Agile Analytics or Lean Analytics

Agile analytics or lean analytics… or both?

Before go into agile analytics or lean analytics, let us first understand the differences between Agile and Lean. Most people assume they are the same and hence use the two terms interchangeably.

Both are effective principles/mindsets in their own rights but that doesn’t make them the same. A car and a plane will get you to a destination, both are effective means of transport, but they are not the same.

Lean was designed to improve operations and eliminate waste within a manufacturing system. It also takes into account waste created through overburden and irregularity in workloads.

Agile refers to a group of methodologies or principles that focus on iterative software development. Requirements and solutions evolve through collaborative effort. Continuous improvement, flexibility to change, early delivery are just some of the philosophies of Agile thinking.

Could we apply one or both of the above disciplines to analytics? If we break down analytics to a “product” and a “service” we could take out parts of each principle and apply them to analytics.

Lean Analytics – Improving operations and eliminating waste.

We can apply Lean to the “physical” outputs of an analytics team. A few examples of “physical” outputs include:

• reports
• presentations
• briefs
• ad-hoc requests

By operationally improving the way the teams produce reports and presentations, we could dramatically optimise our processes; create PowerPoint templates, automate reports, reuse queries or even train stakeholders to pull basic reports.

To eliminate waste audit the requests and reports that go out; those that are not adding any value should be cut. In the context of Lean analytics, the waste you are trying to save is time.

Agile Analytics

Agile analytics can be broken into the two most popular Agile subsets, Scrum and Kanban*.

Of the two, Kanban is more likely to be used in the field of analytics. If the previous section considered the “outputs” of an analytics team as Lean analytics, then the “processes” within the team can be considered as Agile analytics.

Kanban uses a pull system which allows team members to pull new tasks once the previous task has been completed. The tasks in the backlog are prioritised with stakeholders and timelines are agreed when the work begins.

This process is effective in an organisation with a centralised analytics team whereby requests are coming in from all directions. A transparent Kanban board would mean trade offs are more likely to be made between stakeholders and not between analysts.

Hopefully, the ideas above demonstrate how analytics can be approached in both a Lean and Agile way. By reducing the waste from the output and effectively managing the processes we can take a truly harmonised view of analytics.

*A great comparison between Scrum and Kanban can be found on leankit or we have some great videos on reporting and making data your product on our videos page.