Is Your Data Strategy Built Like Fresh Lasagna?

Low-quality Ingredients Can Ruin Dinner – and Predictive Models

Most businesses know these days that they need data to succeed. They have come to understand that quality data is critical to a successful data-driven marketing strategy. Otherwise, it’s nearly impossible to gain insights into how your business is doing, who your customers are or what might happen next. But where businesses struggle is in how to use their data effectively.

Recognizing the importance of data is one thing. Being committed to data maintenance and having a strong data governance in place is another matter.

Being an avid amateur chef, and an even more avid professional data strategist, I started pondering the relationship between great-tasting food and quality data. Any good cook will tell you that the secret to turning a recipe from good to great is the quality of the ingredients – and quality in cooking is often synonymous with freshness. Unfortunately, keeping their data fresh is where many businesses stumble when it comes to their data management strategy.

After launching promising data initiatives (for example, building a new intelligence platform), many businesses find themselves a few months later wondering why they’re not getting the value they had hoped for – or were promised. It comes back to ingredients, and the main ingredient for business success is data that is timely (quality, fresh ingredients) and relevant (the right ingredients).

Great Ingredients Make Anything Great

I love lasagna. So indulge my lasagna analogy as I explain my thoughts on the importance of a data management initiative.

There’s a wonderful Italian deli in my neighborhood. I feel so much potential (and mouth-watering hunger) when I visit, and I’m surrounded by its fantastic ingredients. If I go to that store today and buy handmade sheets of pasta, fresh ricotta cheese, vine-ripened tomatoes, garlic, freshly hatched eggs, locally sourced ground beef, spinach, just-picked herbs, and so forth, I could come home and make a great dinner to serve tonight. I’d plan on using my new cooktop and cookware. I’d even break out my best flatware and fine china for serving. It would be a dinner worthy of my dearest friends. What a fantastic evening!

But what if I went to the store today, got all those ingredients, and then left them sitting on the counter for six weeks? After six weeks, if the smell hadn’t already driven me from my house, I might still be able to assemble a lasagna, but it wouldn’t be edible. My friends wouldn’t be too happy with me if I tried to serve them six-week-old, unrefrigerated beef.

The fact that I once had great ingredients no longer matters, and the fanciest of dishes and serving utensils doesn’t make this dinner any more palatable.

Same ingredients, same tools, same recipe, but dramatically different results.

Stale Data Is Like Stale Food

The issue, of course, is that my ingredients aren’t good anymore. They’ve turned rancid and gone stale. And just like fine foods, data can go stale, too. Age is just one of a number of factors that can negatively affect data quality. Old or incorrect name and address data impacts marketing efforts. Incomplete data about customers’ industries impacts sales territory planning. Poorly managed transaction data can lead to wildly inaccurate predictive models – which can lead to bad forecasts, investor disappointment, loss of market reputation, and worse.

How much worse? As recently as June 2020, digital payments company Wirecard filed for insolvency and its CEO was arrested after auditors were unable to account for more than $2 billion. The company denies any wrongdoing. Could a potential explanation be an underlying data issue? Perhaps this could explain why, in 2017, Wirecard was moving money for a company that ceased to exist in 2012.

Consistent Definitions Count

Data is shared throughout an organization but often ends up siloed in department-specific applications and platforms. A master data strategy is needed to bring order and interoperability so the data can flow from system to system.

If I were to bake a loaf of bread to accompany my lasagna, it would be important to know whether the recipe was in imperial units or metric. When a recipe is clearly labeled (whether it uses cups or liters), the expectation is that anyone using that recipe, whatever their background and wherever they live, should expect the same results. Without clear labels, most baking efforts will turn out inedible.

Similarly, data governance provides a common understanding of data throughout the organization. One could look back to 1999, when NASA lost its Mars Orbiter. The reason? One group of engineers defined distance using imperial units of measurement and another was using the metric system. The numbers may have all been right, but because they never aligned on the units of measurements, the data wasn’t any good across systems.

In the end, failures can boil down to trying to use bad ingredients and expecting a satisfactory result.

Quality Data Drives Data-Driven Marketing

Data-driven marketing is dependent on quality data, and quality data is dependent on sound data management. Not having a data management strategy in place results in missed opportunities, lost revenue, and unrecognized potential in technology investments.