These are the people that can explain you how they are creating value today, but for sure they know what their next step would be. For the engineer it could mean taking more data sources into consideration. The intern might miss out connecting to your factory data system. For the marketeer it is maybe having a way to connect the data to a deep-learning toolkit. Such stories help you understand what the business needs, and moreover: it helps you understand how the data platform could pay for itself.
You don’t know where the data comes from (yet)
Talking with the people who are already wrangling the data teaches you more about where the data comes from. For sure you are going to encounter anecdotes about the complexity of collecting the data, combining it and transforming it into a form that makes it possible run analyses on. This is the next step in your journey to help you understand what you need from the data platform.
Data collection is a tough game. Data sources are spread throughout your organization, across networks, vendors, technology stacks. Knowing what you need to connect helps you to understand what the platform should be capable of. Can the platform connect across your organizational boundaries like different networks, geographies, data centers? It helps shape the requirements in this area. For some a cloud solution is an easy way in, for others it could mean keeping it close to home on a datacenter inside the company.
You don’t know how the data is processed (yet)
Data platforms are not just big silos of bits and bytes. When you want a platform that is truly adding value over time there is a lot of compute and storage power involved as well. Understanding what your organization is currently already doing, and what future needs there are will help you better understand the computing requirements.
What you see often is that the initial business value that is created from data comes from simple visualization and dashboarding. However, listening to the stories of the people already doing the data processing you can teach new insights. Imagine the engineer who is shaping your products of the future, he might be using specific physics or mechanical models that need a place in your platform as well. Or the intern might need image data processing. And the marketeer who needs deep-learning techniques is going to need much more processing power.
Not taking this into the platform architecture might cut off corners that are hard to fix later on.
You don’t know how to make the data actionable (yet)
Getting a platform is not the goal. The goal is to make the insights from your data actionable and integrate them into your day-to-day business processes. The goal is to deliver insights to customers or to enhance your supply-chain. Different ways to embed those insights mean also new requirements for your platform.
Are you going to need a mobile phone app that brings the data at the finger tips of your logistics work force? Or do you need to satisfy your management with real-time scenario planning of demand and supply characteristics of your business?
In the first example you need to make sure your development team can leverage the data through micro-services and more traditional databases. In the latter example you need to connect resource-hungry client applications with data that can be queried in milli-seconds. Can your new platform deliver?
You don’t know how to make it operational (yet)
At the beginning of this journey, delivering solutions in a production environment still seems far away. But very soon, when use cases prove to be successful, the business is going to demand continuity. Here you learn that a platform is not just data and compute power. It needs people to run it, to industrialize your data. Do you currently have people who can debug issues in ingestion of data? Who is going to update the apps once your logistics personal is demanding new features?
The data platform is going to ask for a new approach from a new team of people with new skills and expertise. You want to make sure you grow the team with the platform. Make them part of the decision process and train them on the new capabilities. Move them close to the colleagues who are already creating value and teach them to understand the business language. This way, the team is integrated with business and seen as a valuable partner to create the insights and produce value.
You don’t have your business case (yet)
What you might know already, data platforms do not come cheap. Having to build the business case for the platform before you have created the actual value is going to be a challenge. Taking the approach that has been sketched here is going to remove that burden. You can explain the needs from the actual business value that is created, the business is going to demand the operationalization, and the return-on-investment is easy to explain.
So when you embark on your data journey: start small, talk with people, experiment and fail fast. Work with enthusiastic experts that can help you with each step of the way, and stay agile. Only then you will find the best platform that suits your needs. Since you are going to need one, although it is not the first step.
Back to overview