Return to Insights

Digital innovation requires data modernization


No matter what industry you work in, you likely rely on some degree of digital innovation to help your business get ahead. And while it’s easy to start by thinking about the end platforms that you will either adopt internally or deploy to your customer – after all, that’s the “sexy” part – I advocate for starting with something decidedly less glamorous: the data.

Digital innovation requires modernizing your data approach

You know that in the end your platform should deliver a customer-centric experience. This includes a rich set of capabilities and an intuitive interface. But for the experience to offer value to your customer, you need to feed it something: data. And that means accurate, relevant and timely data to support the work you are doing and to support what your customer is consuming.

Digital experiences accelerate everything, including the speed at which you can lose trust and credibility in the eyes of your customer. Serving up outdated or inaccurate data, or worse, becoming a victim of a data breach, are two of the fastest ways for your company to experience reputational harm.

Data is coming from newer – and faster – sources 

Chances are, your company relies or has relied on data sourced from transactional systems. And while those systems will always hold a high degree of value, it’s important to shift our thinking to new data sources that offer fast and tangible ways to access the data.

As you develop new digital solutions for your customers, you are now being faced with newer and more modern sources of data. Purchased data is now available in real-time or batch-time feed format, abandoning old practices of an annual refresh. Web logs, advertising platform data, social media and search have all made for an abundance of data that can be used to drive your customer’s digital experience.

Digital requires agility 

As you embrace new data sources, you will certainly feel a new sense of urgency. Acting on new customer insights needs to be done as soon as possible, because the data itself is outdated very quickly. Your organization should be prepared for the idea of “real-time everything,” and look for ways to speed up touchpoints.

Getting the right data in the hands of the right people at the right time is now truer than ever. You will be forced to look at your business practices and methodologies; if you haven’t already adopted some degree of Agile, you likely will need to make that a priority.

Simply put, legacy business intelligence inspired data architectures will not give you the speed-to-market that you need to keep up with real-time customer data. Where business intelligence tends to give you a good look into the rear-view mirror, Predictive and Prescriptive Analytics is what will really guide you into the future.

Data ethics should be your compass

Finally, any discussion of data is not complete without a discussion of data ethics. Legislation and regulations in the areas of data governance can’t keep up with the speed of change in our field. It is any data team’s responsibility to instill a culture of ethical data stewardship and security.

The lack of ethical controls on data have time and again led to new laws and regulations to officially govern data usage, two recent examples being GDPR and CCPA. Don’t wait for a law or regulation to approach your customer’s data from a customer-centric privacy perspective. Handle with care.

A good approach is to adopt a DigitalTwin approach for model development to safeguard sensitive customer information that has no or very little analytical value to begin with.

The need for DataOps

As your organization adopts a data integration strategy, I recommend you consider a true DataOps function. This is not the same thing as DevOps applied to data, but instead a dedicated data technology function that focused solely on the management and governance of how data flows throughout your organization.

I often use an analogy of an air traffic control center. Each operator in the center needs pin-point accuracy, an exact view, and cooperation with the centralized team in order to have an eye on what’s in the sky – and to avoid disasters. A DataOps team needs the same rigor, to see what data is coming in and where the data is going – to which platforms and systems – and how it is being used. 

Customer data is still personal

Finally, remember to look at your data as simply an extension of your customer. Data is personal. Data is human. While data can inform machine-learning algorithms and AI, remember that those technologies are not without their pitfalls. Your data is a tool for implementing technology solutions and platforms that benefit your customers. Data is not a replacement for your customer interaction.