In creating my 90 day plan I leant on the overall mission of the data function at Transpower which can be summarised succinctly as “Enhancing decision making”. Then taking assumptions from previous conversations, I feel I have a solid understanding of where the organisation wants to be in 3 years time.

  • Self servicing

  • Automated tests, alerting, pipelines and deployments

  • Predictive analytics

  • Data as a service

  • Integration of DevOps as a practice

The challenge: In 90 days, in a new role, in a new team, how can we deliver success?

Building foundations while delivering milestone achievements

Taking predictive analytics as an example:

  • To have predictive analytics we need to have a prediction model serving production data
    • In order to have that we need to have the framework for production deployment
      • In order to have that we need to have consistent data and tooling to unlock the ability of individuals build out these frameworks

So a foundation building block is providing a consistent working environment that is industry standard, unblocking the analytics function. I propose this is a juypter server. It does not give us predictive analytics but a milestone achievement along the path to predictive analytics.

Prioritization framework

Considering the above, how should we define what can and should be achieved by this role in the 90 day time frame? Leveraging a simple Effort vs Impact quadrant the 90 plan focuses on quick wins and laying foundations for strategic innovation.

My picture

90 days into 6 cycles

Splitting out the 90 days into three key focus areas:

  • GIT and workflows
  • Data and tooling
  • Stakeholder engagement

We are aiming to achieve milestones of

  • Sprint cycles
  • Deployment and production tests
  • Code review
  • Jupyter application for analytics
  • Data pipelines alerting and monitoring

On to the presentation