Data is touted as one of the most important sources of transformation in our world today. In fact, The Economist termed data “the most valuable asset in the world.” And the amount of data out there is only increasing. In 2016, an IBM Watson analysis estimated we were creating 2.5 quintillion bytes of data every single day.
Pretty staggering information, but most organizations are not converting this data wealth into revenue. VentureBeat AI reports 87% of data science projects never make it into production. Gartner shares 80% of analytics insights will not deliver business outcomes through 2022.
MarketWatch recently published three common flaws in data governance and how to solve them, citing a new data blog series by Credera executives, Vincent Yates, Chief Data Officer, and Jason Goth, Chief Technology Officer.
1. Starting from the Wrong Places
Simon Sinek’s famous “start with why” TEDTalk applies here to data strategy. Many organizations don’t truly understand the reason for a new artificial intelligence (AI) or machine learning (ML) project. MarketWatch explains that to avoid this problem you should “start from your business’s need.”
Jason Goth and Vincent Yates explain it this way, “The key to actual success is to remember that the data is not the end but rather the means to the end. Thus, the successful approach starts with the business. What business decision, action, or intervention is the data driving?”
2. Bad Data
Unfortunately, an Experian report found that 98% of companies believe their data isn’t accurate. As many like to say, “garbage in, garbage out.” If the information that is taken into a data model or ML algorithm isn’t accurate, it’s no wonder the applications of the data are not driving real business value.
MarketWatch shares the reasons behind bad data. “The sources of poor data quality are many. When companies begin from the wrong places, the data often becomes irrelevant. Without consistent collection policies, error takes root. In many cases, information is housed in departmental silos that don’t communicate much with each other — or, if they are on speaking terms, offer conflicting evidence. Absent a structured way to store and access this information, data-driven workflows ebb.”
They share that fixing the data can oftentimes be simple, and breaking down organizational silos is one of the best places to start.
3. Failure to Translate Data into Decisions
Tying back to Credera’s advice to “remember that data is not the end, but the means to the end,” MarketWatch explains that one of the main reasons data projects fail is because you have a people problem. People may not trust the numbers or others may be pulled away from the data by political whims.
MarketWatch shares that it takes time to create a data-driven culture, but the rewards are well worth it.
Continue to read the full article here.
Creating a Data-Driven Organization
When taking the next step in your data maturity journey, organizations must see it as just that, a journey. When problems arise, try to focus on the source of the problem, not just the symptoms. And don’t forget, treat your data as a tool for an ultimate goal instead of the goal itself.
Read more of Vincent Yates’ and Jason Goth’s blog series on data strategy below:
Data May Be the New Oil, But Few Know How to Refine It: Data Strategy Series Part 1
Not Your Father’s Governance Model: Data Strategy Series Part 3
Mind the Gap—It’s Not AI/ML Unless It’s in Production: Data Strategy Series Part 4