How to Build a Strong Data Model

With ClicData, you can prepare your data by adding the calculations that you need on your final dashboards. You can perform different operations like fusions, merges and aggregations, and you can create new calculated columns, update existing columns, clean your data and much more.

The advantage of performing data manipulation with ClicData is the automation that comes afterwards. After you build the data model once, all of the transformations and other views are updated automatically once one of the data sources is updated.

Because we provide visual capabilities of data manipulation, not much technical knowledge is required. However, there are a couple of items that BI developers need to take into consideration while building the data model on ClicData.

Before Starting

Don’t jump into importing the data sets and creating different data layers. You can take a pen and paper and draw your data model, but it is preferable to use a tool like Visio to design your flow diagrams. You can use the diagrams while documenting the project later.

This way, you will avoid unnecessarily referencing a data set multiple times in the data model or creating circular references that will slow down or even crash your process.

Challenge: BI developers need to fully understand the business model before designing it. They need to sit with the business/project manager and translate their final requirements into data flow diagrams.

Even if this step is done and we have successfully designed a clean data flow, there are cases when the development continues after the project delivery. Like Heraclitus once said long ago, “The only thing that is constant is change.”

Changing Requirements

Continue optimizing the data model when new requirements come. Try to insert the change in the right place in the flow. Don’t build new layers of data on top of the existing ones if you can do better!

Challenge: Sure, it might be time-consuming to review the flow, decide where to add the change and then re-adjust the data model. It might be easier or faster just to add another view with the change and rebind your dashboards. However, this option might impact the dashboard’s performance.

And finally, keep documenting the data model! This is the only way that you will be able to optimize and maintain a healthy data flow for your clients.

In case of further questions, do not hesitate to open a support ticket at and we will be happy to help you!

Share this on