Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.

Skip to 0 minutes and 0 secondsSo after looking at the mental model, taking all the review text and putting it into a sentiment score for each property, we are then ready to take the rest of the data that are all numbers and plug it into this predictive model that we're going to see. So what is the rest of the data? We have the star rating, we got the price, we have the property attributes. All of these, can then go into a predictive model to predict the number of times a property is saved on Airbnb.

Skip to 0 minutes and 34 secondsThis predictive model, one could use regression which we will see in the module on regression, and so basically it takes all this information and historic information about all the properties and sees which of these factors actually influences you, to say, put a property on the saved wish list or rent the property and that's what we're going to see here. Now what we did for the purpose of this case study, was extract this data for all the properties of Airbnb in Miami and Paris from the web. So all of our public information that is available on Airbnb's website, we downloaded that and ran some analysis and predictive models on that.

Skip to 1 minute and 17 secondsAnd that's kind of what I want to share with you now as insights as to what does all of this analytics lead to in terms of insights about our marketing planning process and objectives that we are trying to address in the first place. You can look at the pictures here. We have, you know, Paris and Miami over here and we collected data from all of these properties and ran it through integration and what we found was, in Paris review sentiment and cleaning fees was important. In Miami price, security deposit and cleaning fees was important. Which is kind of interesting because you know without data, our intuition or our gut would say, you know, review sentiment is important.

Skip to 2 minutes and 6 secondsWe read reviews and that's what is influencing our decisions. But, you know, when we take a check here and go look at the data and dig that information, what it really reveals is that Airbnb needs a regional strategy. Some places are influenced by review sentiment, some are not. They're influenced by price and if you look at the sign, cleaning fees, higher cleaning fees means lower rental prospects in both Miami and Paris which makes sense. And price also has a negative influence on number of renters. So people, in general, like lower priced properties which, kind of, make sense and review sentiment was positive. People like to rent properties that have a high review sentiment.

Skip to 2 minutes and 53 secondsSo this kind of makes, you know, this what is called a smell test. It has face validity. The results make sense that fees have a lower effect and price has a negative effect and review sentiment has a positive effect and now we can see that using the data from all of our properties and putting it through an analysis in a predictive model like regression. We looked at how Airbnb strategy really changes based on the location. Whether it is Paris or Miami and how data actually improves our marketing planning process and decisions really.

Utilizing Data to Improve Marketing Strategy

After running an analysis of the Airbnb’s data from the Paris and Miami rental markets, surprising insights are revealed about what influences rental activity in each region. Learn how Airbnb can improve its strategy based on the results of this analysis.

Share this video:

This video is from the free online course:

Marketing Analytics

Darden School of Business, University of Virginia

Contact FutureLearn for Support