How Data Disrupts Traditional Business Models

Digitally disruptive companies such as Uber, AirBnB, Amazon and Netflix have flipped entire industries on their heads. This disruption of traditional business models help explain why 52% of Fortune 500 companies from 2000 are nonexistent. By leveraging the power of data, these companies were able to understand their customer preferences and give them what they want without asking. Anyone from a startup entrepreneur to a veteran CEO of a big enterprise can benefit from understanding how they used data to drive innovation in their business.


Uber not only drives cars, it also runs algorithms. Every trip is recorded, even when no passenger is involved. Customers and drivers rate each other, the closest Uber car can be dispatched to a waiting passenger within seconds of his request, and the company can see whether a driver seems to be working for a competing company. How are they able to do all of this so efficiently? Simple answer — the data they collect.

Someone unfamiliar with data analytics might say: “Great, but so what?” But it is data which allows them to forecast supply and demand with a high degree of accuracy, predict customer wait times, and identify and avoid traffic bottlenecks. Drivers can make themselves more competitive by taking advantage of positioning suggestions based on historical trends and current events, while a complete view of the market allows fares to be set at a level that keeps both sellers and buyers coming back for more.

The thing to understand is that none of these decisions are made by a bunch of people sitting around and staring at pie charts and trend lines. The entire process occurs in real time without a need for the intervention of fallible humans, comprising complex, multi-dimensional analysis of the entire data set. This is a major part of what makes the Uber model so efficient, the other being that they chose a cool name.


Airbnb supports hosts on their platform by using data analytics to gauge the likelihood of a booking at a certain property on a given day at some price point, based on data collected from the millions of stays already booked. From the host’s perspective, the interface couldn’t be simpler: a price is suggested for every day of the year, while a calendar date is displayed in red or green depending on the chance of a booking being made for it. This also improves revenue by increasing booking rates and optimizing prices.

In terms of the guest experience, classifying search results is an algorithmically challenging process. Geographical proximity to various points plays an obvious role, as do rankings and user-defined preferences. How to best weigh the influence each has on how likely a guest will be to enjoy a stay with a particular host?

In this case, too, the optimum answer was found not through guesswork or industry knowledge, but the intelligent application of data science. Through several iterations and utilizing community insights, the company developed an algorithm which can show guests those locations within a destination where they are most likely to have great experiences. Based on statistical models and data visualization, this system can easily be applied to cities across the world without relying on simplistic measurements, e.g. distance from the city center.

Many people tend to see innovation as a process which results in a patent or product, but digging into data often reveals possible efficiencies where they are least expected. As a company with a particular culture, the AirBnB management team was concerned to realize that only 10% of new hires were women, although the proportion of female applicants was much higher. Their data science team was set the task of improving this, too, and while the exact details of what they found haven’t been made public, small, data-driven changes to the hiring process eventually doubled the rate of female hires at the company.


According to the most recent reliable figures, Netflix now boasts 104 million subscribers worldwide. Each is not only a source of revenue, but also of market and CX information.

In this industry, traditional data-gathering techniques such as Nielsen ratings are too nebulous to be of much practical use, while developing new projects can be expensive. Formerly, television networks relied heavily on intuition and rules of thumb when deciding where to invest their money.

Digital broadcasting, or streaming, turns this situation on its head. For example — Netflix may choose to examine the behavior of every customer who has ever watched a series such as Breaking Bad. Assuming that they limit this data set to those who started with the first episode of the first season and continued sequentially, they can understand and define the proportion of users who continued watching up to the latest available episode.

At this point, serious analysis can begin and they can determine the point where most viewers dropped off. Or try and see what time-based patterns emerge. One can empirically come up with many answers in order to obtain a deep understanding of the level of viewer engagement with this show. This data can be compared with that of other series when it comes to making decisions on which creative projects should be funded, with an excellent chance of attracting the predicted number of viewers.

This is not the end of the story, though. Every online or in-app action can be tracked, and there is no technical reason for not doing so. Events recorded by Netflix include the user pausing, rewinding or fast-forwarding, the date and time content is accessed, the user’s location, content abandoned before the end, type of device used, search queries and browsing behavior and of course user ratings. Key scenes within movies may also be subject to analysis in terms of audio volume, facial identification and similar characteristics amenable to machine analysis.

What does all of this mean to a manager sitting in Los Gatos, California? Simply put, it paints a broad and deep picture of every type of customer “persona”. Netflix has found that movies are more popular over weekends, while users prefer to watch shorter content during the workweek. This may seem like something that’s easy enough to predict, but what about the precise minimum amount of time that a user unlikely to cancel the service uses it?

Innovation is all About Being Customer-Centric

No company can effectively serve a market it doesn’t understand, and compared to older, hit-and-miss techniques of data gathering, data analytics offer a floodlight at midnight.

Facebook knows to a high degree of certainty which advertisements you’re most likely to respond to, while Uber recognizes exactly when and where they can surge prices. Netflix understands your views on subtitles and won’t try to market Shakespeare to you if you’re more of a sitcom fan.

Every customer-facing interaction yields new data to companies such as these, while late adopters of this technology lag behind in every way. Data-driven enterprises can take advantage of more efficient market mechanisms, as well as a far higher degree of personalization when it comes to marketing and their product offering, based on quantifiable data real time.


July Systems’ Proximity MX helps physical businesses gain actionable insights on guest behavior using existing WiFi or Beacon infrastructure.

ASK FOR A FREE DEMO to learn how Proximity MX can help you gain guest insights and engage with them real-time.