10 navigation pointers for your analytics/machine learning journey.


The digital transformation roadmap can seem complicated because, well, it often is. The following tips can help you arrive at the solutions you need to choose the right platform and partners to help accelerate your success far into the future. 

1. Data is king. Treat it accordingly.

Data drives the insights. No analytics journey will be successful without it. This means you’ll need to find a partner who can clean, prepare and enrich your data – a process that can involve reformatting, detecting outliers, and preparing it for model consumption.

2. Commitment-phobes need not apply.

The digital transformation requires engagement. There’s no such thing as a part-time analytics journey. And no experiment is successful as an afterthought. 

“Productivity is never an accident. It is always the result of a commitment to excellence, intelligent planning, and focused effort.” – Paul J. Meyer, Author & Entrepreneur 

3. The IT/OT gap is very real. So is bridge building.

New software deployment and data access requires information technology team engagement. Yet, your operational technology folks are the ones with plant level insights. To be successful, you’ll need a liaison to build bridges. Ops isn’t used to buying and deploying software - if this is your first time doing this you need IT and it’s imperative to understand the implications from a security perspective as well as interoperability. 

Suggestion: Gain credibility with IT by asking their advice from the beginning.

4. Set a vision. Be SMART about it.

A vision for what you want to accomplish isn’t worth much if you can’t execute properly. Goals must be:






Yes, we’ve all seen this acronym before. But it remains a great way to set and manage expectations for both your engagement and leadership teams. Just picture it: you’re halfway through the first project of your analytics journey. Leadership asks you for an “economic value assessment.” And you’re ready with a solid, compelling answer – all because you’ve ensured that the success criteria were measurable from the start.

“If you don’t know where you are going, you will probably end up somewhere else.” – Laurence Johnston Peter, Author & Educator

5. Share the map. Communicate up, down and across.

It’s surprising how many analytics journeys start with no destination in mind or reason for choosing their destination. Make sure your ops team knows ‘why’ you’re all going on this journey. Then, communicate to leadership ‘why’ you’re doing this and how it will make their lives better. Without mutual benefit between all subsets of the organization, buy-in will be difficult when you get to the scaling step.

6. Share purpose. Celebrate milestones. Boost morale.

When you’re exercising, dieting and losing weight, you need to see progress. Same goes for a digital transformation. Analytics isn’t always about the end solution, a lot of times it’s about the insights you can gather in the process. Make sure to celebrate every small success along the way.

Suggestion: Celebrate your big milestones in style. Hold events focused on the success of the team. Be sure to call out specific top-performers to reinforce the culture of transformation.

7. Ace the great POC debate.

Partners of choice. How many do you need? Platforms or tools? OEMs or pure play SaaS providers? To figure all this out, begin with a well-defined scope. Establish your desired outcome. Determine how potential POCs pair with your overall vision. And, finally, decide on a level of transparency between companies if you choose multiple POCs.

8. Pick platforms, not tools.

More often than not, the beginning phase of an analytics journey is more about the knowledge you’ll gain along the way. So don’t begin it with a specific tool. Instead, partner with a group whose capability and expertise are broad enough to educate you and your organization along the way even if your experiment leads to failure…which, well, it happens. In fact, prepare for it. 

”Success is going from failure to failure without loss of enthusiasm.” – Winston Churchill, Author & Former British Prime Minister 

9. When results are delivered, be ready to pour fuel on the fire.

Be intellectually honest about results. Don’t exaggerate the impact of machine learning. Sometimes it’s just not a fit. But when a project succeeds? Be ready to churn, burn and fire it up to scale. Think through how to do this before you succeed. Your ongoing evaluation will allow you to pivot proactively and know what next steps or initiatives look like.

10. Comprehensively capture your economic Impact. And tout it.

Build economic models along the way. There are many reasons your organization should go on their own analytics road trip, but you need to be ready to rationalize the money you’re spending to put “gas in the tank.” along the way. Ask yourself at each mile marker; what is my company getting out of this?

Value is in the eye of the beholder. And for you, the most important beholder is leadership. That’s why it’s critical to use economic thinking, and capture its results, at each milestone of the journey toward solving previously un-solvable challenges. Without it, we will be missing out on the key to leadership’s heart. 

EFT Staff
About EFT Staff

This blog post was sourced and written through a collaboration of various EFT staff members.

Follow Staff on LinkedIn

10 navigation pointers for your analytics/machine learning journey.

Thanks for your interest.