Making Websites Accessible and Personalized
Many agencies are concerned with accessibility, and state and local governments must comply with the Web Content Accessibility Guidelines 2.2 by April 2026 or 2027, depending on their size. AEM sites optimizer tools use generative and agentic artificial intelligence to scan websites and flag accessibility, search engine optimization and security issues, providing a list of action items, along with one-click fixes, to whoever runs the report.
If, say, alternative text is missing for an image, the user can easily insert text suggested by AI.
When it comes to personalizing websites, AEM doesn’t collect personally identifiable information but instead uses unique identifying features and Adobe Analytics for user segmentation.
“The whole point of Adobe Analytics is that it is a highly customizable implementation,” says Peggy Bradley, managing consultant for Adobe Experience Cloud at CDW Government.
It’s not unusual for an agency’s website to feature multiple portals or services that citizens access, and Adobe Analytics standardizes tracking across those to determine if a user is mobility-challenged or taking advantage of other accessibility features. AEM can then modify that user’s experience to help them find what they need on the site faster or simply help them navigate it more easily.
READ MORE: Here are some tips for making government websites more useful.
Adobe Analytics can identify user pain points down to a particular line of a form that is confusing people and causing them to fall out.
“In the government sector, there is a definite concern with being creepy or overstepping because people think you're collecting a bunch of data and using it against them,” Bradley says.
For that reason, Bradley says, agencies should choose one or two groups of people where a personalized experience would improve their lives. Instead of welcoming them by name, which can be unsettling, a more nuanced approach is to surface information a citizen likely wants based on previous visits.
Consent is key; citizens should always be asked to opt into these personalized experiences, and they should be able to audit them, Bradley says. But once they’re set up, it’s just a matter of evaluating, refining and further segmenting down the experiences, starting at the highest levels (often, the users’ industry and site habits).
LEARN HOW: States can digitalize documents to jumpstart IT modernization.
A Note on Customer Data Platforms
The process of user segmentation hinges on customer data platforms, which unify structured and unstructured citizen data across various sources, from online behavior analytics to a property database or driver’s license system.
CDPs are not data warehouses, centralized repositories for storing and querying vast amounts of structured data for analytics and reporting.
The goal with CDPs is to segment the data in one way or another by creating audiences for various platforms, including websites and chatbots. Maybe the audience is citizens whose driver’s licenses will expire in a matter of days, and the data to be pushed is information on renewal, to make the process as easy as possible.
“With heavily regulated industries, it becomes absolutely vital that we go through the business scenarios and ensure we understand their site,” Bradley says. “That way, we have a good grasp of what it is they’re trying to accomplish so that we can retrofit the implementation to them, rather than taking a one-size-fits-all approach.”
