Solving the Right Data Problem. Finally.
Par Christal Bemont
Making the case for universal data health standards
It’s a big data world; we’re just living in it
In late 2020, a CEO at an American bank revealed the thinking that’s becoming common in many businesses these days. “We’re a 103-year-old bank,” their CEO told me. “We’re doing everything on spreadsheets. But we are trying to become a highly profitable, digital-first bank that anticipates financial needs and empowers our clients with frictionless experiences. We need to become a data company.”
Companies in every industry are utterly dependent on their data. Retailers do more than sell goods in stores; their success depends on collecting, analyzing, and sharing data on consumer behavior, preferences, and actions. Financial services companies can mine the wealth of data they obtain on trades for insights and sell it as a source of intelligence, creating additional revenue streams. Healthcare providers no longer just treat their patients’ illnesses and injuries — rather, they collect and analyze data to treat the patient before their condition becomes acute.
Every business is now in the data business. Even before COVID hit, many organizations had already begun that journey. When we found ourselves face-to-face with a global pandemic, the need to transform became more urgent.
Too many trees and not enough forest
But even though companies want and need to become data-driven, they’re not that successful at doing it yet. Surveys show that nearly 70% of companies report that they have not created a data-driven organization, and over half are not yet treating data as a business asset. Companies know that the path to the future depends on using data. So why is using data so challenging?
For decades, managing and using data for analysis was focused on the mechanics of the process: collecting data, cleaning it, storing it, and cataloging it. It turns out this was the wrong problem to solve. The preoccupation with the mechanics of data management created some enormous challenges:
- There’s no connection between the people who prepare data and those who make the decisions or assess the state of the business.
- There’s no way for the people and systems on the front lines to easily validate that the data fueling day-to-day business is reliable or risk-free.
- The piecemeal approach to managing, integrating, and storing data has created silos. It is expensive and difficult to manage, and it also creates dark data where analysis cannot penetrate.
- For the most part, software and platforms for moving, collecting, preparing, and storing data is not helping companies gain a deeper understanding of the data they have or helping drive better data outcomes.
More and more companies finally realize that this piecemeal approach doesn’t work. It’s not enough to simply collect, move, and prepare data more efficiently.
Data management is focused on the wrong things
The data management market, estimated to be worth about $130 billion, has attracted a lot of attention recently, and rightly so. These solutions have become highly effective at moving and storing more corporate data. But in our view, for many companies, this efficiency is creating as much of a risk as it is a reward.
While capturing and storing data was a problem, it was never the problem and certainly never the end game. The old message to companies was that they should collect as much data as possible and figure out how to use it later. Well, later has arrived and many companies are ill-equipped to move from being data-saturated to being data-driven.
When data is moved and stored without any other considerations, it essentially becomes a digital landfill of corporate information. Instead of solving problems, data is making it harder to sort through the chaos. Companies are drowning in their data.
We have a very scary situation on our hands. A huge number of companies, all of which rely on data to stay in business and now also believe they need to become a data company, still haven’t addressed some of the most basic components of the data equation. They still don’t know what data they have, where it is, or who is using it, and, critically, they have absolutely no way to measure its health.
Ask any company how they measure the health of their business, and they will list metrics backed by the data they run their business on. Besides employees, data is the single most important asset any company has, yet it’s the least understood or measured. Data runs the world, and yet it’s the one thing we understand the least.
Getting a pulse on corporate data health
Data management can’t be a simple pass-through as it typically is today. It needs to be an active and intentional system that increases an organization’s understanding of its data — its reliability, risk, and opportunity to provide value for the business. You should have visibility and clarity in your data. The solutions you use to manage data should provide the knowledge that will help make your organization smarter, more agile, and more efficient while avoiding risk.
This may sound impossible. Is it really plausible to understand corporate data — what it is, where it’s located, whether it’s accurate, who’s touched it, and how it’s distributed? Is it really possible to get a measurable and quantifiable view into the most valuable, yet today the most intangible, business asset?
Yes. Every business can do it through a concept called data health.
Data health is Talend’s vision for a holistic system of preventative measures, effective treatments, and a supportive culture to actively manage the well-being of corporate information. The system would include monitoring and reporting tools for helping organizations understand and communicate, in a quantifiable way, the overall reliability, risk, and return of an asset that’s essential to their viability.
In the future, the aim is for data health solutions to help create a universal set of metrics to evaluate the health of corporate data and establish it as an essential indicator of the overall strength of a business.
For too long we’ve treated data as simple, concrete units: cells on a spreadsheet, fields in a database — passive digital objects waiting for an analyst. But that’s no longer a sufficient model. Data is a complex, constantly changing organism. New inputs flow in and out, updated by users and transformed by shifting contexts. Those inputs and actions both provide an opportunity to learn about and change the value of the data itself. To truly understand what our data means, we need a more responsible, holistic view of that data.
Data is complex, and every organization has its unique requirements, regulations, and risk tolerance. This is why we imagine data health as a journey. Just like human health, data health would be different for companies of every age, life stage, and maturity level.
Our initial framework imagines three areas for companies to focus on as they begin the journey to establish data health:
- Preventative measures — preemptively identifying and resolving data challenges
- Effective treatments — systematically improving data reliability and reducing risk
- Supportive culture — establishing an organizational discipline around data care and maintenance
Overseeing all of these focus areas is a comprehensive system of proactive monitoring and reporting to indicate success in achieving your data’s well-being. The combination of technologies and cultural practices that form this system will be unique to every organization, but the standards applied will be universal.
A vision for a better future
We believe there will come a time in the not-too-distant future where we’ll look back and wonder how we ever functioned in business or as a broader society, without a quantifiable way to measure the reliability, risk, and return of an asset so critical to our success. The risks of compromised data — and opportunities for innovation — are just too great.
In 2019, Vyaire Medical, a global respiratory care company, kicked off a digital transformation initiative. Like many companies, their data infrastructure was a patchwork of one-off solutions and inefficient structures, including 12 enterprise planning systems, meaning that employees had to collect data on parts location, production, factory efficiency, and more from wherever they could. Decision-makers often received conflicting data depending on its source. This created confusion and made it difficult to make the fast, data-driven decisions necessary for the company to continue to thrive. There were so many design elements, top to bottom, that were never built to scale to the numbers we needed to scale to,” said Ed Rybicki, Vyaire’s global chief information officer. “So we really needed to rethink the whole thing.”
“Rethinking the whole thing” meant making some key infrastructure decisions — moving to a cloud infrastructure rather than remaining on-prem, building a centralized data repository that anyone in the company could access, and instituting data quality standards to ensure the data in the company-wide data lake is clean, accurate, and available in real-time. In short, they made the call to ensure healthy data to anyone who needed it for analysis and business decision-making.
Vyaire’s wholesale transformation project proved extremely prescient; in 2020, COVID emerged, which meant there was more demand for their ventilators than ever before. “We had to replicate a highly customized manufacturing process for this line of ventilators,” said Rybicki. “It was probably 20 times beyond what had ever been done before, all in six or seven months. At the end of the day, we knew that if we weren’t able to scale up, it would mean that people who needed ventilators might not get them. We were able to scale up these old systems—and help save lives.”
For Vyaire, data health was no longer optional. It’s not something they could live without or could plan for in a year. Ensuring data health was what the business needed as soon as they could deliver it. Vyaire Medical’s experience is not a fluke. Because they made deliberate investments in the health of their data, they had complete clarity into their entire operation from the factory floor to the boardroom. They knew exactly what the capacity of each cog in their supply chain could be and what it should be. And that meant that they were able to answer the call of a lifetime — to meet the moment and produce the equipment that patients around the world desperately needed. Thanks to their data, Vyaire optimized their business and overcame the obstacles a global pandemic presented. Data health offers that exact type of clarity that you could have into your business.
No one should ever have to make decisions on information they can’t find, see, or understand. Data should not be a black box. The ultimate goal of creating a data health practice is to establish confidence and total visibility into your data, and therefore a real quantification of value. You should understand how data is working in every aspect of your business. You should be able to identify what data is doing for you. You should be able to establish an ROI for all your data investments. We believe that once you establish your company’s baselines for data health, it will become as indispensable as Google Maps — you won’t be able to imagine life without it.