Does anyone find AirDNA to be WAY way off in some scenarios?

Question

Esteemed listmates,

I am looking at a particular property in a market that I have some ok knowledge of. (Have been poking around there for awhile, haven't bought yet, know a local PM who is a great guy and gotten alot of data from him, done alot of database research, been in escrow a few times, but never panned out) Not a complete noob, but also very far from anything resembling really knowledgable.

So i am looking at a property, 3/3 2800 sf poolhouse on acreage, has all (or many) of the desireable characteristics that one would want in this market. However- The AirBNB revenue estimator tool from BP (And AirDNA also, as I have that too)- estimates the revenue at $48-50k yearly. I would estimate something like this at about 100k.

AirDNA is literally off by a mile. That bothers me alot to see, as I have the nagging question "Does AirDNA know something about that partucular spot that I don't?" The PM and the agent don't seem to think so, but of course they want me to buy it.

Query: Has anyone had a sitch where AirDNA estimator is just a million miles off? Especially in a market with alot of STRS? (This is right around Yosemite Natl Park so lots of STR data to go from) (also FWIW awning has it at 92k which is a little closer to what I would think- but i also find airdna to me more reliable in general than awning)

Nathan Gesner · Answer

It's a computer algorithm. The lack of accuracy should not surprise you. It is able to calculate based on square footage, number of beds/baths, crime scores, and other hard data. However, it can't assess intangible things like the appearance of the neighbor's lawn, the slope of a driveway, the quality of appliances, the artwork on the wall, or the cleanliness of the pool.

I would only use AirDNA to compare markets. Evaluating a specific home requires research if you want accuracy. The same goes for Zillow sales estimates.

John Underwood · Answer

Yes it can be very inaccurate at time. Best just used for a 500ft view then check manually from there.

Collin Hays · Answer

AirDNA is a lot like carfax.  It’s a single datapoint. I wouldn’t put much stock in what AirDNA says.

Luka Milicevic · Answer

From every scenario I have run and seen with AirDNA I find it completely useless.

This is in particular for my local market of Nashville. It shows revenue projections that are about half of what they should be, and this is the case for every single one I've looked at.

I had one scenario run by an investor that came up with 70k annual revenue on a high end, 4 bedroom unit. I had 4 different property managers run a pro forma on revenue and came up with 120-135k/year. I have a buddy running one down the street that is smaller and not nearly as nice and he is pulling in 120k a year.

From my personal experience, it's not a very useful tool. It's an algorithm, much like Zillow's zestimate that ended up costing them billions of dollars in losses bc they bought houses based on random estimates.

Valerie Budd · Answer

We met with AirDNA at a STR Data conference in August 2022.  In our discussions with them we ( the largest STR Property Manager in our area) found AirDNA was using data of timeshares managed by a company that had numerous timeshares around the nation.  These properties were not being used as STR's but timeshares.  We also found AirDNA was not using any of our 200+ managed properties that consist of 2 bedroom to 12 bedroom stand alone homes.  Their data for our area was primarily based on 2 bedroom condos.  After our visit they began manually adding our properties to use for their algorithms.

Nicole Balsamo · Answer

Have you ever tried data.rabbu? You can look at nearby comps and select/deselect them to get a revenue projection but for free as opposed to paying for a subscription. I'm all for comparing multiple data points, just wanted to pass along this info in case it case save you a few bucks on a subscription!

Will Fraser · Answer

As noted by the other responders so far Airdna indeed is regularly off, and I have found it to usually be low, but I wouldn't build a business around the assumption that is will always be low.

Personally I use it as a starting point, discount the annual rev projections, then make sure my deal will work around those assumptions.  Then I manually check a few comps and see if I can bring my number up.

Michael Baum · Answer

Hey @Richard E., how did you come up with the 100k? That would be my first question. What makes it worth that?

I have found AirDNA to be off as well but it is much more positive in my situation.

Andrew Steffens · Answer

Hey Richard

I am a PM in FL with 80+ units, of which 75% are client owned in a pool of about 40 clients.  I also rep them as their agent on new acquisitions often, so I have dealt with this situation quite often.  I have experienced this on varying levels fairly often, and even just today.  I just comped a a nearly $2M house for a client where AirDNA returned an estimate of $107k, about 1/2 to 1/3 of what is would actually bring in.

- Do you have the subscription to that area where you can see the comps that are making up their estimate?

If not, I implore you to.  We have it and it costs about $200 per month.

I was able to view all the comps, and all of the comps were 2/2 condos, 3/2 non pool homes, 3/2 landlocked homes, etc.  This is a uniquely beautiful property on the intracoastal waterway, with a pool.  There was not a single even really comparable property.  In this situation AirDNA is fairly useless and I would need to search in similar towns to find similar properties or rely on my intimate knowledge of the market.

I hope this helps, feel free to reach out if you have any other questions!

Kyle Wise · Answer

Like some have mentioned, I use AirDNA to quickly evaluate certain markets not compare individual houses. There are so many different variables that AirDNA doesn't take into account however, if the data is coming in lower it might reveal an overall trend happening in that particular market. For example, even though I noticed the data was on the low side, when I did go into comparable calendars they we're in fact charging lower rates and getting lower occupancy in a lot of cases.

People referencing what others made last year need to be careful because we all know the market has changed this year and a lot of individuals aren't seeing that same ADR and occupancy levels they saw in previous years.

Does anyone find AirDNA to be WAY way off in some scenarios?

Most Popular Reply