Editor’s note: Customer data analytics promises outcomes vital for product management, marketing and sales in B2C. At the same time, approaches that work well for financial or operational analytics are not suitable for customer data, diverse and vibrant by their nature. Alex places his 25-year field experience at your disposal. If you feel you need help with the design or implementation of customer data analytics, you are welcome to consider ScienceSoft’s data analytics services.
The Customer is King concept came into being long ago. Businesses have understood that unique customer experience is their competitive weapon that boosts conversions in the short term and builds customer loyalty in the long run. And as companies need to know their customers to create such experience, customer data analytics (aka customer data analysis and consumer analytics) has become important.
However, the adoption of contemporary methods in customer data analytics is slow and many businesses fail to understand their customers as well as they want. The consequences are quite alarming:
- The decisions in product management, marketing and sales are not optimal, which results in unsuccessful products, slow acquisition of new customers, the need to provide more discounts to keep sales stable.
- The forecasting is not precise enough, which sometimes leads to stock-outs or excessive production/inventory.
A burning need to know their customers better is on the mind of many marketing and sales leaders. “If only we knew our customers better, we would be able to increase sales by 5%-10% and improve margin contribution along the way…”
I am always enthusiastic about using data analytics to reach new business heights and happy to see when businesses undertake strategic initiatives in this area. If you are one of them, let me share my experience and recommendations that may be useful.
Customer analytics is a process of collecting and analyzing customer data to learn customer behavior and preferences for making strategic and tactical business decisions, as well as automatically forming personalized recommendations.
Intended to provide the answers to diverse customer-related questions, customer analytics can embrace different types of business analytics. Businesses can apply descriptive analytics and get a clear picture of their status quo. They can use diagnostics analytics to find out the roots of a particular business problem. Companies can even enjoy the advanced capabilities of predictive and prescriptive analytics. Relying on data science and machine learning, these two types can provide forecasts and recommend actions that a business can take.
To give you an idea of how it works in practice, let me share the examples of insights into the revenue generated by different customer segments that an ecommerce retailer can get leveraging customer analytics. The examples are based on a case with ScienceSoft’s customer running an omnichannel retail, hotel, restaurant, and other businesses (numbers have been changed to ensure confidentiality):
Of all customer segments, value-seekers generate the lowest revenue (descriptive analytics). While an average customer from this segment used to buy three products (two of which are at a discount), now they buy just two discounted products, which leads to the revenue drop by 12% (diagnostic analytics). The analytical system estimates how much revenue can be gained if these customers are offered a 3%, a 5% or a 7% coupon for their next purchase upon the condition that they buy four products (predictive analytics). Of these three scenarios, the system recommends offering a 5% coupon as it will lead to the maximum revenue increase for this segment (prescriptive analytics).
To run customer data analytics, companies can use diverse customer-related data (which, nowadays, mainly belong to big data rather than traditional data). Here, we look at 4 main customer data types to find out how companies in different industries can use them.
1. Transactional data
It’s easy to explain transactional data through retail, where each purchase deepens a company’s understanding of its customers’ journeys. For example, the analytical solution we implemented for an FMCG company gathers data from retailers and allows identifying sales trends, find out which SKUs and stores showed the best performance, estimate growth potential as well as optimize sales and marketing activities.
2. Data about service/product use
Manufacturers can examine the data about product use to create a better customer experience and innovate. For instance, a couple of ScienceSoft’s clients from the connected car domain gather vast information about car location, a driver’s behavior, the level of fuel and fluids, the condition of the brakes and any faults detected by the on-board vehicle control units. This info can be used by car manufacturers to improve car design and understand usage patterns.
3. Web behavior data
A company can analyze every move that their website visitors make: where they come from, which pages they open, how deep visitors’ engagement is, etc. With this data at hand, the company can create relevant content to increase conversion rates. Ecommerce retailers apply this logic to track customer behavior, identify customer preferences and make product recommendations with the help of predictive analytics.
4. Data from customer-created texts
Customers take the opportunity to share their personal impressions about a product or a service in the form of an online review or a social media post. Companies can study this content to get a clue about what their customers think about their brand, product or service by identifying trends, recognizing a positive or negative emotional tone of each piece of text, revealing complaints and problems to solve. For example, Samsung uses social media analytics to attract customers from their competitors.
Root cause analysis
Having information at hand seems useful and even mandatory. However, if implemented without a purpose in mind, data analytics may result in hundreds of reports that show the situation but do not give a hint about what the reasons are and what to do about it. You’ll probably think: “Some categories grow, some decline, but what’s next?”
Fortunately, modern analytical tools go far beyond just a collection of reports and charts. For example, Power BI, a tool from Microsoft that is my first choice for enterprise BI, provides drill-down capabilities and convenience for exploratory analysis to get to the root of the situation. As a result, managers may avoid a reactive approach and balancing greed and fear, and make grounded decisions.
Forecasting and predictions that work
Many clients I talk to see forecasting as a tempting but suspicious initiative. Many marketing managers ask “Is it at all possible to achieve the needed precision in forecasting and prediction? So many factors influence buying decisions – the economic situation, fashion, personal circumstances…”
Predictive analytics is no miracle, as you may see in this example of how it helps to solve a manufacture’s day-to-day issues. Artificial neural networks, more precisely, convolutional neural networks, take the analytics to the new level as they capture a trend based on multiple factors and give the best possible forecast. The result is not avoiding uncertainty completely (which is impossible), but reducing it to a minimum.
Forecasting is a big topic and ScienceSoft does a lot in this area. You are welcome to look at the detailed description of demand forecasting prepared by my colleague Irene Mikhailouskaya.
Agile analytics for dynamic markets
Who is the best friend of a marketing manager or a product manager? The answer I often get is – It’s Excel with all its power and flexibility.
However good Excel is, it can hardly cope with current customer analytics tasks. The good news is the self-service BI tools like Power BI, Tableau, Qlik allow being agile in analytics. Power BI (my favorite) provides close-to-Excel flexibility. Besides, it’s reliable, brings human error down to a minimum, and can be used enterprise-wide. Power BI can work with big data using, for example, Azure capabilities.
Integrating external data
Using external data is a common practice today. Cloud and API technologies allow getting data from supply chain partners and data providers quickly and cost-effectively.
Remember the FMCG manufacturer I mentioned earlier? They get all their customer data from retailers.
One more example, now on a larger scale: our client’s business is consumer data analytics and they obtain data from advertisers from around the world.
There are four options of how a business can get customer analytics:
- Buy a ready-to-use customer analytics tool.
- Implement a customer analytics solution with the efforts of an in-house team.
- Involve a consulting and implementation vendor.
- Outsource customer data analytics.
Before choosing any of these options, a business needs to know in detail the advantages and disadvantages of each. We invite you to check the comparison table below and weigh the alternatives.
If you consider data analytics outsourcing or consulting and implementation services, you can follow the relevant links and read about each service in more detail.
As you might see from ScienceSoft’s practice I shared, no matter the company size, customer data analytics helps to attain the following 10 goals:
- Increasing conversion rates.
- Segmenting customers based on similar behavior patterns to target their needs better.
- Understanding customer journeys (what products and when they buy, what channels they prefer).
- Forecasting sales.
- Predicting customers’ response to marketing activities and offering them relevant products and promotions.
- Reducing the costs of marketing campaigns.
- Establishing targeted/personalized communication.
- Improving customer satisfaction.
- Increasing customer loyalty and retention.
- Optimizing product portfolio to better satisfy consumer needs.
Implementing customer data analytics your company needs is not going to be an easy journey. But rest assured it is going to be beneficial for sales, marketing, product development, and customer service. My colleagues at ScienceSoft and I are always happy to support companies at any stage of a data analytics journey, with everything from consulting to complete implementation. If you are interested, please let me know.
Time to Benefit from Customer Data Analytics
Partner with a professional consulting, implementation and outsourcing vendor with 32 years of experience to convert your customer data into insights.