correct.

Results

Top 10 Destinations by Tourist Arrivals (2015)

Rank 2015 Arrivals City Country 2014 Arrivals (‘000) 2015 Arrivals (‘000) Y-o-YGrowth 2014-2015
1 Hong Kong Hong Kong, China 27,770.50 26,686.00 3.91%
2 Bangkok Thailand 17,031.70 18,734.90 -10.00%
3 London UK 17,404.00 18,580.00 -6.76%
4 Singapore Singapore 16,795.60 16,869.40 -0.44%
5 Paris France 15,058.10 15,023.00 0.23%
6 Macau Macau 14,566.00 14,308.50 1.77%
7 Dubai United Arab Emirates 13,200.00 14,200.00 -7.58%
8 Istanbul Turkey 11,843.00 12,414.60 -4.83%
9 New York City US 12,230.00 12,300.00 -0.57%
10 Kuala Lumpur Malaysia 11,629.60 12,153.00 -4.50%

Full table.

Hong Kong welcomed 28 million tourists in 2015, 10 million more than the second touristed city, Bangkok (17 million). It also saw the fastest growth (4%) for arrivals in 2015 (top 10 only). Jerba, Tunisia saw the biggest increase in tourists (top 100) (17%), whilst Osaka, Japan saw the biggest decrease in tourist arrivals (top 100) (-52%).

Top 100 Destinations by Tourist Arrivals (2015)

Histogram-of-2015-Tourist-Arrivals-000

Download chart.

To put things into perspective, while all cities in the top 10 received more than 10 million arrivals the majority of cities received less than 5 million (67).

Map-2015-Tourist-Arrivals-000

Interactive map.

The top destinations in the top 100 fall primarily in Europe and Asia.

Top 10 cities by number of accommodation options

Rank Accommodation count City Country count Hotels count B&B count Specialty Lodging count Holiday Rentals total accommodation
1 Orlando US 335 2 17 15,796 16150
2 Rome Italy 1291 3763 1097 8549 14700
3 London UK 1088 574 837 7,654 10153
4 Beijing China 6449 1292 925 44 8710
5 Paris France 1801 129 157 6,265 8352
6 Shanghai China 4999 728 292 62 6081
7 Mugla Turkey 1238 912 954 2,792 5896
8 Mexico City Mexico 1691 1796 1299 964 5750
9 Phuket Thailand 754 997 849 3064 5664
10 Barcelona Spain 521 326 764 3,984 5595

Full list.

Orlando does not have the most hotels (Beijing does, 6449) but it does have the highest number of accommodation options, 16150 in total (most of which are classified as holiday rentals by TripAdvisor).

The top 4 cities with the most hotels are all in China: Beijing (6449), Shanghai (4999), Guangzhou (3394), Shenzhen (2754).

World largest hotels by rooms

Rank Name Country City Rooms/suites
1 First World Hotel Malaysia Genting Highlands 7,351
2 The Venetian and The Palazzo United States Las Vegas 7,117
3 MGM Grand Las Vegas + The Signature United States Las Vegas 6,852
4 CityCenter United States Las Vegas 6,790
5 Sands Cotai Central Macau Macau 6,000
6 Izmailovo Hotel Russia Moscow 5,000
7 Wynn Las Vegas + Encore Las Vegas United States Las Vegas 4,750
8 Mandalay Bay + Delano + Four Seasons United States Las Vegas 4,426
9 Luxor Las Vegas United States Las Vegas 4,407
10 Ambassador City Jomtien Thailand Pattaya 4,219

Full list.

The First World Hotel in Malaysia holds the crown for the largest hotel by number of rooms (7351). 6 of the top 10 largest hotels by number of rooms can be found in Las Vegas.

Fewest accommodation options per tourist

Arrivals / accom rank City Country Accommodation options total Arrivals 2015 total Arrivals per accom
1 Macau Macau 110 14,308,500 130,077
2 Edirne Turkey 58 3,190,400 55,007
3 Pulau Pinang Malaysia 53 2,774,500 52,349
4 Burgas Bulgaria 97 2,842,800 29,307
5 Manama Bahrain 234 6,413,200 27,407
6 Hong Kong Hong Kong, China 1469 26,686,000 18,166
7 Singapore Singapore 963 16,869,400 17,518
8 Doha Qatar 169 2,930,000 17,337
9 Antalya Turkey 749 10,868,700 14,511
10 Mecca Saudi Arabia 516 7,175,200 13,905

Full list.

As an average figure, each accommodation option (hotel, B&B, etc) in Macau must serve 130,077 tourists over a year. Assuming each stays for an average of 7 days and the majority of those bookings are 1 room for 2 people, that means there must at least 1,251 rooms at each accommodation option ((130077/2)/52), assuming maximum efficiency. After further research, I found official figures (2015) that stated there are 29,725 rooms spread across 73 hotels in Macau (note, the above calculation factors in other accommodation options Рnot just hotels).

Compare that to Orlando, ranked 100 for arrivals / accommodation, where there is 1 accommodation option to every 309 tourists. Tourists have significantly more choice here.

Improvements

I used number of accommodation options, not rooms for these calculations. Cities like Macau and Las Vegas tend to have “mega-hotels” of 3000+ rooms each. It would be more accurate to calculate scarcity of hotel rooms against tourists, as opposed to individual hotels (and other accommodation options) as has been used¬†in this post.

As discussed in the Methodology section of this post, the data does not include rentals from popular sites like Airbnb and other equivalents resulting in a lower numbers reported for this category than actually available. If it was possible to get an accurate count of these rentals this would improve my current workings.

tl;dr

Each hotel (and other accommodation type) must serve 130,077 tourists in Macau.

Get the Data

Get all the data used in this blog post on Google Sheets.

Pay Off A Mortgage In Less Than 3 Years As An Airbnb Host

Micro-housing markets around the world change the shape of local communities. Those with skyrocketing costs, London or San Francisco for example, are starting to see an exodus of young people. Whilst Detroit, with comparatively low housing costs, is seeing an influx of younger people.

In San Francisco it will cost you on average $199 USD per night (April 2016) to rent a 1 bedroomed apartment in the city using Airbnb. If you’re spending this much a night on accommodation you quickly start doing some back-of-napkin maths. At 100% occupancy that’s a gross rental income of $6169. Even at 50% occupancy that’s still well over $3000 USD per month.

Now, anyone who lives in San Francisco will tell you long-term rent is expensive too. The average traditional rental price for a 1 bedroomed property is $2850 USD according to Zillow. In this case subletting on Airbnb at 50% occupancy would leave a gross profit of $150 USD (before taxes, utilities, etc are factored in). If you owned the property, the profit (even with a mortgage) would likely be much greater.

But where are hosts making the most money?

Methodology

Using Airdna data I examined the average nightly cost of “Entire place” for studio, one, and two bedroomed rentals on Airbnb (house and apartment) during the month of April 2016 for 30 of the largest metro areas in the United States. My calculations factor in the Airbnb service fee, however ignore utility costs (and other associated fees) to the hosts (thus, returns should be¬†considered as gross).

Occupancy rates for Airbnb rentals are harder to find and vary significantly as a result of a number of factors, namely time and location. However, after researching a number of sources online I came up with average occupancy figures of between 50% Р75%, like this Quora answer for NYC and this SmartAsset blog post, as a base estimate.

I then turned to Zillow to obtain both average April 2016 rental values and average April 2016 property values for the same 30 metro areas.

Analysis

Buy-to-Airbnb (1 bedroom)

Property prices in some cities have increased wildly over the last few years. Take San Francisco, CA, a new one bedroomed home will set you back over $500K USD. In Cleveland, OH you could buy 8 homes with that money with some left over (average 1 bedroomed cost = $63.3K USD).

Highest ROI Cities

1 BR purchase price rank (April 2016) 1BR ROI Rank (April 2016) 1BR Actual Earnings Rank (April 2016) RegionName Buy 1 BR ROI Yr @ 50% occ (April 2016) Buy 1 BR ROI Yr @ 75% occ (April 2016)
30 1 20 Cleveland, OH 22.72% 35.65%
27 2 15 Tampa, FL 21.52% 33.83%
26 3 23 Pittsburgh, PA 19.50% 31.19%
29 4 26 San Antonio, TX 18.74% 29.71%
28 5 28 Orlando, FL 18.04% 28.68%
24 6 21 Cincinnati, OH 17.88% 28.17%
22 7 10 Houston, TX 17.78% 27.87%
18 8 6 Miami-Fort Lauderdale, FL 16.48% 25.75%
21 9 17 Las Vegas, NV 16.40% 26.14%
23 10 27 St. Louis, MO 14.95% 23.79%

Full list.

Cleveland, OH, the cheapest city to buy a 1 bedroomed place on our list ($63.3K USD) gives a 26% yearly return at 50% occupancy. At 75% this increases to 36% — in 3 years the Airbnb revenues would exceed the purchase price of the property!

In comparison, San Francisco, CA, the most expensive place to by a property, offers average returns of around 6% – 9% yearly to homeowners renting property using Airbnb.

Own 1 BR Yearly Airbnb ROI @ 75% Occupancy

Download chart.

All owners hosting on Airbnb at 50% occupancy will make a gross profit. The best ROI for 1 bedroomed places is generally found in the cities with lower property values. The spread of returns at 75% occupancy is between 9% and 36% yearly. At 50% occupancy this drops to 5% and 22%.

Buy-to-Airbnb (2 bedroom)

Highest ROI Cities

2 BR purchase price rank (April 2016) 2BR ROI Rank (April 2016) 2BR Actual Earnings Rank (April 2016) RegionName Buy 2 BR ROI Yr @ 50% occ (April 2016) Buy 2 BR ROI Yr @ 75% occ (April 2016)
30 1 21 San Antonio, TX 30.62% 47.53%
22 2 23 Orlando, FL 29.72% 46.21%
20 3 11 Detroit, MI 29.45% 46.77%
27 4 15 Cincinnati, OH 27.78% 43.01%
16 5 7 Miami-Fort Lauderdale, FL 27.65% 42.51%
21 6 25 Tampa, FL 26.92% 41.92%
28 7 26 Pittsburgh, PA 25.81% 40.64%
24 8 17 Houston, TX 23.81% 36.92%
29 9 29 Cleveland, OH 23.42% 36.70%
17 10 18 Las Vegas, NV 23.25% 36.42%

Full list.

Detroit, a city that¬†has been in decline over the past few decades, is beginning to see somewhat of a resurgence. Whilst the city is not the cheapest to purchase a 2 bedroomed property (ranked 20/30), it ranks high for actual Airbnb dollar earning (11/30) and offers the third highest yearly ROI¬†at 30% and 46% at 50% and 75% occupancy respectively. I don’t have the data to back up wether these¬†occupancy statistics are¬†easily realised, but 50% occupancy seems relatively low to¬†me (for an average property in a big¬†city, competition not considered).

Own 2 BR Yearly Airbnb ROI @ 75% Occupancy

Download chart.

And such high returns are not unique. The median return for 2 bedroomed places at 75% occupancy is over 30% —¬†that equates to 15 properties out of 30 returning more than 30% of their value each year.

Buy-to-Airbnb (summary)

Median returns by occupancy and home size buyers (April 2016)

Download chart.

You can see the gross return on 1 bedroomed properties in most cities is roughly 10% less than 2 bedroomed properties.

Rent-to-Airbnb (1BR)

Highest ROI Cities

1BR rent rank 1BR actual earning rank 1BR ROI rank Region Name Rent 1 BR ROI Yr @ 50% occ USD Rent 1 BR ROI Yr @ 75% occ USD
30 21 1 Cincinnati, OH 99.21% 198.81%
27 21 2 Cleveland, OH 72.51% 158.77%
28 27 3 St. Louis, MO 66.93% 150.39%
24 13 4 Las Vegas, NV 66.80% 150.20%
25 20 5 Pittsburgh, PA 59.32% 138.98%
29 30 6 Detroit, MI 51.59% 127.39%
22 16 7 Tampa, FL 51.30% 126.96%
26 26 8 San Antonio, TX 49.30% 123.95%
3 1 9 Boston, MA 45.67% 118.51%
23 25 10 Phoenix, AZ 40.91% 111.36%

Full list.

Your first observation is that ROI will be much higher for renters (but remember we are only considering rent for a year which is significantly less than the cost of buying a property).

Returns of almost 200% can are obtainable if renters are able to sub-let their 1 bedroomed home for 75% of the year. The top ten cities above all have ROI greater than 100%.

1BR rent rank 1BR actual earning rank 1BR ROI rank Region Name Rent 1 BR ROI Yr @ 50% occ USD Rent 1 BR ROI Yr @ 75% occ USD
8 9 27 Seattle, WA -3.54% 44.69%
5 7 28 Washington, DC -5.04% 42.43%
4 4 29 Los Angeles-Long Beach-Anaheim, CA -6.42% 40.36%
1 2 30 San Francisco, CA -10.19% 34.71%

Full list.

Unlike buying, I have calculated that it is likely you could lose money subletting your home (based on rent paid for the year). In 4 cities, subletting with less than 50% occupancy yields¬†returns less than total rent paid. That means if you sublet on Airbnb in San Francisco and your occupancy rate is 50% for the month, you will have to cover 10.19% of the rent yourself — essentially you pay 10% of total rent to stay in the home 50% of the time. Not a bad deal.

Rent 1 BR Airbnb ROI @ 75% Occupancy

Download chart.

At 75% occupancy renters in all cities subletting their home will receive positive returns with a huge spread returns of between 35% and 200%.

Rent-to-Airbnb (2BR)

Highest ROI Cities

2BR rent rank 2BR actual earning rank 2BR ROI rank RegionName Rent 2 BR ROI Yr @ 50% occ USD Rent 2 BR ROI Yr @ 75% occ USD
28 15 1 Cincinnati, OH 122.09% 233.14%
25 9 2 Detroit, MI 114.77% 222.16%
26 17 3 Las Vegas, NV 99.41% 199.12%
29 27 4 St. Louis, MO 86.08% 179.12%
4 1 5 Boston, MA 85.27% 177.91%
24 21 6 San Antonio, TX 82.46% 173.70%
15 8 7 Baltimore, MD 73.43% 160.14%
19 14 8 Charlotte, NC 66.95% 150.42%
22 21 9 Phoenix, AZ 66.85% 150.27%
23 25 10 Pittsburgh, PA 61.67% 142.50%

Full list.

As was the case for property owners, sub-letters of 2 bedroomed homes¬†get better returns when compared with 1 bedroomed homes — in Detroit returns are 55% higher. All cities yield positive returns at both 50% and 75% occupancy for 2 bedroomed homes.

Again Cincinnati offer sub-letters the best returns for 2 bedroomed places, as was true for 1 bedroomed homes.

Rent 2 BR Airbnb ROI @ 75% Occupancy

Download chart.

Most returns are between the 60% – 180% range at 75% occupancy.

Rent-to-Airbnb (summary)

Median returns by occupancy and home size renters (April 2016)

Download chart.

Sub-letters see a larger difference in returns on property sizes. 2 bedroomed homes¬†offer much better returns than 1 bedroomed homes — at 75% occupancy returns for 2 bedroomed places are almost 40% higher.

Improvements

I didn’t consider utilities — a major overhead for hosts. I have obtained numbers on Numbeo which puts average rents in these cities at about $2360 yearly, a range of between $1938 USD¬†(San Diego) – $3936 USD (Detroit).

That said, my calculation did not consider cleaning fees, which can amount to $100’s per month, and thus potentially offsetting utility costs incurred.

For a more accurate comparison, both of these factors should be considered.

tl;dr

Airbnb hosts renting 2 bedroomed homes make more money than 1 bedroomed homes. In many cases hosts subletting can pay their yearly rent twice over from Airbnb yields. Likewise many homeowners in the cities analysed could make the purchase price of their home in only a few years through Airbnb guests.

Footnotes

  1. Data sources + data used in this post.

$100 Per Week To Have Your Airbnb Cleaned in Miami

Airbnb rentals can be a great alternative to hotels, not least because they’re often cheaper.

As the company expands it’s reach into the global tourist market an ever increasing number of cities have hosts making their properties¬†available to rent through the company.

For travellers on a budget the choice of destination is often heavily weighted on cost of accommodation, one of the largest expenditure of any holiday.

So where can you find the cheapest Airbnb rentals?

Methodology

To draw up a list of cities for comparison I turned to GoEuro’s “Accommodation Price Index” published in 2016 (using data from 2014). This list of 150 cities consists of the top tourist destinations chosen based on several factors including hotel availability and size of hospitality industry.

I then turned to Airdna for data on Airbnb rentals.¬†I examined the average nightly cost of “Entire place” studio, one, and two bedroomed rentals on Airbnb (house and apartment) during the month of Feb¬†2016 (May 2016 for cleaning rates) for the cities collected from GoEuro. Airdna had data for 87 of the 150 cities listed on the Accommodation Price Index.

Many listings on Airbnb are not actively rented. Airdna has defined its own definition of what constitutes an “active” property. I am only considering “active” properties in this post, although admittedly I’m not 100% sure of the definition.

airbnb active rentals feb16

Full map.

Rank (most active rentals) City Active rentals total Feb16
1 London, UK 29718
2 Paris, France 29344
3 New York (Manhattan), USA 16648 (26795 NYC)
4 Rio de Janeiro, Brazil 16057
5 Berlin, Germany 13931
6 Barcelona, Spain 13550
7 Rome, Italy 12383
8 Sydney, Australia 11800

Full ranking.

It is surprising more US cities are not on this list of most¬†active rentals, especially given this was the Airbnb’s first market. It is also surprising to see European cities, London and Paris, in first and second place respectively.

Analysis

Studio price index

Entire Studio House and Apartment Nightly Rental Ave USD Feb16

Download chart.

Studios can be rented as cheaply as $26 USD before fees (St. Petersburg, Russia) although most rentals (58/87) have a nightly cost of between $35 – $75 USD.

Airdna studio rentals Feb16 Map

Full map.

Rank (most expensive) City Entire Studio (House & Apartment) Nightly Rental Ave USD Feb16
1 Manchester, UK 169.50
2 Singapore, Singapore 154.50
3 San Francisco, USA 153.50
85 Mumbai, India 31.50
86 Sarajevo, Bosnia and Herzegovina 29.00
87 St. Petersburg, Russia 26.00

Full ranking.

San Francisco (2nd) and Singapore (3rd) are expensive places to live and as such would be expected to be cities with high rental values.

Manchester in 1st place ($169.50 USD p/night) almost $15 USD more than second placed Singapore ($154.50 USD) is¬†unexpected — not least because rents are on average 63% cheaper in Manchester than London (ranked 9th @ $109.50 USD p/night). My assumption for this would be that there is a greater supply of studios geared to transient populations in London, San Francisco, and Singapore. Whereas in Manchester¬†studios¬†are likely to be of¬†higher specification, more centrally located, and mainly designed for traditional renters. What do you think?

1 bedroom price index

Entire 1BR House and Apartment Nightly Rental Ave USD Feb16

Download chart.

There is a wider spread of prices for one bedroomed apartments compared to studios, perhaps because of more supply (greater choice), although this is just an assumption as Airdna only provides a single aggregated “active” rentals number for all property types.

airdna 1br rentals feb16 map

Full map.

Rank (most expensive) City Entire 1BR (House & Apartment) Nightly Rental Ave USD Feb16
1 San Francisco, USA 172.50
2 Boston, USA 167.50
3 New York (Manhattan), USA 164.50
85 Sochi, Russia 35.00
86 Budapest, Hungary 34.50
87 St. Petersburg, Russia 26.50

Full ranking.

In most cases, one bedroomed properties are around 14%¬†more expensive than studios. What constitutes a studio and a 1 bedroomed is somewhat of a grey area though. I’ve stayed in Airbnb rentals listed as 1 bedroomed that official measures would deem a studio. Frustrating, ever-so-slightly. However, could be one reasons that accounts for this ever-so-slight increase in cost.

2 bedroom price index

Entire 2BR House and Apartment Nightly Rental Ave USD Feb16

Download chart.

Whilst most 2 bedroomed places are below $200 USD p/night before fees, 5 cities have an average above this figure. 2 cities cost more than $300 USD a night, Dubai, UAE ($326.50 USD) and Frankfurt, ($311.50 USD).

airdna 2br rentals feb16 map

Full map.

Rank (most expensive) City Entire 2BR (House & Apartment) Nightly Rental Ave USD Feb16
1 Dubai, UAE 326.50
2 Frankfurt, Germany 311.50
3 Boston, USA 281.50
85 St. Petersburg, Russia 44.00
86 Sochi, Russia 40.50
87 Sarajevo, Bosnia and Herzegovina 38.00

Full ranking.

At the other end of the scale, a 2 bedroomed places in Eastern Europe can be picked up very cheaply. The 10 cheapest cities for 2 bedroomed cities are all in Eastern Europe (all below $60 USD).

Cleaning fee index

Once you’ve got your heart set on a place, don’t forget to budget for the additional fees charged. These include a cleaning fee set by the host, and it’s not always as cheap as¬†you might expect.

Cleaning fees help hosts account for extra expenses they have getting their listing ready before guests arrive or after guests depart.
Airbnb

Ave Cleaning Fee USD May16

Download chart.

You’re probably wondering what city has average cleaning fees that amass on average $100 USD per rental. That city would be Miami, USA.

airdna cleaning fees may16 map

Full map.

Rank (most expensive) City Ave Cleaning Fee USD May16
1 Miami, USA 100.00
2 Boston, USA 95.00
3 Ibiza, Spain 91.00
84 Bucharest, Romania 12.00
84 Sarajevo, Bosnia and Herzegovina 12.00
87 Kiev, Ukraine 10.00

Full ranking.

You might notice that 2 of the top 3 cities with the highest cleaning fees could both be considered “party towns”, Miami and Ibiza. Parties = mess = more cleaning?

Improvements

Whilst this post identifies the most expensive cities in which to rent an Airbnb apartment, it does not look at potential profits hosts can achieve. Or put another way, the differences between cost of ownership or longterm rental against Airbnb income.

tl;dr

The cheapest “Entire¬†place” Airbnb rentals can be found in Eastern Europe. For a studio or 1 bedroomed place go to St. Petersburg, Russia (ave. $26 USD p/night exc. fees), and¬†Sarajevo, Bosnia and Herzegovina for 2 bedroomed places ($38 USD).

Footnotes

  1. Data sources + data used in this post.

Your Airbnb Rental Is More Spacious, Homely, and CHEAPER Than A Hotel

Airbnb is increasingly becoming¬†an ingrained part of searching for a place to stay when away from home — just look at the companies¬†very rapid growth rate¬†as an indication.

One common belief held by travellers is that an Airbnb rental is usually¬†cheaper than staying at a hotel. A quick glance at the prices shown in¬†GoEuro’s Accommodation¬†Price Index, suggests this statement would appear to hold some weight.

Are Airbnb rentals really cheaper?

Methodology

To draw up a list of cities I turned to GoEuro’s “Accommodation Price Index” published in 2016 (using data from 2014). This list of 150 cities consists of the top tourist destinations chosen based on several factors including hotel availability and size of hospitality industry.

I then turned to Airdna for data on Airbnb rentals.¬†I examined the average nightly cost of “Entire place” studio and one bedroomed rentals on Airbnb (house and apartment) during the month of Feb¬†2016 for the cities collected from GoEuro.¬†Airdna had data for 87 of the 150 cities.¬†I chose 1 bedroomed and studio rentals as I deemed them the closest equivalent to hotel rooms.

Results

Airbnb vs hotels

USD Difference between hotel and 1 bed, entire home Airbnb rental

Download chart.

USD Difference between hotel and Airbnb map

Download map.

I found Airbnb rentals are on average $7.29 USD cheaper then hotels.

In 26 of the 88 cities I analysed hotels were cheaper, compared to 60 cities where Airbnb’s were cheaper. 2 cities had no price difference, Riga and Warsaw.

USD Difference between hotel and entire home Airbnb rental

Download chart.

40 of the 60 cities that were cheaper for Airbnb rentals, were only slightly cheaper ($0 USD – $15) than hotels.

Top 10 cities where hotels are cheaper (by actual price difference)

Top 10 cities where hotels are cheaper than Airbnbs by price difference

Rank City Hotel-Aibnb diff USD EM Hotel Ave 2014 USD Airdna Median Entire Home Studio 1br Feb16 USD
79 Frankfurt, Germany (13.75) 62.00 75.75
79 Shanghai, China (13.75) 41.00 54.75
81 Boston, USA (14.33) 147.00 161.33
82 Mexico City, Mexico (14.67) 42.00 56.67
83 Reykjavik, Iceland (15.75) 85.00 100.75
84 Chicago, USA (16.75) 97.00 113.75
85 San Francisco, USA (22.00) 141.00 163.00
86 Ibiza, Spain (27.75) 71.00 98.75
87 Manchester, UK (31.25) 100.00 131.25
88 Singapore, Singapore (38.00) 94.00 132.00

Full table.

In the cities where hotels are cheaper, the best savings by choosing a hotel can be found in Singapore ($38 USD cheaper then Airbnb rentals).

Top 10 cities where Airbnb rentals are cheaper (by actual price difference)

Top 10 cities where Airbnbs are cheaper than hotels by price difference

Download chart.

Rank City Hotel-Aibnb diff USD EM Hotel Ave 2014 USD Airdna Median Entire Home Studio 1br Feb16 USD
1 New York, USA 95.50 246.00 150.50
2 Canc√Ļn, Mexico 70.00 134.00 64.00
3 Bruges, Belgium 63.00 131.00 68.00
4 Miami, USA 62.50 167.00 104.50
5 Rio de Janeiro, Brazil 45.00 99.00 54.00
6 Cape Town, South Africa 40.00 99.00 59.00
7 Seoul, South Korea 33.50 85.00 51.50
8 Paris, France 33.25 117.00 83.75
9 Montreal, Canada 30.00 82.00 52.00
10 Venice, Italy 28.50 146.00 117.50

Full table.

In New York you can save on average $95 by choosing an Airbnb over a hotel. The city is a popular with Airbnb hosts with 16648 Active Rentals (3rd highest worldwide). Perhaps an oversupply of rentals is pushing down prices for guests allowing for such a large saving?

Did you know? The cities with the most Active Rentals in February 2016 were: 3. New York (16648), 2. Paris (29344), 1. London (29718).

Price¬†difference between hotels and Airbnb’s by region (by actual price difference)

USD Difference between hotel and Airbnb by region

Download chart.

On a regional level, Airbnb rentals are generally cheaper everywhere. It is interesting to note that savings in Europe, which I would assume to be the companies second largest market, are relatively low ($1.95 USD).

Top 10 cities where hotels are cheaper (by % price difference)

rank (actual price diff) rank (% price diff) City EM Hotel Ave 2014 USD / Airdna Media Entire Home Studio 1br Feb16 USD
84 79 Chicago, USA 85.27%
83 80 Reykjavik, Iceland 84.37%
76 81 Dusseldorf, Germany 83.90%
79 82 Frankfurt, Germany 81.85%
78 83 Rotterdam, Netherlands 81.70%
87 84 Manchester, UK 76.19%
79 85 Shanghai, China 74.89%
82 86 Mexico City, Mexico 74.12%
86 87 Ibiza, Spain 71.90%
88 88 Singapore, Singapore 71.21%

Full table.

When looking at percentage price differences, Singapore hotels are over 28% cheaper than Airbnbs. Singapore also has the largest actual price difference between hotels and Airbnb rentals ($38 USD).

Top 10 cities where Airbnb rentals are cheaper (by % price difference)

rank (actual price diff) rank (% price diff) City EM Hotel Ave 2014 USD / Airdna Media Entire Home Studio 1br Feb16 USD
2 1 Canc√Ļn, Mexico 209.38%
3 2 Bruges, Belgium 192.65%
5 3 Rio de Janeiro, Brazil 183.33%
16 4 St. Petersburg, Russia 170.89%
6 5 Cape Town, South Africa 167.80%
7 6 Seoul, South Korea 165.05%
1 7 New York, USA 163.46%
4 8 Miami, USA 159.81%
9 9 Montreal, Canada 157.69%
14 10 Istanbul, Turkey 151.96%

Full table.

Whilst New York had the largest actual price difference where Airbnbs were $95 USD cheaper (hotels were 63% more expensive than Airbnbs), Cancun has the largest percentage price difference between hotel and Airbnbs (hotels are over 109% more expensive!).

A developing market

A number of savvy Airbnb hosts increase their pricing to match high demand, much like most hotels do. Requiring significant effort by hosts, many have previously missed out on opportunities for increased revenue as a result. Similarly during quieter periods static prices for rentals often leave potential guests looking elsewhere because they are set unusually high for the time of year.

Airbnb’s pricing tool that debuted publicly in November 2015¬†is likely to remove much of this burden from hosts. It is too early to tell if more dynamic pricing will¬†increase or decrease the cost¬†difference between hotels and Airbnb rentals.

Improvements

Having to use aggregated city data for both hotels and Airbnb rentals is not the best way to compare prices. If I was able to compare specific hotels and Airbnb rentals that were more closely matched (location, quality, etc) would result in higher accuracy of analysis.

tl;dr

Airbnb rentals are typically cheaper then hotels.

Footnotes

  1. Data sources + data used in this post.

Check If A Hotels WiFi Sucks Before It’s Too Late

Heated pool? Cable TV? Ghosts in the reception hall?

There are lots of factors you can use to rank hotels thanks to online reviews, but for both business and leisure travellers alike WiFi download speed is often one of the most important.

The problem is that many hotels don’t really seem to care about connection quality. Slow WiFi is arguably worse than no WiFi.¬†As long as a hotel can display¬†that little ‚ÄúWiFi Available‚ÄĚ box on the amenities list, they‚Äôre happy. And travellers think they will be too… until they arrive and realise that anything more than checking email is going to be a problem.

Thankfully HotelWiFiTest is continually crowdsourcing data on hotel WiFi speeds so you can be confident you’ll be able to stream the latest TV series from your room.

So which hotels should you avoid?

Methodology

Using a Python script I scraped data from HotelWiFiTest. Data was collected for the following cities: London (384 hotels), New York (321), Bangkok (242), Paris (240), Berlin, Barcelona (196), Singapore (136), and San Francisco (134). In total WiFi data from 1890 hotels was collected.

HotelWiFiTest provides an expected speed in Mbps for each hotel. This indicates the speed you would likely get at a randomly chosen time of the day.

Hotel WiFi Speed Confidence Percentage Distribution May 2016

Download chart.

The site allocates a confidence score to each result, that is a value showing how thoroughly the WiFi has been tested at the hotel. The confidence value depends on several factors, including the number of speed tests taken, how recently the tests occurred, and the diversity of tests in terms of the time of day, day of the week, and point within the travel season.

Out of the 1890 hotels, the mean confidence score was slightly less than 25%. Only 828 hotels had a confidence score above 25%. More than¬†450 hotels had a confidence score less than 7.5%. For the purpose of this analysis we only considered hotels with a confidence score greater than 30% — 644 hotels in total.

Hotel Wifi Speed Expected dl Mbps Distribution May 2016

Download chart.

Of the remaining 644 hotels, 412 had download speeds slower than 10 Mbps.

In addition to HotelWifiTest data I used Akamai’s, “Q4 2015 State of the Internet” report, that contained average¬†internet download speeds by country for comparison.

Hotel room rates used are the cheapest available based on average rates for a standard room on Tripadvisor.

Analysis

Cities with best hotel Wifi Speeds

Rank (Mean d/l speed) City Mean Hotel WiFi Download Speed Mbps Median Hotel WiFi Download Speed Mbps Country Mean Download Speed Mbps Q4 2015 Hotels Tested
1 London 20.88 11.30 13.90 384
2 Barcelona 17.23 11.85 12.10 196
3 Berlin 12.02 6.95 12.90 237
4 New York 11.67 5.60 14.20 321
5 Paris 10.56 4.40 8.90 240
6 Singapore 9.06 6.70 13.90 136
7 Bangkok 8.91 6.50 9.30 242
8 San Francisco 6.88 4.30 14.20 134

See calculations.

London comes out as the best city for hotel download speeds (by mean d/l speed) of those considered with a mean figure of 20.88 Mbps (median of 11.30 Mbps) for the 384 hotels considered.

San Francisco, a city bubbling with some of the latest and greatest internet technologies fares worst with a mean download speed of only 6.88 Mbps. To put that into comparison, the average download speed within the United States is 14.20 Mbps.

Did you know? The country with the fastest average download speed can be found in South Korea — 26.7 Mbps (Akamai Q4 2015).

Best hotels for Wifi download speeds

Rank (Hotel WiFi Speed Expected d/l Mbps) Hotel Name Hotel City Hotel WiFi Speed Expected d/l Mbps Hotel WiFi Speed Confidence Percentage
1 The Mark New York 235.00 59.70%
2 every hotel Piccadilly London 201.00 57.30%
3 Amba Hotel Charing Cross London 178.00 58.50%
4 Thistle Marble Arch London 145.00 53.00%
5 Sofitel London St James London 119.00 58.80%
6 Hotel Costes Paris 112.00 53.90%
7 Ham Yard Hotel London 88.70 60.00%
8 Hotel Crowne Plaza Berlin City Center Berlin 80.50 51.00%
9 Barcelona City Centre Hostal Barcelona 71.30 54.40%
10 San Francisco Marriott Marquis San Francisco 67.10 57.90%

Full table.

Confidence is relatively high for the top 10 hotels, all above 50%. 6 hotels have reported speeds of above 100 Mbps, 4 of which are in London, UK.

The Mark, New York and every hotel Piccadilly, London have estimated download speeds of over 200 Mbps — plenty of bandwidth to stream video (even at 4K resolutions).

Download speeds drop off quickly though¬†—¬†for hotels ranked below 25th¬†speeds are below 50 Mbps.

Worst hotels for download speeds

Rank (Hotel WiFi Speed Expected d/l Mbps) Hotel Name Hotel City Hotel WiFi Speed Expected d/l Mbps Hotel WiFi Speed Confidence Percentage
635 Hotel Observatoire Luxembourg Paris 0.37 37.30%
636 Hotel Xenia London 0.35 31.90%
637 Hotel Paris Louis Blanc Paris 0.32 30.50%
638 Holiday Inn London Kensington Forum London 0.31 41.00%
638 Journeys London King’s Cross Hostel London 0.31 35.50%
640 Park Plaza Wallstreet Berlin Mitte Berlin 0.30 41.60%
641 Novotel London City South London 0.29 33.00%
642 Victor’s Residenz Hotel Berlin Berlin 0.21 30.40%
643 Hotel Air in Berlin Berlin 0.15 33.50%
644 InterCityHotel Berlin Hauptbahnhof Berlin 0.08 36.90%

Full table.

The slowest hotel tested, the¬†InterCityHotel Berlin Hauptbahnhof (rank 644) in Berlin (ranked 3rd), had a download speed of only 0.08 Mbps — only slightly faster than dial up (0.056 Mbps). At this speed checking (a few) emails is about the best you’ll manage.

41 hotels have an expected download speeds slower than 1Mbps that would make even basic use of many sites frustrating (although YouTube recommends a minimum download speed of 0.5 Mbps).

Speed benchmark

10 Fastest Hotels May 2016 vs Country Download Speeds Q4 2015

Download graph.

To put this in comparison, you can see how much faster the hotel connection speeds for the top 10 are compared to the average connection download speed for each hotels respective country. The hotel with the fastest expected speed, The Mark (New York), has a download¬†speed that’s over 220 Mbps faster than the average download¬†speed in the US according to Akamai (Q4 2015)!

10 Slowest Hotels May 2016 vs Country Download Speeds Q4 2015

Download graph.

In contrast, the hotels with the slowest download speeds are in some cases well-over 100 times slower than the hotels with the fastest download speeds (and all significantly slower than their countries average download speeds).

Cost per Mbit

Top 10 Hotels With Fastest Download Speed Cost pMbit May 2016

Download graph.

Top 10 Hotels With Slowest Download Speed Cost pMbit May 2016

Download graph.

If you’re paying hundreds of GBP per night for a room you’d likely be expecting good WiFi speed as part of the service. You might even assume a linear relationship between money spent and download speed.

The average nightly room rate for the top 10 hotels with fastest download speeds is £212 GBP (range £62 Р£331 GBP), and £71.50 (range £16 Р£122 GBP) for the bottom 10 hotels. Although a small sample (only 20 hotels), this does indicate a relationship between hotel room rate and WiFi download speed.

At the hotels with the fastest download speeds the cost per Mbit (expected WiFi speed / average cheapest room rate) ranges between £0.83 Р£4.24 GBP. At the slowest hotels the lowest cost per Mbit is, at best, £16 GBP and, at worst, £122 GBP!

Neighbourhoods

Looking at this on a slightly more macro level, I wanted to know if the neighbourhood of a hotel had any bearing on the speed. For instance; would hotels found in business districts typically offer better download speeds?

London

Hotels with Wifi confidence greater than 30pc heatmap London

Download full map.

New York

Hotels with Wifi confidence greater than 30pc heatmap New York

Download full map

In the city with the largest number of hotels analysed, London (384 hotels), there is a spread of speeds across the capital. Although some of the fastest speeds are to found in the more traditional tourist areas of the West End, Victoria, and Marble Arch.

Similarly in¬†New York (321) the fastest download speeds can be found in areas with hotels geared mainly towards tourists — around Broadway. However, in New York the hotel spread is more dense when compared to London. This is because the layout of Manhattan being more compact than the urban sprawl of London.

Improvements

HotelWifiTest download speed data points are:¬†“the speed you would likely get at a randomly chosen time of the day”.¬†Clearly more detailed data on WiFi speeds would increase the accuracy of any analysis.

It would also be interesting to compare average room rates for all hotels and see how strong the correlation between room rate and hotel WiFi download speed really is.

tl;dr

  • If you want a blazing fast connection go to The Mark, New York, where the expected WiFi download speed is 235 Mbps. You’ll pay for it though. Prices from ¬£332 per/night (TripAdvisor – checked May 2016).
  • London hotels offer the best average WiFi download speed at¬†20.88 Mbps.
  • Generally you’re probably best staying in hotels located within a cities tourism district for the best WiFi download speeds.

Acknowledgements

  • HotelWiFiTest for putting together a great public resource for checking hotel WiFi speeds (you should contribute when you can)
  • Akamai for their brilliant “State of the Internet” quarterly reports (I used Q4 2015 for this post).
  • …and finally¬†TripAdvisor for hotel room rate figures.

Get the Data