Thanks, data science, for another record Cyber Monday

The numbers for Cyber Monday are staggering, but shouldn't be surprising, argues Dataiku CEO, Florian Douetteau.

Since its inception in the early 2000s, Cyber Monday quickly rose as the shopping day of the year around the world, even usurping long-time favourite Black Friday, which had been around since the 1950s. And for the past several years, we’ve woken up Tuesday with a deal-hunting hangover to news of record-breaking profits. 

Just like Singles Day in China, a day on November 11 created to celebrate young Chinese people who are proud of being single, e-commerce giants reported record-breaking online sales on November 28, 2016. The numbers are staggering - more than $3 billion collectively in online sales for Cyber Monday, and more than $17 billion in online sales for Alibaba alone in China on Singles Day. Staggering… but surprising? It shouldn’t be.

Big data: Big player for online shopping, thanks to user data  

In recent years, businesses in every industry (from agriculture to insurance) have started to see the value of big data, a buzzword that essentially means processing large, complex datasets. Corporations are investing large amounts of money in curating, analysing, and predicting based off of all that data, and this is especially true in e-commerce. 

Big data is particularly important to and interesting for e-commerce businesses because of the trove of information they have about customers shopping online. Broadly-speaking, there are three types of data (transaction, interaction, and external) that, when combined, provide a holistic view of user data. 

Understanding user data and its representative elements does not only help retailers to define where user information is coming from and how its complexity has increased over the last decade, but moreover how it can be used to dynamically segment user databases.

3 types of user data

User data results from three kinds of data:

  • Transaction data, which gives a peek into customers’ buying habits. This is one of the oldest data types and reflects a wide variety of customer-centric data including time, location, price, payment methods, discount values, quantity purchased, etc. All of this data can be combined to convey a precise picture of customer shopping habits and interests.
  • Interaction data, which enables a shift from snapshot to screenplay. In the 1980s, consumer-to-business communication channels were limited to a handful of options such as customer’s written feedback, complaint letters, etc. The digital era now enables companies to follow customers and prospects on all channels whether it’s website interactions, social media, email, phone conversations, or text messages. An almost one-to-one relationship can be created and, when combined with other points of interaction, allows e-commerce businesses to formulate a global customer view.
  • External data, which means going from tunnel vision to global perspective. External data is defined as all data outside of an organisation’s internal operating systems. Historically, this type of data was hampered by the traditional segmentation approach: external data was limited and, when available, only data that fit within the confines of segmentation rules was considered (e.g., average age group, interests filtered by location, etc.). But organisations can now tap into a variety of dimensional data in order to add layers of meaning to customer behaviour. For example, geographic and socio-demographic datasets can be used to provide deep customer insights: how will traffic congestion in a specific area affect retail outlet visits? How will the weather affect foot traffic for outdoor locations?

Better segmentation for better use cases  

Making sense and predicting future trends in user data is made possible by machine learning. Compared to the process of traditional segmentation, advanced analytics (like machine learning) empower companies to explore data on a much more granular and dynamic level. 

Advanced modelling, testing, and visualisation techniques combine to provide detailed predictive insights turned into specific use cases like surfacing the right offers for specific shoppers or reducing customer churn.

Enhanced and advanced product recommendations

Of course, the most obvious way in which e-commerce uses data is to determine which products go where. In terms of data analysis, this means expanding the range of customer signals that could be captured and analysed. Retailers can now capture and make sense of large amounts of data, develop effective customer segmentation, and eventually implement an entirely new non-rule-based approach for analysing incoming and historical data.    

The strategy starts with establishing a mechanism for collecting data from customers’ online behaviour, such as click paths and bookmarking. With the collected data, the focus shifts to creating a machine learning-derived score for each customer - essentially a value that reflects the likelihood of members pursuing specific offers.    

This coupling of online behavioural data and tailored offer selections enables retailers to automatically present relevant buying opportunities that have the highest likelihood of customer acceptance, resulting in significant increase in revenue per user.

Pricing

Just like the business of targeting, properly pricing is deeply rooted in big data. This is one of the biggest ways that online retailers can stay competitive and not lose any sales to other e-commerce sites - with big data, they are able to monitor in real time any better pricing available and make adjustments with little human intervention.   

Of course, there’s also deciding what deals and sales to offer on Cyber Monday, which (surprise!) is also a science, not an art. With optimal segmentation and yield experiences, the most optimal deals for both the customer and the business can be offered.  

Targeting

Not all advertising is created equal, and no one knows that better than e-commerce businesses. Even on Cyber Monday where everyone is looking for a deal, almost every online store still seems to send out a marketing email.   But in general, everyone gets the same email on big shopping days, which is a missed opportunity given that big data today allows for targeting based on a global customer image and a combination of transaction, interaction, and external data sources.    

This means that companies now have the capability to truly understand their customers: how do they pay for goods? What do they like on social media? What is the traffic like when they typically visit your store? Model segmentation data breaks the mould of older techniques by collecting data directly from the customer instead of relying on marketers to frame data questions ahead of time. The new granular, model-based segmentation allows for perfectly crafted email campaigns.  

Who gets what email and at what time with what featured deal and messaging is carefully planned out per model-based segmentation. And many e-commerce businesses are able to pinpoint customers likely to convert with incredible accuracy.   

Customer loyalty: Customer service and technical considerations  

Even more back-of-house details on and in preparation for Cyber Monday all tie back to big data. For example, predicting technical outages and issues or decisions like how much staff to provide on customer service phone lines or chat.    

What’s more, when a customer does inevitably call or reach out on email or chat for help, customer service representatives might have troves of data (like historical transactions and even insights from their social media presence) to use in assisting that specific person and providing the best possible experience.    

Going one step further, many business (including e-commerce) are using machine learning techniques to move toward much better automation to remove humans altogether in a way that doesn’t infuriate customers. Though there’s still a long way to go on this front, it’s definitely in the realm of possibility given advances in artificial intelligence on unstructured datasets.

No surprise here

If all teams from product to business are involved in and completely devoted to making Cyber Monday a smashing success, is it really a wonder that it breaks new records every year? Each year, further sophistication in predictive modelling and advances in machine learning technology mean getting highly motivated customers to take certain actions and, essentially, buy more. 

So with the ever-growing amount of data available to e-commerce businesses and the buzzing Internet of Things (IoT) poised for success in 2017, don’t be surprised at the headlines about Cyber Monday on Tuesday, November 28, 2017. 

Image source: Shutterstock/Maxx-Studio
Florian Douetteau, CEO and co-founder, Dataiku