Four examples of possible data science analyses

In the last two years alone, 90% of the world’s data has been generated, and that’s a lot of data. With such a drastic increase in the data flow, new data tools have emerged. Tools that aim to create value from the data that is collected.

Data Science analyses is the perfect blend of data inference, algorithm development, and technology. Mix them together, and the data scientist can solve complex analytical problems. Data Science analyses is like a chameleon – it adapts to the environment, the question asked, and available data. It is not a standalone field but a combination of different areas, data sets, and business strategies.

Data Science encompasses all the tools, techniques, and technologies that help businesses use and understand data to their advantage. 

Data science is the use of data to make better decisions, with analytics for insights, statistics for causality, and machine learning for predictions.

Annemette Møhl | Co-founder Borbaki

The number of possible data science analyses is endless. We have selected the following four analytics for companies in the early stages of integrating data into their workflows and are most likely to achieve the best results.

We will cover:  

  • Product recommendation analysis
  • Customer churn analysis
  • Customer segmentation
  • Customer lifetime value

All four analyses can be connected and influence each other. But to create the best understanding, we have chosen to divide them as much as possible. 

Product recommendation analysis

Your customers have many preferences, and as a business, you are constantly trying to understand and meet the wants and needs of your customers.

In cities like Silicon Valley and Shanghai, recommendation analysis is one of the most common and normal data science analyses. Recommendation analysis allows you to determine the buying behavior of your customers, thus giving you the ability to predict which products are most likely to be brought together and which product the customer will demand after x amount of time.

The analysis gives you the ability to predict which product your customer most likely would have purchased if they were presented with the product before checkout. So you can use recommendation analysis to specify cross-sell products in your e-commerce webshop or in our physical store. Recommendation analysis gives you the insight to personalize your email marketing, online advertising, and sales to the different customer segments.

A recommendation analysis is a smarter approach to filtering information than simple cross-sell plugins. A recommendation analysis is based on your customers’ previous buying behavior, the behavior of similar users, and the content.

From real life: 

  •’s conversion rate increased 20% after they implemented recommendation analysis-based cross-selling on their webshop.
  • Research by Barilliance found that 31% of the revenue from e-commerce webshop is coming from cross-selling products

Customer churn analysis

Losing customers is costly: The lost sales and the fact that the customer now buys from the competition.

The cost of acquiring new customers can be as much as 700% higher than the cost of retaining a customer. However, if you retain 5% of your existing customers, you can increase your profits between 25% and 95%. Yes, up to 95%! Those are pretty significant numbers.

So it’s clear that customer retention analysis is one of the most important areas to focus on. Retained customers are also more likely to be engaged and open to up-selling and cross-selling.

The heart of churn management is recognizing the warning signs of potential attritors. If you can identify at an early stage that a particular customer is likely to leave your business, you can take proactive steps to prevent this from happening.

By understanding your customer churn and retention rates, you can identify how well your product fits the market and optimize the areas where it needs improvement. In addition, it can help you rebuild customer relationships with those customers who are about to churn.

From real life: 

  • In the banking sector, it has been shown that just by targeting customers who are most likely to churn early on, 11% of churn can be avoided.
  • The SaaS company reduced customer churn by 71% through the results of a customer churn analysis.

Customer segmentation

Understanding your customers is critical if you want to maintain or build a strong customer relation. has created a list of the top 8 reasons for customer churn, and in the first place, we find: Attracting the wrong customers.

You have put tons of time and effort into creating great e-commerce with cool marketing materials – and to be honest. There’s nothing like turning that hard work into money. If you are using the same marketing strategies, content, and approaches for all of your clients, we suggest changing your mindset. I mean, if you go into a store to buy a new pair of jeans, you do not want the cashier to show you the latest running shoes.

By dividing your customers into different clusters, you can create specialized marketing campaigns and offers tailored to each segment. We are not just talking about the usual demographic segmentations here, like age, gender, or income.

Identifying different customer segments is primarily about creating a better understanding of your customers, their buying behavior, and how they interact with your brand.

Segmenting your customers also allows you to divide your customers into groups based on which customer groups generate the most revenue for you. This knowledge will help you focus on the customers that actually make you money, and it can help you change the way you work so that your customers in category 2 or 3 can be grouped into group 4.

From real life: 

  • The insurance company Metlife used customer segmentation to determine that their customers could be divided into five groups, with two groups representing 80% of all customers. Therefore, they were able to adjust targeting to the personas that generated the most negligible revenue.  
  • Webhosting Canada realized that its users could be divided into three groups. The people in group 3 disliked everything about the company, and there was nothing they could do about it. So they stopped trying and instead focused their marketing spend on the people they were most likely to turn into loyal customers. 
  • Apicbase, a restaurant management software, increased its conversion rate from leads to MQL from 2% to 6% by personalizing emails created based on its newly discovered customer segmentation.

Customer Lifetime Value

Many companies do not know which of their customers is most valuable to them. This leads to business strategies where marketing spends, targeting, sales efforts, etc., are spread evenly across all customers, rather than spending the most money on the customers that bring the most profit to the business.

By understanding when a customer is at risk of churn, you can focus on the customers that bring you the most value. In short, the higher the customer lifetime value, the greater the profit.

Customer Lifetime Value is about dividing your customers into groups based on the number of orders and total earnings you expect to receive in the future. Instead of spending your entire marketing budget on customers who will most likely make sporadic purchases, you can focus your marketing on the customers who will generate the most value for your business.

From real life:

  • Hear and play music, a provider of music education products, developed personalized marketing that increased customer lifetime value by 416% by understanding its customers based on their lifetime value.
  • By calculating the lifetime value for half of its customers, Retina created a similar audience based on the 20% of customers with the highest lifetime value, resulting in an 8.1x.