Agile Analytics. How to Get Started?
From Alteryx Inspire in San Diego to Tableau Conference on Tour in London, we've been following a stellar data trail this year. With more yet to come, let's discuss agile analytics, a topic that's been growing in importance when we talk about data.
What is agile analytics?
The essence of agile analytics is on speed to value by focusing on the insights and actions that can be drawn from multiple and disparate data sets. It involves rapid prototyping where you can test and learn as you go along to build up models, requirements and insights based on solid answers to your business questions.
Your business questions can be anything from how to maximise revenue, mitigate risk, minimise churn and optimise efficiencies. However, the agile approach to conducting analytics will be the same when looking to initially unlock that value from your data.
So how to get started with agile analytics?
Here are 7 points to consider:
• Think about outcomes: Since agile analytics takes away the load of traditional implementation, shift your time and effort into thinking about the business impact. Ensure you have a holistic approach to track and report on performance rigorously.
• Drive focus: Prioritise. Decide on which problems to tackle first and which areas of the business will benefit the most from this initially. How much can you sensibly chew off to start showing results?
• Don't wait around: Poor data is often used as an excuse to delay analytics projects. But you don't need perfect data to make a start. As the project moves along, there are processes to help identify the data gaps and develop robust systems for collecting and managing both the data you have and need.
• Build the right team: Domain knowledge is key for understanding the business area, whether it's translating outcomes or identifying anomalies. Make sure you get proficient data analysts and data integration experts on board. Even in building a prototype, it is crucial to get the data captured, collated, and tested correctly from an early stage.
• Select the tools that fit: The number of self-service tools that support data blending, advanced analytics and data visualisation is growing all the time. You may already have the right ones but with almost all vendors offering free trial periods, you can always test out something new.
• Collaborate with others: Partnership between the business, IT and the data analysts is key - whether it's gaining access to data, adopting new tools or evolving a prototype design to an enterprise deployment. Share outputs and findings as you go along, so there is continuous involvement in evolving requirements.
• Step by step buy-in: Build organisational momentum, celebrate small wins and find your champions! Ultimately, you want to highlight how new approaches/tools are enhancing rather than challenging existing wisdom.
I hope the above helps and do tell us how your agile analytics project goes.
New to Alteryx? Get a free trial!
New to Tableau? Get a free trial!