Data is the new oil – crude and dirty

What oil and canaries in mines tell us about digitalisation

by | Feb 29, 2024 | General

Data is the new oil

People often say that data is the new oil. Except it is crude oil. Almost useless in its original form.

The useful stuff is petrol, diesel, jet fuel, LPG and more. But out of the ground, it’s a dirty, smelly black liquid. Valuable but not useable without expending energy and expertise.

Back in my Big 4 consulting days, I accompanied a private equity team on a tour around an oil refinery they were thinking of buying. Over breakfast, they said, “We’ve looked at the numbers and we don’t understand why they don’t make more aviation fuel.”

While munching on a pastry, I attempted to think of a way to explain an inconvenient truth: crude oil needs refining to separate it into useful products and the ability to make more of a product was dictated by the (expensive) capital equipment and the properties of the crude oil purchased. And however much they wanted to make more aviation fuel, that was not going to be possible without investment.

To paraphrase, if it were easy, the people running the refinery would have done it already.

Like those private equity guys looking at oil products, the opportunities to use data and generate profit are often easy to identify. But in the same way, if it were easy, your team would have done it already.

Just like our oil refinery, data needs upfront investment to turn the crude raw information into actionable insights.

It’s a useful metaphor to help leaders understand why transformation teams often have to invest in platforms, data standards and data governance before they can deliver the riches promised by “Big Data.”

If you’re thinking that I’m suggesting that data projects should be fast-tracked to the “Too Hard Bucket” — not at all. They are one of the best places for companies to start their digital transformation journey. Understanding how to make more profit from the core business activities your company is already doing is a great entry point into digitalisation.

Photo by Peter Olexa on Unsplash

 

The Good News

The good news is that all organisations are already doing data analytics in some way. However, that is usually in the form of massive spreadsheets in Excel, used to generate one-off charts that are copied and pasted (as a picture) into Powerpoint.

More good news, improving data quality is a pre-requisite for more complex digital opportunities. Core ERP functions can often carry on, even with poorly maintained master data. However, supply chain software and other forms of decision intelligence can often be a “canary in the mine” — an early death from toxic gas or in this metaphor contaminated data.

Investing in data improves the probability of success for those IT investments and reduces the pain of implementation.

And even more good news, CFOs clutching the purse strings for IT spend love seeing data and insights on how to improve business profitability.

The Pitfalls

  • Quick Wins: Yes, you can get quick wins on data projects. Perfect to prove the value of the investment in your data “refinery” — but those quick wins are often not scaleable/ secure/ automated /re-useable. So, be careful not to give the impression that your quick dashboard that’s held together with staples is enterprise-ready.
  • Beware the big in Big Data: It might be nice to pull in the entire history of your business. But that’s a lengthy data cleansing and migration activity that will take months. You might be better off starting today and collecting new data. By the time those months have elapsed, you’ll have a fresh trusted data set that’s good enough for a lot of your decision-making.
  • Download to Excel: One of the more depressing meetings of my career was with an IT team promoting their data analytics project, who said that “download to Excel” was the most requested feature from their users. If you want one source of the truth, then beware of your user base creatively implementing workarounds that avoid using your tools and also avoid awkward conversations where they might have to tell you why they don’t like your tools.

So what, now what?

Before rushing off to chase the latest AI fad, start with what data you already have and how you are using it to help the customer experience. Process data mining can be a good way of seeing where transactions fail, leading to customer frustration. Supply chain inventory is another classic pain point – theory is well understood but most companies still have too much of the wrong stuff.

Do you agree?