Data Gathering 
Currently Market Dashboards is using scraped data from Airbnb (and VRBO too). For a listing, all data we gather could be found by going to that listing’s public Airbnb/VRBO page and looking through their calendar and the listing info they provide. We then keep records of how their calendar has changed, what dates have become unavailable (or re-available), how their prices have changed, etc. and we build up a history for that listing. We currently do this for all listings that appear on Airbnb/VRBO. Using scraped data enables us to provide Market Dashboards for any location around the world (regardless of whether we have customers there or not). 

Block Removal 
One of the main challenges for scraped data is that there is no guaranteed way of determining if specific dates are not available because they were booked or if the owner has decided to block those dates. Everyone using scraped data faces this issue and generally has some way in which to try and remove these blocked dates from the data, and no method is perfect. PriceLabs has its own block removal logic that looks at patterns in the whole market as well as individual listing data to help us determine whether a booking is real or a block. Some factors of a booking we look at to determine if it is a block or not are: Length of Stay, Booking Window, Market Occupancy, extreme Price variations, and more. We also automatically remove any stay greater than 60 days as we feel they do not fall under the Short-Term Rental category and can have a large impact on the data. Once a block has been found, the corresponding dates for that listing are changed to available and do still count as the listing being empty when calculating market level Occupancy for those dates. All other booking info is also removed.

Bookings made on other OTAs or Direct Website 
As long as a property is listed on Airbnb in addition to other OTAs, we still are able to deduce bookings on other OTAs as they block the calendar on Airbnb. As described above, once a date is unavailable (either due to a booking on Airbnb, or from any other OTA) we use our block detection logic to identify whether it's a booking (made through any channel) or a blocked date. 

Dynamic Pricing Analysis for listings in your market/comp-set
When creating listing comp-sets, you will see a column named "Dynamic Pricing." Here's what it means:


We track the prices of each listing and score them on a 0 to 1 scale on both day to day to day price variation as well as how much their prices have changed for the same day since the last time we checked. A 1.0 would indicating their prices vary strongly from one day to the next and are constantly changing, while a zero indicates the price is always constant. I can't divulge exactly how this score is calculated but to give you a rough reference an average PriceLabs user generally falls into the 0.6 - 0.8 range depending on their customizations.

For the table we then bin these scores:
None: score 0.0 - 0.1, these listings almost never change their prices
Low: score 0.1 - 0.25, these listings may change their prices for holidays and events, but their overall price week to week is pretty constant
Moderate: score 0.25 - 0.5, these listings prices vary week to week but the variance is overall small and prices don't update too often (someone manually updating their prices every week may fall into this category)
High: score >0.5, these listings prices account for DOW and holiday/event demand changes and update every few days (listing is most likely using some dynamic pricing software to update their prices on a daily basis).

Frequently Asked Questions

1. Why do we sometimes use Median instead of Mean? 
   There tends to be a few outliers in every market that set extremely high prices or only take extremely long bookings, the listings are not indicative of the market but would have a large effect when calculating the mean for these values. Median on the other hand is more stable and resistant to outlier behavior.
   
2. Why is the area covered in the listing map smaller than the inputted radius? 
   We currently cap the number of active listings in each Dashboard to 1,000. If the area you selected has more than 1,000 listings in it we will only show you data for the closest 1,000.
   
3. In the listing table (under Location Map) how do you classify professionally managed listings? 
   We use 4 categories: Individual, Small, Moderate, Large. The individual category is for property managers that only have a single listing tied to their account, Small means they have between 2 and 10 listings, Moderate between 11 and 50 listings, and Large is 51 or more listings. Unknown may appear if we haven’t yet determined how many listings the property manager has.
   
4. Is there a way to search for Airbnb Luxe listings using the dashboard? 
   Not currently
   
5. Why do some listings have extremely high occupancy/revenue in the past 30 days even though the market as a whole isn’t doing well? 
   It is likely that at least some of the bookings we have assigned to that listing are actually blocks that we missed. Our block removal isn’t perfect but do send us cases you find suspicious so that we can try to improve!