How important is Rose Bike’s customer data?
We at Rose Bikes have recognized the relevance of customer data and how this data can be used to add true value to the customer experience. We are therefore investing considerable effort in developing our competence in this area. We have set up a separate Customer Intelligence department in the company to manage this. The department deals exclusively with the generation, preparation and analysis of customer data. Many companies fall into a state of shock when it comes to the issue of data, because it seems like a black box. On taking a closer look, however, you will see that although it is infinitely complex, you can separate it into feasible steps. In principle, the same mechanisms are always at work. We are not trying to tackle the whole issue of big data all at once, but to approach the topic in a very tangible way based on use cases. For example, this could include using returns data to improve size recommendations. We first try to understand how results can be tweaked hands-on and which economies of scale could result from this. Then we look for a technical solution for this either within or outside the company.
How do you collect data on your users?
We are very lucky that we have been around for quite some time, more precisely since 1907. Due to our company’s long tradition, we already know a lot about our customers. We also have a long mail order history and have therefore always had to provide our customers with the right answers from a distance, because there was no salesperson standing next to them. In addition to this, we generate 80 percent of our revenues online. This means that a large part of the customer journey takes place digitally, making it easier to track. All in all, we try to collect data along the entire customer journey and the entire customer lifecycle. In order to achieve this, we use a number of data-producing points, such as loyalty programs, transaction data at digital touchpoints, contacts with customer service and, of course, product and purchase histories. An important future field is the Internet of Things. The connectivity of products will also allow a lot of exciting data to be generated in the future. In general, we try to collect huge amounts of data because nobody knows what we may need in the future. It’s better to have too much than too little. It is also very important to us to achieve a good mixture of quantitative and qualitative data collection. Quantitative data only indicates what is happening, but not why it is happening. That’s why we organize a lot of user labs on our own premises.
You also measure your customers. Why?
That’s right. In the stores we collect measurement data to be able to offer our customers the right product in the right size. For example, bicycles are adjusted based on body measurements. We also use a foot-measuring tool to be able to recommend the best cycling shoe. It is hard to try the shoes out in a store. By measuring their feet, however, we know if someone has a raised arch and can provide them more targeted advice. We also want to gradually transfer these added data values into an online experience and thus relieve our customers of having to decide which size suits them best. Product data is crucial to achieving this. After all, we not only require customer data, but also need to know what the product can provide.
Where do you see low hanging fruit in the use of customer data? And what is more complex?
I think intelligence that recognizes patterns in customer behavior for individual use cases can be realized within just a few weeks. This is not too hard to achieve using existing services, but it does add a lot of value. If, for example, you can provide correct size information, it will relieve your hotline noticeably. Another use case is the issue of when a package will arrive. Why should a hotline employee have to look up this information in the database for a customer? The information can also be provided automatically – and the service employee can spend more time with a customer who needs more personal advice when buying a bike. For traditional retailers with many offline sales, I quote Manuel Ludvigsen-Diekmann from Shopmacher, who always says, “digitize the information on your receipts first, because they contain many hints as to which items customers buy together. You can then use this information online for your cross-selling recommendations.”
To achieve success, however, it is important to establish data-driven work in the company first before trying to tackle big data as a whole. The Customer Intelligence department should avoid serving as a data gatekeeper, because this can quickly turn into a bottleneck. Instead, all relevant employees should have access to data. We are currently working on a self-service BI solution to make it easy for employees to display individual reports that clearly present the important information. There will also be alerts for employees to indicate that a KPI is changing significantly, for example. In this way, we will ensure that employees can actually work with existing data.
Data Scientists are the talk of the town. They are being celebrated as the new heroes, but are difficult to find on the market. What do you think about this?
We are also seeking Data Scientists and have received applications. In addition to this, there are now agencies that send Data Scientists to companies, create use cases with the companies and gradually train them to deal with the subject independently.
How do you use the data you collect?
In a multitude of ways. We use it to personalize the customer experience. And we use it for marketing and customer loyalty. The persona strategy that has been customary so far is far too unrefined. People don’t just behave according to seven basic patterns, but in entirely different ways in many situations. Customer data is also relevant for developing new features, avoiding returns and optimizing conversion rates. Pricing is a very important topic as well. We see ourselves as a price-performance leader and have the same prices across all channels. To achieve this, it is extremely important to be able to work in a data-driven way.
What are your plans for the future?
We are currently approaching the big topics by way of small use cases. As I mentioned, we first want to understand mechanisms manually, then achieve initial quick wins with simple plug & play solutions, and then develop in-house expertise and make the solutions great. In the future, our main focus will be on automating standard analysis processes and translating them into direct actions in real time. This will take us another giant step forward.
The bottom line
If you want to work in a data-driven way, the first thing you have to do is create the conditions within the company and give employees access to the data they need. After that, the issue must be approached bit by bit. By starting with small use cases, you will soon find that even Data Scientists just cook with water.
You might also be interested in these stories: