Everyone’s getting excited about data science. It’s all very well building analytical models and using artificial intelligence and machine learning, but what’s the point if you can’t use it?

Data analytics is not just about building a technical model that no one understands. It’s about working through the whole process; making sure the model answers the business question and that any analyst or data scientist will be able to use it. For example, creating documentation that anyone could take anywhere and use to rebuild the model.

Once implemented, the next, equally important, step is to monitor the model and ensure its working to best answer your business question. So, we set KPIs such as tracking model usage or are they driving incremental revenue?

Designing and building the data-model is one part of the puzzle

After deciding which operating model to adopt, you must be able to bring it to life across the business. How?

Tap into your analytical communities. If you’ve got different teams of analysts across the business, get them to meet up and share ideas and feedback on what’s worked well. It’s an easy way to solve problems of data misinterpretation and an excellent way to engage with users. 

3 Top tips for embedding a data model

  1. Enforce documentation and define common standards and tools by which documents are created and maintained.
  2. Make sharing knowledge the right thing to do. For success, you want everyone in the organisation to be able to see what methods you have applied and then be able to access them to use in their work.
  3. Support analytics execution. You need to help embed the model into the business, not leave it all up to the analysts.

By following these tips, you’ll have consistent tools across the organisation, with shared knowledge and definitions, and everyone will follow the same process in terms of implementation.

One of our global clients has taken this analytical approach, and by starting to share their models, and by having processes in place, it’s working for them. In turn, by tracking and measuring the KPIs, we understand how the models are working and where we can make improvements.

At Merkle Aquila, we can help you develop the right processes and the right analytical communities within your company for data-driven decision making. Just ask us how.

Back to News