AirBnb Data Analytics - Lake Livingston TX

13 Replies

Hi all, considering buying a property in Lake Livingston, TX as a short term/Airbnb rental.  I have seen AirDNA and a few other sites that have historical analytics data.  Are these sites any good?  Is there one that is better than the others? Since Lake Livingston is a relatively small market area I am not sure these sites will be much use...?

Thank you!

I have tried Airdna and AllTheRooms and I like AllTheRooms better. It's like a fourth the price for what was basically the same thing for analysis. 

Huge bonus: ATR let me export CSVs, which let me plug it into Google Data Studio much more easily. 

With Airdna I had to manually copy each month's ADR and Occ Rate for the past 3 years. Because I'm looking at everything, when you're doing all bedroom options individually and separating by 50th and 75th percentiles, that gets really annoying really fast. So yeah. CSV exports are where it's at.

Hey @Jessica Lindner , your best bet is to search AirBNB and VRBO for listings in that area that are similar to what you want to buy.

Take a look at their nightly rates and their calendars. That will give you a good baseline to work from. See what their amenities are. What do the pics look like. Etc Etc.

@Michael Baum   Thank you, I did that for June/July/Aug which gives me a decent idea however I don't think there's a way to view the historical calendars for the listings - they are just grayed out (unless you know of a way?).  I am more interested in occupancy rates during the off season for a lake house which is why I was considering AirDNA, etc.

Originally posted by @Michael Baum :

Hey @Jessica Lindner , your best bet is to search AirBNB and VRBO for listings in that area that are similar to what you want to buy.

Take a look at their nightly rates and their calendars. That will give you a good baseline to work from. See what their amenities are. What do the pics look like. Etc Etc.



This is a good point. If you're seeing that everyone has Wifi, be sure you've got that. If no one has this or that, ask why that is and you can maybe get away with not having it (or get it so you stand out!). 

But also be aware that looking at Airbnb itself only shows what people are asking for per night, not what they're actually booking for. People can ask for whatever they want, of course. Another bonus that I think Michael alludes to is with the data sites like AllTheRooms, they show you the best performing properties in that area based on the data. So you know for sure what the best in the area looks like and what amenities they have and what their listing looks like. If you search for a certain weekend to gather info, you might not see the best properties in the area because those are already booked. 

 

I don't get why people would pay to use sites like AirDNA when all one has to do is go on AIR or VRBO and do your own searches for comparable listings. You can find out so much more it seems......but maybe I'm missing something.... :-)

Originally posted by @Bruce Woodruff :

I don't get why people would pay to use sites like AirDNA when all one has to do is go on AIR or VRBO and do your own searches for comparable listings. You can find out so much more it seems......but maybe I'm missing something.... :-)

We can't assume we understand the used Ford F-150 market because we saw one guy trying to sell his on the side of one road for $25,000. Just because he's asking that, doesn't mean he's getting it. The software and analysis is the same way. Just because you search Airbnb once and see someone trying to rent theirs for a certain price, that's not reflective of the market. Just that "street" at that time for that particular type of truck. 

But what these data sites do is tell you, "Okay, in October 2019, all the 2015 Ford F-150 extended cabs sold for an average $19,232." So when you try to sell your own, you can collect all the months, see the trend, see how many are selling these days, to know what is fair. To expand the analogy, these softwares even show you what the median is, as well as the top 25th percentile and sometimes the top 10th of listings. THOSE are the ones you want to explore. Not the ones that happen to be available for the couple of timeframes you chose. Those really good ones are already booked and aren't going to show up. 

And if you're looking to invest, you can begin to see how buying a bigger or smaller place is reflected when you see all the months and trends and for however many bedrooms you're looking to rent. Not many 1 BR in an area? Good to know. Most 3 BR homes feature xyz? Great. This area is actually decreasing in the number of total Airbnb listings overall for the last three years? Better see what's going on there before I invest. 

I can charge $10 a night and book every night, but that doesn't mean I'm making the most money. When we can see both how much everyone actually books for AND how many nights per month they book, we can see a much better picture of what's possible in the market without leaving a lot of money on the table. 

Hope this helps. :D

@Jacob Murphy , the biggest issue with AirDNA is the algorithm. There is nothing wrong with it really, but in my experience, it overestimates occupancy and nightly rates.

In our case, it was just about 3x the actual nightly rate for our area (Lake Coeur d'Alene). Our place was lumped in with small lake cottages up to multi million dollar log mansions.

I will say that if the area has a ton of comparable units, the more accurate the data will be.

This has been proven across Bigger Pockets as many have found this to be the case. If you rely strictly on these kinds of data miners for your decision making, you will leave yourself under performing those stats in most cases.

@Jessica Lindner , I wish I knew a way to scrub data from AirBNB's past months but I don't. I just went through all the rentals in my area that were remotely similar to our home on both AirBNB and VRBO. That got us right in the target nightly rate. We are out performing the occupancy for our are at the moment.

The bottom line is that you will never get 100% accurate data. All you can do is get a good idea. Much of your performance will come down to the area, your level of customer service, design, etc etc. You can have a cheap place, furnished cheaply with no style and never book a night even if it is the lease expensive place. Or you will get the dregs staying there who will trash your place into the dirt.

In the case of the F-150, if you base your selling price off of one single unit you saw with no other data, then you will have bad info to base your selling price off of. Just like relying on AirDNA.

Originally posted by @Jacob Murphy :
Originally posted by @Bruce Woodruff:

I don't get why people would pay to use sites like AirDNA when all one has to do is go on AIR or VRBO and do your own searches for comparable listings. You can find out so much more it seems......but maybe I'm missing something.... :-)

We can't assume we understand the used Ford F-150 market because we saw one guy trying to sell his on the side of one road for $25,000. 

No one would do that. When you go to AIR or VRBO and search within , let's say, 5 miles of your house, with similar amenities and size, you can easily find usable comps. You can also search different dates to see if they are actually being rented, in other words are they getting their asking price consistently?

So.... if you find that the top 10 (or 20, or 30) places that are very similar to your place are being rented for, let's say, $125 per night in June....Bingo. At least that's enough research for me, some people may want to research this to death, that's their choice. I still can't see how it's necessary.

As a matter of fact, your own research may be the only that is valid. The metrics that a certain company uses are most likely based on the way their brain works...they may not give you what you really need at all. It would be like getting all of your news from MSNBC (or ?) and assuming that you now know what's going on in the world. You actually would have no clue what is going on......

I still have no idea why one would use an Air DNA type service, it's waaay too easy to get your own results and probably more accurate. But thanks for attempting to explain :-)

 

I love this discussion. Fantastic stuff.

@Michael Baum Yeah for sure more data is always going to produce better results. An area near me I looked into had about 250 units to draw from, which made the percentiles look odd. It looked like your area is similar, so I know what you mean. 

Well tell me what you think about this, because this problem fascinates me. I'm hearing you say - and correct me where I'm wrong - that even when you a) filter for how many bedrooms and b) how many guests you accommodate, what's left in your data are the couple of properties in your area with those details, but the spread of properties in your area is so different that it makes the data shown not very useful? 

Is this because there are too few properties with those stats? Do you have any other property areas you've explored that show this, and how many are listed in that one? The other thing I wonder is if the percentiles are any help. We can assume at least that if you get all the data and you can see what the median earns/books, and you know about what features are typical for your area, you can assume that by doing a little more than that you could be in the 75th percentile. I'd think the 90th percentile is going to be for those ultra-mega ritzy, great-view places inside your particular category of matching # BR. 

Last thing might be that you're suggesting under performing the stats means the data is wrong and can be given less weight, but (and this is more of a meta question) how can we as owners/hosts rule out that the data is fine and we're not the ones under performing?

@Bruce Woodruff Same as I said above. Great discussion. This is great. :D

You say no one would do that, but I think you'd be surprised! People with a LTR mindset come in from doing market research on a couple because that market is pretty dialed in. That market has been refined and matured to a point where someone asking for something per month is coming from a place of more time and information.

STR, on the other hand, is full of people with a spare place who are guessing about what to price stuff at. I don't think the market is that mature yet to where the majority are that confident. BiggerPockets users and more data-driven owners aside, I'd argue most people seem to ballpark it, or go off a handful of other users who are also ballparking it.

I wouldn't disagree that it can be overwhelming to the point of throwing your hands up like eh how much effort is this really saving me. But my point is just that searching for a certain week or weeks and looking at 20, 30, or even 50 places is going to take a bunch of time on its own. The data shows everything in a downloaded CSV done for you or in easy to view charts the second you purchase.

And when you take from the site in that way, again, you're taking ones that aren't booked and only from that timeframe. It's no use at all to calculate based on asking price, because we want to know what people are actually earning. It's just like comping a LTR investment off apartments in town with a sky-high vacancy rate. The data lets us know which properties are actually performing and how well. And when lumped all together, it shows us where most perform and where the best of the best perform.

There's also seasonality to consider. For a lot of places, you can split the performance into four seasons a year, because it fluctuates so much. How much hurt would you be in if you calculated for an investment off high season and assumed that performance all year? That's bad news.

So what I'm hearing you say is that you'd be okay collecting info from 50 different places. Are you accounting for seasonality? If you're doing that every month (50x12) just to get a better feel for the year round aspect, that's a start, but that's a lot of research. And you couldn't do that ahead of time accurately because everyone opens their calendar different days into the future. So you're collecting 20-50 places, every month at a month in advance, for a whole year to get a proper investment analysis?

So all I'm saying is, there's gotta be a point where it's easier to pay the $30-90, see the graphs, grab the monthly numbers (from all the properties across the last three years, not just the few you choose) it gives you in a spreadsheet in 10 minutes and move on. 

Hey @Jacob Murphy . Ok so I spoke with the owner of AirDNA about the algorithm. He approached me after I checked it out with a free demo and left some feedback. Like I had said, there is nothing fundamentally wrong with the process. Here is my example.

On Lake CdA, there are many 100+ year old, traditional cabins. They are 1000 sqft with 3 bedrooms. They were more for weekend use rather than full time occupancy. On the other end of the spectrum is some massive log homes. Some of these only list 3 or 4 bedrooms even though they are 8000 sqft. They have other rooms that sleep multiple people but not listed as bedrooms.

The other thing is that I don't want to get that specific. The more specific you get, the fewer data points you have. I wanted to have a broad idea of comparable units. Some folks have 2 bedroom units that sleep 16 people!

I am not saying that AirDNA is worthless. What I am saying is that it is a tool that gives you a general idea, but for some reason, it always seems optimistic. If you base your planning off of AirDNA alone, you will think you are going to kick a ton of a$$ and end up getting kicked in the a$$ instead.

Also, it doesn't take into account the intangibles. Things like customer communication, design style, etc. Those are the things that take a OK vacation rental into the Awesome range. I like to think that we are awesome. We have all 5 stars on both AirBNB and VRBO.

I will also say that I haven't tried AirDNA in about 4 years. Much could have changed. Every review I have read praises it's process, but I also noticed that the used saturated markets to test. On our lake there are less than 200 rentals. Some of the reviews use areas that have thousands of comparable units. AirDNA will get more accurate with more listings to compare.

If you are using it to get an idea of performance in Orlando or Gatlinburg area, then you will get a pretty solid idea. If you try it for Coffeeville Kansas, you will most likely get questionable results.

Bottom line, use it, but don't rely on it exclusively for your planning. Nothing you do will get you perfect results. In the end there might be some trial and error when doing pricing. We were pretty set on the value of our rental and priced it accordingly. I am getting ready to raise our prices for 2022.

@Jessica Lindner We've used AirDNA and gives the base framework to start calculations. However, in small markets as you can see the data is limited. So when you have small amounts of data on a specific asset type, it's hard to really get an accurate measurement. In any market, always verify with some local boots on the ground. Perhaps call some property managers or agents in the area. This is something we do every time for our short term rental strategy.