In our first post, we discussed how in our work, as in our homes, we can apply rules for bringing order to our practices and our tools not just to clean things up but to create efficiencies going forward. This second post in our two-part series will continue to apply some KonMari Method™ principles to your Domo instance to do just that.
A messy and un-governed Domo instance leads to data security problems, user frustrations that can impact further adoption, and overall maintenance nightmares. Are you having issues figuring out where to start cleaning up an instance or with data governance in general?
Let’s continue on our tidying journey…
The Last Three Rules
These sections are based on the KonMari Method rules, and they aren’t steps to perform in order. Instead, develop processes in your governance ecosystem that consider each of these rules. Rules 1-3 live here…
Rule Four: Tidy by category, not location.
In the KonMari Method, everything gets tidied according to category and in order. This helps get everything organized at once to help it stay organized later. For your Domo instance, let pages, cards, datasets, users, and groups be your categories, and begin organizing them in order (see below). Even though the methods and processes will be different for each category, aim to outline and execute a comprehensive plan for each category so that all users can adhere to a set of standards that are easy to abide by.
Next, perform security audits to determine who has access to different cards. Provide access based on what adds value, and keep in mind that any process you introduce to the ecosystem should be evaluated by a governance perspective. For example, if you aim to enable security that puts up a firewall to restrict access, you’ll need to determine (and communicate) what IP can access the instance, how someone outside of the organization might get access, and how to onboard new users.
Rule Five: Follow the right order.
In data governance, all categories are interconnected, so changing a part changes the whole. This is all the more reason to ensure that the steps you take toward tidying are efficient. We recommend tidying pages, then cards, then datasets.
For each category ask at a minimum, the following questions:
- Who owns this? If no one, assign an owner
- Is it being used? If not, mark it for deletion
- Do you need this? If not, mark it for deletion
- Does it follow all documented standards? If not, mark it for cleanup
In addition to that, you’ll want to ask more specific questions within each category:
- Are all cards organized into collections? If not, think about the hierarchy and organize accordingly (“at a glance” type information at the top, more detailed data underneath)
- Is this a production-ready page? If so, consider locking it so it’s not inadvertently altered.
- Does every card on this page belong here? If not, remove the stragglers.
- Has this card been viewed in the last couple of months? If not, verify need with owner
- Does this card belong to a page? If not, mark for deletion.
- Has this dataset been updated in the last couple of months? If not, verify need with owner
- Has this dataset been viewed in the last couple of months? If not, verify need with owner
- Is this dataset being used in a card or dataflow? If not, mark for deletion
- Is this dataset a duplicate? Is so, consolidate all usages to one dataset
- Does this dataset contain sensitive information? If so, ensure that access is properly granted/restricted
- Has this user been inactive? If so, consider removing them
- Is this user assigned to a group? If not, assign them as applicable, creating groups as necessary during this process
- Does this user have the proper access level? If not, ensure that they do at this time
- Does this group not have access to any pages? If not, mark for deletion if it’s older than 30 days
- Does this group not have any users? If not, mark for deletion if it’s older than 30 days
If something is not being claimed or used, or if it isn’t adding value, then it needs to be tagged for deletion or deleted.
Rule Six: Ask yourself if it sparks joy.
While the phrase “adds value” may not make your heart leap as much as “spark joy,” there is a certain tactical cheer involved in intentionally governing your data to answer business questions. When you finish your process and know that your Domo instance or data strategy is streamlined and actionable, you can be confident that you and your organization are doing your best, masterful work with optimal ease. If that isn’t analogous in data governance to joy, then I don’t know what is.
What sparks joy for me is helping people implement their ideal data governance. I’d love to help you discover tidy governance solutions. Contact us for more information.
For some other practical tips on how to stay on top of things going forward, here are Five Steps for Using Your BI Tool to Manage Your BI Tool by Senior Business Intelligence Manager, Alan Hyman.
All of this tidying would be a lot easier with a sound data strategy! Make sure to check out Plotting Your Data Strategy by our Senior Director of Data Strategy, John Lovett.