A data catalog can be an excellent resource for a business. An important element of modern metadata management, it can make many workflows and business processes clearer, encourage better cross-departmental collaboration, and boost productivity in several key ways.
Gretchen Burnham and Becky Lyons of First San Francisco Partners recently sat down with DATAVERSITY® to talk about some of the big questions in data catalog adoption and development, including when to get started and how to gain executive buy-in.
What Is a Data Catalog?
A data catalog is a comprehensive directory listing and describing all available enterprise data assets a business uses across its entire org chart. Lyons likened it to an old library card catalog. All that information in the data catalog points to the location of various data resources and what’s in them. “A data catalog at its heart is an inventory,” she said, “a living inventory of data assets that has to continuously be revised and refreshed.”
Organizations must regularly edit the contents of their data catalog to avoid just “collecting, collecting, collecting,” warned Burnham. The value of a catalog is in its usage, not its collection. “It has to be useful for what your organization is trying to achieve.”
A data catalog is essentially the same as a metadata repository, said Burnham, but the term “data catalog” sounds more approachable to a less tech-savvy audience. Like data catalogs, metadata repositories store metadata that is easily accessible to anyone in the business.
Benefits of Using a Data Catalog
When properly implemented and optimized, a data catalog can help people find valuable data within their organization – some of which they may not even know exists – and work together to understand and use it in a meaningful way. “A data catalog is a communication mechanism,” explained Lyons. “It’s a way to bring data to the people who are using it every single day.”
Similarly, Burnham emphasized the importance of driving alignment as a primary benefit of catalogs. “All sorts of different people, every day, work with an organization’s data,” she said. A data catalog allows everyone to get on the same page about all that data and how it relates to business goals and issues: “What does this data actually mean to all of us?”
A data catalog can create visibility and solve redundancies. It can also solve problems for business analysts. Since it is such a Swiss Army knife for business, both Lyons and Burnham recommended that organizations really figure out what they need – looking at the pain points of their enterprise and how they aim to solve them – before moving forward with catalog implementation.
Does Your Organization Need a Data Catalog?
Businesses may start to show tell-tale signs or “symptoms” of needing a data catalog, said Lyons. For instance, if you’re spending business meetings arguing about the meaning of the data and “whose numbers are right” rather than making decisions, that’s one clue that a data catalog could help.
Lyons and Burnham pointed to other scenarios indicating that a company might benefit from a data catalog: if your data analysts and data scientists are spending a long time looking for data rather than analyzing it; if much of your workforce is getting close to retirement age (or you have high employee turnover) and you risk losing valuable business knowledge; or if you’re exploring self-service analytics for business users and stakeholders. A data catalog is also essential for mergers and acquisitions.
“You’re bringing together two very different organizations, cultures, understandings, applications, data, all of that – it’s all very, very different,” said Lyons. “Data catalogs are a great way to understand the current state, as well as create the future state.”
Types of Data Catalogs
Once you’ve decided that your organization should adopt a data catalog, you’ll want to determine what type of catalog would best serve its needs. Businesses with a highly collaborative approach will benefit more from a “Wiki-style open catalog where everyone is intended to contribute,” said Burnham. In contrast, more traditional, tightly controlled businesses might want more curation and approval processes built in.
Talk to your database administrator (DBA) and data architecture team about your enterprise’s current technology stack – as well as where it is going – to ensure that you can easily scale your catalog efforts in the future. Different tools may connect with your environment better than others and offer varying levels of governance support and automation.
Next, consider who will use the data catalog and how their skill set will affect their user experience. Less tech-savvy employees, for instance, might need a user interface that is better at holding the user’s hand.
“More and more, we’re seeing business analysts – not even developers – wanting to use data catalogs,” said Burnham. “And they’re going to have a very different comfort level with a user interface than someone who knows how to write SQL for a living.”
Lastly, how will these employees be using the metadata found in a data catalog? Will they use it to write reports? Manage databases? Resolve business alignment issues? “Get specific,” advised Burnham. “Have a common list to present to all vendors so that you can see apples to apples across the different products.” This will help you make objective decisions rather than lean toward the product from the most likable sales agent.
How to Get Business Buy-In
Lyons and Burnham agreed that the key to getting buy-in is simplicity. Describe the benefits and use cases in a way decision-makers understand. When discussing the merits of metadata, for instance, speak the language of your stakeholders:
“You may be tempted to say metadata is ‘data about data,’ and while it’s technically true, your stakeholders won’t get it,” said Burnham. “So, say something about metadata being useful context about the data of your organization. ‘Data about data’ is a fun little catchphrase, but it doesn’t tell the audience enough if they’re not already data people.”
Use analogies that resonate with decision-makers, suggested Lyons. For younger audiences, she likes to compare a data catalog to a Spotify playlist: “Think about if you had that kind of list of all the reports – and information about those reports – that you look at every day,” she said. This type of description helps promote buy-in and get everyone excited.
Using another analogy, Burnham sees data catalogs as a bridge, connecting people and data:
“Data catalogs are a way for a business to talk about its data – how it’s used and what they think about it, as well as the more technical aspects like where it lives and how it’s structured – and then connect those things together.”
With executive support and plenty of careful planning, a data catalog can provide the necessary tools for organizations to make those connections and put business-critical information to use – organizing data assets, enhancing communication around data, and helping everyone get the most value out of their organization’s data.
Image used under license from Shutterstock.com