Seattle is a big city, at least in terms of its reach. It gave birth to Starbucks and Grunge; it plays the backdrop in films that have embedded themselves in the fabric of American culture. Its Space Needle has pierced the firmament of architecture since 1961. Still, while Seattle’s presence as a metropolis is undeniable, it’s actually not that big of a city. In fact, it doesn’t even have its own airport, but rather shares Seattle-Tacoma International Airport with neighboring Tacoma, Washington. Both places combined have a population of fewer than a million people. But, interestingly, it is one of the most-traveled airport in the country. Sea-Tac is a small city airport with a big city flow of passengers.
Throughout this series we’ve looked at a spectrum of major airports and their social data. The focus is always on how Lexalytics’ solutions can show you the big picture while preserving the individual voices behind high level insights. After all, changes will not occur if the brand can’t drill down to specific customer feedback. The Lexalytics web-based GUI dashboard, Semantria Storage & Visualization (SSV), allows stakeholders see how one comment can tell the whole story.
Jerry is angry
In our Sea-Tac social data, there is a lengthy complaint posted to Facebook that seems to touch all threads of the narrative contained in the full data set. It’s long, over a hundred words. Essentially the traveler tells a tale of rude staff, overly aggressive security and airport police, while referencing wayfinding issues. On its face, it reads like the testimonial of a person who just went through a very bad day. Yet, with the easy-to-use functions available on the web based dashboard, this review becomes an entry point to the entire story of this airport. By using SSV’s point-and-click filter feature, we can drill into this review and see the common themes running through thousands of the airport’s social media comments.
For the sake of this article, let’s name the Facebook commenter Jerry. Using the dashboard, you are able to view Jerry’s comment in their entirety. Notice, that when we drill down into the comment using the Document Browser we’re able to immediately ascertain the topics being discussed thanks to the topics at-a-glance in the right pane.
The dashboard scores all of the topics, themes, entities, and other variables for this single document in expandable lists within the right pane. So, for this review, the three most negatively scored topics are “Baggage-Management,” “Attitude,” and “Bathroom.” We can see that Baggage itself is relatively neutral, which reflects Jerry’s original sentiment. An encounter with security personnel in the bathroom caught Semantria’s eye. Yet, since this complaint isn’t technically about the bathrooms, it might just be an outlier. While you would have to pore over the data to determine this for certain before, the dashboard can answer that question with a single click.
A picture is worth a thousand words
How you visualize your data is up to you, and SSV provides plenty of options. You can have as many widgets on a dashboard as you like. Widgets are simply blocks containing an individual visualization. Widgets allow you to compare different elements of the dataset alongside unique viewpoints. This means you may break up the whole data set into relevant chunks, and then filter it across one specific topic, entity, theme, or category. For example, if we want to see how Sea-Tac bathrooms are being criticized in relation to all topics within this data set then we simply filter for that topic and sentiment value — bathrooms – very negative — with a click.
Dirty bathrooms and expensive food
It doesn’t take long to see that while this reviewer didn’t comment on the cleanliness of the bathrooms, many others did. From this we’re able to hear the broader conversation around bathrooms in the context of the airport as a whole. Bathroom criticism tends to be tied to complaints about cleanliness generally, no surprises there. However, negative reviews about bathroom facilities at Seattle-Tacoma appear frequently alongside concerns about cost. A number of customers feel the bathroom cleanliness adds insult to injury after what they deem to be an expensive meal. Sea-Tac can respond to one of these proximal complaints, bathroom cleanliness, while the other is the purview of third party vendors. In so doing, the customer experience may be improved overall.
We may repeat this process for any topic, theme, entity, or category Semantria picks up from the data set; to wit, insights based on thousands of unique opinions may be unlocked by simply exploring elements within a single document. Taking it further, we’re able to identify an issue with baggage management and baggage costs while examining positive reviews mentioning the attitude of the staff. Using these tools we can see if people are having an all-around bad experience or if the problem is more localized. Inherent to Semantria Storage & Visualization is the ability to store these insights and track them over time.
The tip of the (urinal) iceberg
For Sea-Tac, we see that when it comes to cleanliness they’ve still got a ways to go. When it comes to staff attitude, airport employees are improving, but there’s still room for growth. As one reviewer puts it: “The staff is professional, but not as attentive as they should be.” Checking these insights did not require our full team of experts to crunch the numbers. On the contrary, one person on a marketing team using the SSV dashboard handled it in seconds. It was easier than checking Facebook. That’s what makes this solution perfect for airports like Sea-Tac. SSV puts the full power of artificial-intelligence powered text analytics in a web-based dashboard that anyone can access. You don’t need a whole team just to make the most of your data. Rather this is a data tool your whole team can utilize on a regular basis.
The downside of “big data” is that it feels very impersonal. Sure, you get bigger insights from big volumes of data, but the individual voice of the customer may be lost. When this happens, the effort ultimately fails. That’s not something you have to worry about with Lexalytics. Using our natural language processing machine learning platform means airports and other companies don’t need to worry about missing a thing.