How a SCRUM team builds a customer data platform

How a SCRUM team builds a customer data platform

Gaining data and insights can be a complex process. As a major player in the market, you have to deal with data coming from many different internal and external systems. On its own, this data is interesting, but in combination, it is worth its weight in gold and can provide insights that allow customers to have a personalized experience.

To do this, you need customer data platform software that removes this complexity. Because a customer data platform takes care of combining and orchestrating all this data within the organization, and also to systems outside the organization, such as inventory management at Amazon or sending shipping information to an external system.

At Crystalloids, we develop data and analytics platforms in a SCRUM Agile way. I am Will Rulof, scrum master for the brands Body & Fit and Optimum Nutrition which are part of Glanbia, one of the Crystalloids customers, about what a customized customer data platform brings and how Crystalloids approaches development.

A customer data platform is the heart of all data flows

"Body & Fit had no (combined) insight at all into the data coming from the website, ERP, email, PIM, payment, and other sources. Another challenge was that they were migrating from Magento to SAP Hybris. All the history coming from Magento had to be easily imported into this new system."

The solution was a customer data platform. Because of the combination of components, we also call it a platform. With a NoSQL database at its core and an analytical data store, this acts as a kind of man-in-the-middle. Historical data is extracted from Magento, stored, and delivered to Hybris in the right format. “We have created a Contact API. When a customer account is created on the website, it first connects to this API. The API does a lookup to see if the customer exists. If it doesn't then the API creates a new customer and sends the information to Hybris and other systems that need the information such as Email Service Provider, ERP, reporting, etc. We also store all order and product information in the Data Platform. This data is sent to Google Cloud Pub/Sub and gets queued and then stored in Cloud Datastore and in analytical datastore BigQuery, and orchestrated to all the other systems connected to the platform that need the data."

Insights from data deliver a better customer experience

BigQuery is used to do analysis from combined data. From this, predictions can then be made. Besides customer and product information, BigQuery also stores other information. For example, information about customers who have left the shopping cart. "This makes it possible to send an e-mail with a request to finish the order as soon as someone leaves the shopping cart. Another requirement we have been working on is an "out of stock notifier." If a customer is interested in a product that is out of stock we can automatically send the customer an email once the product is back in stock." 

"In addition, we have also implemented Google Recommendations AI. With Google Recommendations AI we can show visitors exactly that which is interesting to them on the website. Suppose a customer comes to the homepage and clicks on some product pages and products, then the top 10 of products on the homepage are automatically determined by Google Recommendations, so there is a product range that exactly fits the behavior of the customer, the purchase probability is much higher.

How we get things done

Within Crystalloids, we work Agile and use the Scrum method. We work with biweekly sprints. Every day we have a standup where the product owner is also present. The product owner is usually provided by the customer. Together we talk through the Jira dashboard to see if we are still "on track". Every Thursday we do a backlog refinement and ask for input from the customer to create new user stories. Often we then schedule a meeting with stakeholders who know a lot about a particular topic". Then we get to work on estimations. We get to work and deliver the functionality. With each functionality we develop, the customer experience gets better and data flows more efficiently.


A customer data platform allows you to gain insight into the behavior of visitors and potential customers and ensures that systems use the same controlled data with a central data model. In this way, you can influence this behavior by offering a personalized experience based on combined and analyzed data and offering the visitor exactly what he is looking for at any time. Want to know more? Please contact us.



Crystalloids help companies improve their customer experiences and build marketing technology. Founded in 2006 in the Netherlands, Crystalloids builds crystal-clear solutions that turn customer data into information and knowledge into wisdom. As a leading Google Cloud Partner, Crystalloids combines experience in software development, data science, and marketing, making them one of a kind IT company. Using the Agile approach Crystalloids ensures that use cases show immediate value to their clients and frees their time to focus on decision making and less on programming.

No Comments Yet

Let us know what you think