11

Designing Exceptional Experiences

So far in this book, we have focused on the parts of ServiceNow that most often form the core of the business case such as how to optimize the processes to maximize business value while avoiding common challenges. One area we have not spent much time on is the creation of an exceptional user experience (UX) in the platform or the tools that can be used to develop this experience.

In this chapter, we’ll cover the parts of the ServiceNow platform that most strongly support an exceptional UX, helping you understand the various options available for building a great UX and how to leverage advanced AI capabilities such as AI Search and natural language processing (NLP) to provide consumer-grade experiences to your customers. We’ll cover such diverse areas as the following:

• Types of ServiceNow user interfaces
• Portals
• Workspaces
• Conversational interfaces

As we did with other chapters, we’ll focus on the what and the why of your implementations, leaving the how to ServiceNow’s abundant documentation and training programs.

Types of ServiceNow user interfaces

As of 2022, the ServiceNow platform has had three major waves of user interface (UI) technology deployed and used in the platform. In this section, we’ll make you aware of these and their context in the history of ServiceNow to inform your decisions on how you’ll develop experiences for your users.

Jelly

We will cover Jelly briefly because, as of the San Diego release and the Polaris UI, the majority of ServiceNow’s interface has moved away from this technology. Apache Jelly, or Jelly as it’s known in the ServiceNow ecosystem, is a scripting and templating engine that allows for the relatively efficient creation of customized UI elements integrated with the ServiceNow JavaScript engine. For over a decade, Jelly was the main tool in a developer’s toolbox for creating UI components in the Native View or non-portal view of ServiceNow. Jelly was also the primary tool for creating end user-facing portals until the Geneva release in late 2015. Jelly was a great tool for the time and an important part of ServiceNow’s technology stack for over a decade. Today, however, there is virtually no case where a modern customer should need to make use of Jelly for UI configuration or customization except for limited support of legacy use cases that have not yet been modernized. As a rule of thumb, as of late 2022 and for customers on the San Diego release or later, we would recommend extreme caution when any new functionality is suggested to be created in Jelly code. Jelly also had a fully functional end user portal experience called the Content Management System (CMS) but over time, ServiceNow customers began to call for more flexible and powerful development tools to allow rapid, modern web design to be incorporated into ServiceNow end user experiences.

Service Portal

In response to the need for a more modern and flexible end user experience, ServiceNow developed a way to build AngularJS widgets in ServiceNow and connect them in a modular way to create even more appealing and capable portals for users. ServiceNow also made use of this Angular framework to develop some of the early workspaces in the platform because the technology lends itself well to interactive and dynamic pages.

The vast majority of customer-facing portals in 2022 are still developed using this Service Portal technology, which provided fully customizable portals that could be tailored for end users. However, as of the Rome release, an Employee Center application is available as the recommended tool for the creation of customer-facing web portals. Employee Center uses the Service Portal technology but provides a far more structured and prescriptive approach than traditional service portals.

Most of the rest of this section will cover the end user experience in the context of the Employee Center journey, as this is currently the recommended end user experience framework and Employee Center provides a general approach to developing an end user portal that does not vary a great deal based on the implementation technology. Before we look at that implementation journey, it’s worthwhile to consider the next wave of frontend technology that is likely to become the standard in an upcoming release, which is branded by ServiceNow as the Next Experience UI.

Next Experience

Building on the success of ServiceNow’s Service Portal, Next Experience is likely to improve the ability of customers to configure web portals for their end users. As of the San Diego release, Next Experience portals can be used as the portal interface for custom applications but are not recommended for building your Enterprise Service portal, although this is expected to change in a subsequent release. As the Next Experience UI is an emerging area, we’ll cover the key differences from Service Portal so that you understand when to choose one or the other. Let’s look at the application of UI technologies in the context of the major use cases for the ServiceNow UI. We’ll start with portal experiences before moving on to the fulfiller experience.

Portals

Portals in ServiceNow are web pages targeted at a broad population within an organization (typically all employees) with the goal of providing effective service and information to a broad customer base. Portals host interactive processes such as requests for goods and services, articles, and informative content as well as outstanding tasks and approvals. Portal content may be the same for all visitors but is often targeted based on information about the user such as country, job function, or other targeting criteria.

Types of portals

End user portals tend to fit into three categories depending on the scope of their coverage and the degree of specialization in their functionality. We can divide portals into three portal types:

• Enterprise portals: An enterprise portal serves as a one-stop shop for your organization to access corporate services and can span Human Resources (HR), Information Technology (IT), Legal, Finance, and more.
• Departmental portals: Departmental portals are focused on providing service and information from a single department or group. An HR portal or an IT portal is a clear example of this portal type.
• Application portals: Some ServiceNow custom applications have an end user-facing component that calls for a portal to be created for that specific use case. Examples of application portals can include portals created for third-party applications, external customer support, or tailored interfaces for complex business processes that don’t exist in the out-of-the-box processes.

Some portals even evolve through multiple types, often growing from a departmental portal and expanding their scope to other processes. By planning your portal journey correctly from day 1, you can ensure it will scale as your needs expand.

The journey from department to enterprise portals

Let’s look at the first two cases together because they currently have a single recommended approach, Employee Center, and because they typically represent a continuous journey from departmental to enterprise portals. It is common for ServiceNow implementations to deploy portals for one or two departments in a major release and then to expand to new use cases and departments in a subsequent release. By adopting Employee Center even where your use case calls for a departmental portal, you will be setting a foundation that will scale to other processes far more efficiently than if you develop a department-specific portal. If your initial release calls for multiple departments to be providing end user services, then the case for the Employee Center approach is only strengthened as it will enable you to deploy each department’s processes into a common framework. Once your portal is serving as the enterprise hub for service management, you may even choose to expand that scope to encompass employee engagement and content delivery:

Figure 11.1 – An illustrative Employee Center portal with the default configuration

Employee Center will allow you to realize the vision of a single place to provide services to your employees but the potential doesn’t stop there. Let’s look at how Employee Center Pro allows you to further increase the impact of your portal.

Employee Center Pro

The final stage in the enterprise portal journey is to go beyond service management and to make Employee Center a destination for your employees to find information, even replacing your corporate intranet. To help facilitate this, ServiceNow offers an Employee Center Pro app as an optional value-added capability that provides enhanced capabilities that allow ServiceNow to act as a corporate content repository in addition to being the service delivery framework. The Pro level of Employee Center adds content publishing, communities, and enterprise-level search capabilities. In your deployment, a good rule of thumb is that if you are aiming to collaboratively engage customers or to implement or replace an Intranet solution, then you should be considering Employee Center Pro.

Content taxonomy

A key concept for deploying Employee Center is the creation of your information taxonomy, which is a structured tree of topics that can be associated with content in your portal. The taxonomy is used to render the menus and topic-based pages and enhance the search experience, so it is worth investing time in creating a logical taxonomy that makes sense for your content. Remember that the creation of content such as catalog items, knowledge articles, and so on should be aligned to value, as we discussed in the opening chapters; in the same way, the taxonomy should serve to provide value to users by helping to organize the items you’ve created and improving the user’s ability to find the right content to solve their specific needs.

Important note

ServiceNow uses the term taxonomy in multiple contexts. There is the Service Portfolio Management taxonomy, GRC taxonomy, and Employee Center taxonomy. The content in this section refers only to the Employee Center taxonomy. It is particularly important when consulting the documentation site and system menus to ensure you’re looking at the right taxonomy.

ServiceNow provides an Employee Experience Taxonomy out of the box that will be populated with topics based on your installed applications. It is suggested to clone this taxonomy as a starting point for your own taxonomies and modify it as needed. Once your taxonomy has been defined, you should consider the visual presentation of the content. In the next section, we will look at tailoring the look and feel of your portal.

Custom app portals and portal customizations

ServiceNow’s portal experiences can be fine-tuned to meet the visual and functional needs of your organization. This can range from simple color updates to modifying the structure and behavior of pages and their component widgets.

Many customers choose to modify and tailor their Employee Center portal’s style using Service Portal’s Branding Editor to adjust the look and feel by updating the colors of the portal theme to reflect their own color palette. This is useful if you want to give your portal a look that matches your corporate identity with brand-approved colors and your own organizational or portal logo:

Figure 11.2 – Service Portal’s Branding Editor

Using this branding editor, you will be able to navigate through the various pages in your portal to see how the changes appear on each page. This will help identify any changes that need to be made or to confirm that the design is ready to be used.

Employee Center also allows for the configuration of the items that are displayed in the out-of-the-box widgets, including the following items:

• To-dos
• Available requests
• Chat functionality
• Recommendations
• Active items
• Cross-channel favorites
• Launchable business applications
• Employee profiles
• Custom footers

These configurations allow you to fine-tune the behavior of the portal and act in conjunction with your content taxonomy to form the basis of your portal experience.

If you wish to make further functional changes to the Employee portal, you can use the Service Portal configuration tools such as the Designer, Page Editor, and Widget Editor (as well as the backend forms for these objects) but this should only be done with an abundance of caution and a strong business case. Employee Center continues to be invested in by ServiceNow and modifications to out-of-the-box capabilities can lead to you missing out on new features or issues during upgrades.

Application portals

These warnings against customization are appropriate when considering a modification to an out-of-the-box portal such as Employee Center but, in some cases, you may choose to develop a new portal for a custom application. This is where the third category of application portals comes in. These portals are intended to facilitate more complex processes where you have established a clear business need. As of San Diego, the recommended approach for custom application portals is to use the portal Experience option in App Engine Studio, which uses the Next Experience UI framework as its UI technology:

Figure 11.3 – Adding an experience from the App Engine Studio Experience tab

When adding an experience in App Engine Studio, you can select from some pre-existing experience templates. For most end user use cases you’ll find the Portal experience to give you the best starting point.

Figure 11.4 – A Portal experience can be selected in App Engine Studio when building a custom application

It’s also still possible to create Service Portal-based custom application portals, which is a choice that some customers might make if the availability of out-of-the-box Service Portal widgets suitable to their use cases would drastically simplify portal development when compared to the efforts using UI Builder.

Portals provide an excellent way for your organization to interact with ServiceNow-enabled processes, but they are used by the consumers of services rather than the providers of services. To provide maximum value, ServiceNow also provides a strong backend UI capability that helps your teams to deliver these services more efficiently and effectively.

In the next section, we’ll explore the interfaces that can be used to get work done most efficiently within ServiceNow and to help you navigate the decisions you’ll need to make when deploying workspaces for your fulfillment teams.

Workspaces

The most basic interface in ServiceNow is the native view, which consists of the forms and lists that make up the core of ServiceNow’s traditional UI. While it is possible, in most cases, to use this ServiceNow native view to complete almost all tasks that are required to provide service and maintain operations, the use of the native view is less efficient than a specially designed type of interface known as a workspace.

Workspaces have been developed specifically to optimize service delivery and represent the most efficient way for Level 1 and 2 agents to engage with the platform. You should note that administrators and power users will continue to use the native view, as many of the configuration and administration features are only exposed in the native view. 

Until the Next Experience UI framework is fully integrated into all ServiceNow product workspaces, there are two types of workspaces – Legacy Workspaces (also called agent workspaces) and Configurable Workspaces (sometimes referred to as WEP workspaces in ServiceNow documentation). 

New workspaces in the Now Platform will be Configurable Workspaces and use the Next Experience framework and can be tailored in UI Builder. You are also able to develop your own workspaces using UI Builder. Since there are still some workspace experiences that have not been moved over to the Next Experience framework, in some cases, you will have the choice of a Legacy Workspace or a Configurable Workspace, so it is useful to understand the difference and when to use each.

Determining which workspaces to use

Legacy Workspaces are most notably still used for ITSM and ITOM service delivery as of the San Diego release of ServiceNow, while CSM, vendor management, cloud migration, and several other workspaces are already available as Configurable Workspaces, with more expected as store applications and in future releases. 

For the time being, Legacy Workspaces may continue to be used by organizations that have not yet migrated but for new implementations, it would be best to use a Configurable Workspace if it were available. Developing and customizing Legacy Workspaces is not well supported, so all new workspaces developed as part of a custom application should be created using Next Experience and UI Builder. 

Because all ServiceNow user interfaces operate on a unified data model, it is possible and relatively seamless to use different workspace types together, even where processes interact. For example, a CSM Configurable Workspace can be used to serve customer cases, while incidents created from cases can continue to be worked via the Legacy ITSM Agent Workspace. Configurable Workspaces will tend to offer far greater versatility to be configured to meet specific business needs using UI Builder. 

Just as it is recommended to think carefully before making changes to a Service Portal, you should carefully consider the value and cost of changes to Configurable Workspaces; when the value justifies the effort to build and maintain the changes, you will be able to make the changes using UI Builder. 

One of the advantages of the workspaces is their ability to integrate AI recommendations and search into the context of service delivery. In the previous chapter, we addressed the Predictive Intelligence framework, which powers intelligent recommendations in the workspaces. Specifically, when you are viewing a ticket record in a workspace, Predictive Intelligence can find similar records and even provide field predictions to reduce manual effort and errors. Another ServiceNow AI capability that plays a part in the workspaces is the AI Search capability, which you can tailor to surface relevant content in the context of your workspaces even without UI customization. AI Search is just one of the advanced AI capabilities that can be used to help shape a better user experience for both agents and fulfiller users; in the next section, we’ll look at some other examples.

Optimizing experiences with user-facing AI

We have covered the various types of user experiences and outlined the approaches you should take to deploying them in light of the changing landscape of ServiceNow UI technology so far. In this section, we will cover some of the supporting but highly value-generating features of the ServiceNow platform that allow us to shape experience, not only through design but also through intelligence.

AI Search

The AI Search capability in ServiceNow is another multipurpose solution that is useful in improving the search experience of users in the Employee Center portal as well as surfacing relevant information to agents working in their respective workspaces. 

AI Search utilizes sophisticated algorithms to find and rank information in a way that generates more meaningful results than prior generations of search technology. Specifically, AI Search utilizes a form of language processing known as semantic search and enhances this with a ranking model that combines administrator input and aggregated user behavior to determine the correct results to display and the order in which to display them.

AI Search configurability

When configuring AI Search, you can specify the content that should be indexed and various other parameters, such as stop words and boosting and blocking rules that help surface the most relevant content. You can also configure a special type of result known as a Genius Result, which can be triggered with the ServiceNow Natural Language Understanding (NLU) features.

Genius Results should be generated only in the case that the results are most strongly relevant to the user’s query. Some good examples of this would be when a search can return a closely related catalog item or highly relevant knowledge article. In the language of the previous chapter, Genius Results have very low coverage but high precision, meaning that a Genius Result is unlikely to be provided for any given search term, but when shown, a Genius Result is very likely to be correct. You can create Genius Results from records in ServiceNow or from other available data sources accessible via an API.

AI Search works with the help of an external search service that runs alongside your ServiceNow instance and provides a consumer-grade search capability in the portal, workspaces, and via APIs that allow access to custom applications and features. AI Search can index content both inside and outside of the Now Platform and offers seamless integration to both portal and workspace experiences. In addition to the presentation through these interfaces, you can use AI Search in a Virtual Agent conversation, which brings us to another form of user experience that is often embedded in ServiceNow’s portals, Virtual Agent.

Conversational interfaces

One capability that can provide a lot of value is the Virtual Agent chatbot capability that ServiceNow offers alongside its NLP engine. Virtual Agent allows users to interact via text chat with an automated system to quickly handle simple tasks, even when no human agents are available. If a Virtual Agent is unable to handle a query, then it can be transferred to a human who can carry on the support conversation using the chat interface built into the agent workspaces.

Incorporating Virtual Agent into your user experience

Virtual Agent is typically presented through a service portal in the form of a webchat window. Virtual Agent serves as a highly automated first line of triage and can address user needs that would otherwise have required the services of a human agent. 

Virtual Agents can also be exposed through an external chat channel such as Microsoft Teams, Slack, the ServiceNow mobile application, and even custom conversational interfaces.

When introducing Virtual Agents into your environment, you should be careful to consider how a chat-based channel provides different user experience trade-offs relative to a form-based interface. There are both advantages and disadvantages, and not all use cases lend themselves well to Virtual Agents.

Advantages of Virtual Agent as a user interface

Just like a conversation with a regular human, conversational interfaces such as ServiceNow Virtual Agent can take different paths through a conversation in response to the choices of the user. This allows for a highly dynamic experience that reacts to the user and asks only for information that’s not already available in the system or in the user’s past inputs. Virtual Agent topics can be configured to take action automatically based on the user inputs, often orchestrating work inside of ServiceNow or using external integrations; this allows a user to get an instantaneous resolution to their needs and results in both improved customer satisfaction and lower support costs because a human agent was not required.

Disadvantages of conversational interfaces

Virtual Agent interfaces do have some significant drawbacks relative to a standard web portal. The logic behind Virtual Agent conversations is specified by the topic developers and follows predetermined paths, unlike a human conversation. If you’re not careful about the design of these paths, the user can feel locked into an incorrect path and get frustrated with the experience.

The chat interface is also low-bandwidth, meaning that you’re limited in the amount of information you can convey over the chat channel. The chatbot interfaces, such as Virtual Agent, require you to think carefully about what information you want to present to the user with each new message. 

Recommendations for Virtual Agent

The following recommendations will help you make the most of your Virtual Agent deployment and to create more value than just a portal alone:

• Consider enabling topic switching when the user is asked for input, as this enables a user to get away from a “wrong” path as early as possible to reduce frustration. If agents are available, the option to chat with a human operator is an excellent fallback. Virtual Agents should be seen as the first line of support (often called Tier 0), not your only support offering.
• When asking for user input, filter the options to as small a list as possible to minimize the effort for the user. As a rule, consider five to seven items as a maximum number of options per interaction. If you need more, you may prefer a text input with NLP enabled.
• Reusable topic blocks can be used to create consistency and reusability across multiple conversations. Remember the goal of the Virtual Agent is to get things done quickly and without human interaction and that a familiar experience contributes to that goal. The out-of-the-box topic blocks can also be composed into your topics to provide a rich capability very quickly.
• Not every form is appropriate for a Virtual Agent conversation. When a large amount of user input is required, the chat interface can feel clunky. Users are comfortable tabbing or clicking through forms and may want to review all their responses before submitting.
• Where information or inputs are very complex, a Virtual Agent can also serve to find the appropriate resource and link to it for the user rather than forcing that information to flow through the chat channel. If your friends or family wanted to share a long news article with you, they would send you a link, not copy the article text into their messages to you. Consider taking the same approach here.
• Surfacing the Virtual Agent through a common organizational channel such as Teams or Slack can save the user the effort of navigating to the portal and cut down friction in the process. Also, consider that these channels make the Virtual Agent available on mobile devices, expanding their reach significantly.

Virtual Agents are made much more powerful and useful when integrated with the ability to understand written language. Let’s look at ServiceNow’s language processing features.

ServiceNow NLU

When a user first engages a Virtual Agent in a conversation, the task of the chatbot is to match the user’s needs to a defined topic or a fallback. This can be accomplished with basic keyword matching or by using ServiceNow’s own implementation of NLU. NLU models can be trained directly in ServiceNow and broadly serve two main purposes – intent recognition and entity recognition.

Intent recognition is the task of taking some input sentence and mapping it to a predefined intention that can trigger a specific flow. An input sentence might read “Hey there ServiceNow, I’d like to order a new laptop.” This could be mapped to an Order Hardware intent. Note that the actual intents do vary from customer to customer so one organization may match this text to the Order Something intent while another might map it to Request Replacement Laptop. A good language model is one that most reliably finds the intents closest to the user’s requests and one that does not often trigger the wrong topic flow.

Entity recognition is a second function of the ServiceNow language models and allows us to recognize not just the type of topic but even some specific data that might be useful in acting on that intention. If a user was to say to the Virtual Agent “Please order me a black iPhone 13 Pro,” the system could not only detect a request for a new phone but also key facts that we might otherwise have had to prompt the user for, such as the model (iPhone 13 Pro) and the phone color (black).

By combining NLU and Virtual Agent, you can allow users to provide a great deal of information in a short message that puts you well on your way to intelligently responding to their query in an automated way.

Summary

In this chapter, we’ve covered the major types of user interfaces in ServiceNow and addressed portals, workspaces, and the supporting AI technologies that make these experiences more relevant and intelligent. Throughout the book, we’ve aimed to create a useful store of information that applies across releases but in this chapter, we’ve had to get into the details of an important shift in the ServiceNow UI that makes some of the advice necessarily specific to the version available as of the time of writing this book., By explaining the motivations and technologies involved, you will now have the tools to make similar decisions well into the future as well.

With this chapter, we also conclude the final section of this book in which we’ve looked at how to drive innovation on Now Platform and to create efficient, intelligent experiences for end users and agents alike.

With the foundation you’ve gained in this book, we hope you’ll be better prepared for your journey to make an impact with ServiceNow. From this point on, you should prepare for those capabilities most likely to deliver value and be useful to you in your roles.

No matter where you are within the ServiceNow ecosystem, you will benefit from getting hands-on with the software. Go to https://developer.servicenow.com and sign up for a personal Developer Instance where you can build and explore on your own.

For those pursuing a more technical path as a developer, the book ServiceNow Development Handbook (Tim Woodruff) provides excellent implementation tips.

For those focused on product and process, the ServiceNow documentation for each product area remains the most complete and exhaustive guide to the product and best practices.

Finally, ServiceNow has a vibrant community at https://community.servicenow.com, where a global community of your ServiceNow peers is willing to help answer questions and offer advice.

It’s almost impossible to learn about everything there is to know about ServiceNow given how broad the capabilities are but if you start somewhere, keep learning, and always stay focused on delivering value, you’ll most likely be on the right track.