Analytics is still in a phase in many organizations where selling it internally to the stakeholders is the biggest challenge.Creating analytics is a cost- resource intensive investment for enterprises and evangelizing the trust in data driven thinking and optimizing the opportunity cost due to delay in adoption is the most crucial problem to solve.
All the frameworks which talk about the Analytics mix, various technology & tools stack and maturity roadmaps, are confined to an on-boarding and transformational objective view. There is still a big gap in terms of adopting analytics in the right spirit, trust in the value add and incorporating a culture of data driven thinking.
With many organizations incubating analytics as an intrapreneurial setup and driving it in the startup mode, they encounter some of the common problems which a start up faces when trying to scale up. A lean thinking is required towards driving adoption of analytics , than working top down as a big bang transformational initiative.
Here is a thought process around how you can think lean while setting up an analytics hub and driving the adoption of the same:
Live in Chaos but think of processes, Lean processes
Set up a way for experimentation. Fail fast and cheap but have a feedback loop for reasons of failure and a fallback mechanism to maintain the continuity. Pick the low cost projects which need low cost of creation and moderate value in consumption. Do not pick Optimization as your first project. It is more viable to do an opportunity sizing exercise and tell a business stakeholder what they are missing out on. For example, Tell your digital marketing guy that if he is only looking at the problems around conversion optimization, he is actually only focusing on 3-4%(mean conversion rate in e-commerce) and missing out on a huge amount of data and opportunity with 97% of other potential/owned customers. Learn from each proof of concept which fails and set a process around it.
Usability is more important than feasibility
Do not set your objectives to achieve things through analytics which cannot be achieved otherwise. Take up a project based on “ whether there is a need and usability” rather than “whether it can be done or not”. Think from a consumption point of view. Do not just build predictive models because they make you more mature soon as an analytics team, descriptive projects most of the times have more value as they help drive frequent decisions and more connection with stakeholders. Use the concierge approach ( setting up a manual kiosk before you build the automatic one). Learn about the usability, collect feedback and redefine your maturity roadmap.
Get your MVP fast
So what is your Minimum Viable Product in analytics journey? A prototype, A dashboard, A formulae? Think of it, find it soon, build it faster. Collect plenty of feedback, create advocates of your MVP. Let your stakeholders play with the first analytics tool you give them. Imagine, your end goal is self –service analytics, so serve it raw from the start. Choose one metric you want to move for the target audience. Is it automation? Is it more data? Is it cool visualization? Or is it path-breaking insights? Stick to it and deliver.
The key is to Plan big, Start small, Iterate faster.
