How to build a data platform.

And what you need to know to make sure it is a success.

​Today we share a few tips on how to build a data platform, why you need to understand your intended audience and the questions they want answered. When developing a data platform, we also look at how it should be built and why you need to understand what your broader organisational objectives are.

Identify Your Audience

What questions are you trying to answer?

The first step in identifying what questions you are trying to answer is to first identify the audience. Who has questions that need to be answered and what information do they need? 

Different users of the data will want different things out of it. For instance, a Data Scientist will want access to more raw data tools so they can write their own code and do their own thing with it, whereas a Manager or Executive might want a pre-built dashboard that gives them a high level of aggregate data that allows them to accurately make decisions and plan for the future A Data Analyst is more tech-savvy and may want access to data through an easy to understand semantic layer.

A semantic layer is a business representation of data that helps end-users access the data using common business terms, such as product, customer, revenue. This gives the user a unified and consolidated view of the data across the business. 

These requirements are all valid and will help determine what data needs to be captured, how it needs to be captured and what outputs are ultimately required for your audience.


What data do I need?

You have identified your audience, the questions they have and the answers they are looking for, so now you need to identify what data you need to provide these solutions for your audience.  Some key questions to ask yourself include: 

  • What data currently exists in my organisation that I can use? 
  • Where/how is my data stored?  
  • How can I get access to it? 
  • How many data sets need to be brought together? 
  • Can I use external data (ie. From other organisations, industry bodies, publicly available, etc?)  
  • Do I need to use external data or is my internal data sufficient? 
  • How will I source the data? 
  • Is the data of sufficient quality for my intended use? 
  • Who owns the data and is able to answer questions when data issues arise such as quality, usage, meaning, security, etc. 

Data Availability

What ways do I need to make this data available?

As we identified above, the output of the data will depend on the user and their requirements. There are a spectrum of tools that are available for you to use. 

Does it need to be available on a report, on a dashboard, or on a device? How are you going to display your data? Do you need the data in real-time? 

At Data Agilitywe work with tools and platforms including Power BI, Tableau and Azure (and more). We work with these tools and implement them so the data is available as required.

Data Architecture

Designing the data analytics platform.

As you work through the considerations of designing a data analytics platform, you will need to determine which architectural approach for your data that you will use. 

Depending on the data subject areas (single domain vs. enterprise-wide), whether the data is structured or unstructured and whether you plan to prepare and aggregate the data or leave it in a raw state are all considerations.

Traditional data architectures typically involved data warehouses and data marts that served up structured and often aggregated data ready for reporting and analytics. 

With ever-increasing volumes of unstructured data, we need to take a different approach.

Traditional architectures are now giving way to other approaches, such as data lakes, which can benefit from the scalable and distributed processing/compute power that is available on demand in the cloud.

Final Thoughts

Important considerations.

At Data Agility, we have worked with hundreds of organisations to support them with their data needs across decades and a wide range of industries. Our experience has shown that when there is consideration to the broader design of your data requirements, this will support the successful development, design and implementation of your data analytics platform. 

While there’s a growing trend towards agile project delivery, it’s important to have an understanding of the source data and appropriate logical data models in place to drive development of the physical implementation during iterative delivery. 

Incremental building of data structures during the agile process can result in a data model that is less than ideal and is not optimised, leading to rework.  

At Data Agility, we specialise in this work and bring years of experience and expertise to your data analytics project so that you can get the best possible outcomes. 

Supporting You

Help building a data platform.

If you would like assistance to develop your own data platform, contact us today for a no obligation discussion.


Submit a Comment

Your email address will not be published. Required fields are marked *

Most popular insights.

Effective use of technology and data in sports

Effective use of technology and data in sports

Sports across codes and the world expect decision makers, like referees, to make perfect decisions with the tech and data they have at their fingertips. The passion of the sports fan and the desire for the ref to make the correct ‘call’ is not going anywhere. So is technology to blame? Is the data inaccurate? Or is human error unavoidable? Let’s explore.

Data Lifecyle Management eBook

Data Lifecyle Management eBook

This eBook shows you the best practices on data lifecycle management; collect the data you need, store it securely while you need it, dispose of it when it is no longer needed.

The Healthcare CFO Guide

The Healthcare CFO Guide

Maximise the benefits of your health information management team. Learn how to unlock operational efficiencies and extract more value out of budgets and resources.