Data governance and data intelligence is not just about content. It is not just about having the information. It is about having processes to help keep that information up to date and a knowledge base to access that information. This blog post will discuss the human curation activity around your data, points of engagement, and the content creation loop.
In this blog post we will discuss these questions:
- How do people request things?
- How do people find things?
- How do people search for things?
- What do people do if they can't find information around their data?
- How do people engage with the data stewards who have the answers to their requests?
- How does information get captured?
We feel that the best approach to data governance is a just-in-time, customer service, help desk approach. Organizations need to focus on the current and most critical, and not the past. Critical reports are a good starting point for creating a bunch of other data-related content. Do not do report specifications on your 1000 reports. Focus on the new ones or the current ones that people have questions about. The energy and the resources are put on requests, not the past. Rather than spinning your energy to document all your reports before you roll data governance out, just document a couple of the critical reports. See this blog post for more information on this approach.
There is a great deal of data governance-related content: business glossary definitions. information about data systems, data models, information on reports and integrations, reference data (code values, lookup tables, valid values), data quality rules, and data-related polices and monitoring those policy attributes such as data access, privacy, security sharing, retention, and backups. There is a significant amount of content which requires a great deal of effort. That is why we feel the content creation loop is a great way to create usable data governance-related content. The idea is to put this data governance information in some organized way that is easily accessible, findable, and maintainable over time. And a tool like the Data Cookbook is a great help.
One of the most critical things you can do in a documented report is to associate the data getting presented on that report with business glossary terms. For example, in a report there are columns for student status and GPA. Are these defined in our business glossary? If so, does that definition match what we are looking for? If not available, you might then go create that glossary definition. In documenting a report then you might end up with creating 5 to 10 different glossary definitions.
Move forward. Documenting an old report is reverse engineering and will take longer than if done when the report was created. It is a much smaller effort to capture information as you build new content than when you go back in time. You do not want this information around your report catalog or glossary definitions to be just a binder on a shelf that nobody ever looks at. It doesn't add any value. You have wasted time. You want your data-related content to be used.
I am looking at this report and trying to understand if this report is useful. Is this a good report? How is a particular value calculated? Or what does this field mean? Data governance content could answer these questions. By having this information, you gain trust and knowledge in the data. And with this information you make better decisions. That is the point of having this information so people can have more trust or have more knowledge or more reuse.
If data users are not engaging with the data-related content, they are not getting any value out of it. It is not about the content created. It is how many people are engaging and using the content. This is where we talk about this concept of points of engagement, the searching and discovery for this and how that gets started. The points of engagement of searching and discovery can lead to requests. The assigned data steward creates content that needs to be reviewed. There is an ongoing assessment or approval or review of this content.
I am looking at a report/dashboard and I have a question on it. It would be nice if I could link over to information or connect to the information on the glossary terms. And can I make a request for information or curation right inside the report/dashboard that I am viewing. This is an as needed request that then can lead to a data steward defining a term or curating the report.
Often when you are talking about faculty information there are multiple stakeholders. You might have HR people who have a vested interest in how things are defined such as faculty load. And you also have financial people who are interested in that because it might impact grants and payments. And you also have student data involving faculty which means that the registrar is involved. You have a bunch of people that need to agree on what this definition means or have a slightly different context when we're using that definition. Often the content needs a review and goes to others to review and approve.
The assumption here is that your organization has limited resources and time to curate and create all of this content. We have never met an organization that has a bunch of subject matter experts (SME) sitting around with nothing to do.
How do you prioritize the effort and still be sure that you are supporting the necessary engagement? If someone has a question about a report, how do you make sure that they can get the answer they need without understanding that you have limited resources? This is where this customer service/just in time/help desk approach comes into play. You want to prioritize the content creation based on requests. Your goal is to provide data consumers with the information they need to understand the report or build the report. And this helps you to leverage expertise when you need it. Like an IT Helpdesk. If you have a question, then you look in the knowledge base. If the answer is found, you take care of the issue yourself. Self-service is good. If the answer is not found, you make a request, and the appropriate data steward is assigned to solve your issue. Hopefully the resolution will be put into the knowledge base for the next person that has that same issue.
In the content creation loop, people come in, search for discovery, and hopefully find what they are looking for, and they leave the loop. But what happens if they have questions or do not find what they need. They continue with the content creation loop.
You want to implement processes as opposed to waiting for data governance content perfection. Make it easy on your data users. Have them access the report specifications (purpose, description, owner, etc.) along with the business glossary right in the report. And have them make requests regarding that report right in the report. Maybe that data request accesses your ticketing system that is already used in your organization. The request should be put into a workflow that sends the request to the appropriate data steward or, if none defined, a support triage person.
Hope that this blog post was beneficial to you and your organization in creating your data governance related content.
IData has a solution, the Data Cookbook, that can aid the employees and the organization in its data governance and data intelligence efforts including data requests. IData also has experts that can assist with data governance, reporting, integration, and other technology services on an as needed basis. Feel free to contact us and let us know how we can assist.
(image credit ContentCreationLoop Diagram #1246)
