How to create a seamless user experience through administrative governance

Administrative governance—meaning the roles, processes and controls that are used to manage systems at the tactical rather than organizational level—can be a critical lever in maximizing the value of your GIS. But the best governance isn’t even noticed by the everyday user. So where’s the sweet spot between too much governance and too little that will allow your users to get the most out of your GIS? Esri Canada’s Julie Roebotham has a tool that can help.

Is there any topic of conversation more exciting than governing the administration of your GIS?

Okay, well, maybe for some there is. But there’s also something very satisfying about having a well-functioning GIS, which is, in part, the result of some well-planned-out governance. 

So what do I mean when I say “administrative governance of GIS”? Governance at a high level can be defined as the roles, processes and control mechanisms used by organizations as a formal framework for managing their systems. In the context of GIS, and in particular web GIS (e.g., ArcGIS Online, ArcGIS Enterprise), we use these mechanisms to govern geospatial programs whose purpose is to maximize the value delivered by the GIS.  

Governance is a key part of many of the geospatial strategies our clients develop. On the organization level, this includes defining the structures, controls, processes and performance metrics required to evaluate the benefits of the investment in the geospatial arena. We work with our clients to review their technology, data and delivery mechanisms; evaluate aspects of their workforce, people and skills; and to identify budgeting mechanisms such as approval boards, prioritization frameworks and methods by which to measure benefits realized. 

Just below the organization level, there are governance mechanisms that we can develop and apply to the tactical, technical and administrative levels of a geographic information system. 

Have you ever asked yourself any of the following questions when considering how your GIS will be used on a day-to-day basis?

• “How do I decide what privileges a user gets?”
• “How will our users know which datasets are authoritative?”
• “Do we allow everyone the ability to publish data? Or should we restrict it in some way that still allows for innovation?”
• “Who performs regular or ad-hoc administrative tasks?”
• “How should we manage our public content? Who shares it and how can we be sure that it’s good enough?”

If you answered “yes” to any of these, then you’re wading into the waters of administrative governance. 

If you have well-developed governance, your users will not be aware that their seamless experience creating, using and sharing geospatial products is due to some careful consideration and planning on your part. Their experience is that their GIS just works—they can create, innovate and collaborate without knowing there are controls and processes at play.

A good analogy to this is the preparation required when hosting a big party—your guests just show up and enjoy a harmonious evening where there are no food shortages, the entertainment is flawless and there are no line-ups at the bar.  It took work to get it that way, but your guests can just focus on enjoying the fruits of your labour.

When developing governance for a system, be aware that there is a sweet spot. Too much governance, and your users are going to find it difficult to complete their tasks or know which processes to follow. Not enough governance, and you may find yourself with a chaotic space, with multiple versions of the same data, not enough licenses and no idea how to request help.

So how can you, as the GIS system administrator, determine what the sweet spot for administrative governance is? And how can you apply it?

Administrative governance from a web GIS perspective can be thought of in five broad categories:

1. User management: How are users managed in the system? 
2. Content management: What are the rules, roles and responsibilities around creating and maintaining content?
3. Data management: What are the rules, roles and responsibilities involved in storing, publishing, sharing and maintaining data?
4. System management: How do you assess impacts to system resources?
5. Change management: What is considered a change to your system?

A note on change management: this is obviously a large topic and there are innumerable volumes that focus on that topic alone. For the purposes of this blog post, implementing governance is, in itself, a change to the system. Working in a dynamic content management system such as a web GIS elicits near-constant change. The trick is to decide what requires change management processes, and what tasks are part of normal day-to-day operations.

Before we discuss these topics, I want to highlight that how you address each topic, and what items are included under them, will be largely influenced by your organizational structure. My colleague Matt Lewin has a great blog post that outlines some common organizational models. The first step in any sort of organizational governance is to understand where your specific hierarchy falls. Knowing this will inform how you will develop governance under each of the five categories I mentioned above.

For example, let’s assume we’re working with an organization that loosely follows the hub-and-spoke model. This means they have a central group (possibly IT, or some other central GIS team) that is responsible for management of commonly used datasets, GIS support and administration. There are likely GIS professionals in the central team, but expertise is de-centralized so that most business units have their own GIS subject matter experts (SMEs) and manage their own department-specific data. The administrative governance defined by this organization should reflect that dispersion of talent across the organization.

Take publishing, for example. A hub-and-spoke organization may decide that SMEs should have the ability to publish their own data to the organization without requiring additional processes. They may decide to apply their governance on the onboarding of SMEs, and they may decide to limit the privilege of sharing data to the public. They may decide that content itself has its own branding roles, and public content requires certain logos, and ownership, and therefore becomes the task of the subscription admin.

Determining the organization’s requirements for these types of controls and then fleshing out the accountabilities, processes and technology required becomes the fulfillment of the governance needed to satisfy the controls. As with all things governance, the input and output very much depend on what type of organization you are. 

So back to the components of operational GIS governance—what are they, and what are some examples? 

It’s helpful to think of each category using the three principles of governance: people, process and technology. I’ve created the matrix below to help you understand what kinds of topics may fall under each category for each of the three principles, but this is very much just a sample of the topics that could be included.

Now that you understand at a high level the categories at play and how your personal organizational model can influence what items you want to control under each category, you can use the matrix above to work out what each item looks like for your ArcGIS organization.  

Let’s take our example of a Hub-and-Spoke organization from earlier. Once an organization understands their specific organizational model, the next step is to workshop the high-level user roles that will be required. 

For a Hub-and-Spoke model, the team may decide that the central group will provide administration of the organization as a whole, but most departments have SMEs and can retain their own departmental administrator to handle their needs. Just this decentralization alone will increase departmental efficiency as it will enable them to perform tasks without requiring the organizational admins.

The central group may also decide that the departments can nominate a SME to be the data publisher and ultimately can answer for data completeness and correctness, and that each department will have editors. They may also recognize from time to time the organization will hire consultants who may need access, and that these will be handled on a case-by-case basis. (Maybe down the road they will define more rigorous governance, but for now we’re looking for that sweet spot.) 

Here is the initial output of that brainstorming session:

GIS Administrator

Departmental Administrator

Data Custodian / Publisher

Editor

Consultant

After you’ve completed this brainstorming activity, you’ll have a pretty decent list of the accountabilities, processes and technology you want to develop governance for. 

The next step (for the purposes of this blog post) is to align your governance with established best practices. If you find something isn’t working, discard it. There is no need to keep governance that doesn’t work just because you went through the trouble of developing it. The governance you develop is a living entity and needs to be regularly audited to ensure that it’s contributing to the functioning of your GIS and not taking anything away, including unreasonable amounts of your own time.

Lastly, don’t be afraid to seek help. I highly recommend reviewing Matt Lewin’s content on the Esri Canada Blog—it’s a wealth of information on governance and strategy. And sign up for his newsletter, The Geospatial Edge. Finally, you can always reach out to Esri Canada for assistance identifying and implementing the right level of governance for your organization. We have helped many clients across the country develop, implement, refine and benefit from the right governance mix.

By constantly seeking to identify and apply the right level of governance at the right time for your organization, and by being committed to regular, honest review, you’re setting yourself up for the GIS organizational equivalent of a well-planned social engagement, where no one is waiting at the bar and all the snacks are delicious.