Dynatrace RUM Configuration Question for Micro‑Frontend Setup

[riship1699]

I have a question about how to best configure Dynatrace RUM for our micro-frontend architecture.

Our shell application loads the DynatraceRUM cookie globally through a shared footer, and the individual MFEs are deployed independently and mounted into the shared shell.

Given this setup, should each MFE be configured as its own Web Application in Dynatrace, or is it better to keep everything under a single Web Application? I want to follow best practices before making any changes, especially since each MFE is owned by a different team in our org.

[t_pawlak]

Hi,
IMO single Web Application works best. 

From the end-user and browser perspective, this is still one continuous application and one user journey.

All MFEs share the same session cookie, so splitting into multiple web apps often breaks session continuity and makes analysis harder. You keep full end-to-end visibility (navigation, errors, performance, conversions) across MFEs.
Instead of creating multiple web apps,

You can add a custom session property, e.g. mfeName, set by each MFE (via route, JS variable, or some CSS selector)
You can also standardize user action naming to include the MFE name.

[t_pawlak]

Hi,
With your setup (shell + MFEs mounted into the same browser experience, one globally loaded RUM snippet and shared cookie), the usual best practice is to keep this as one Web Application—because from the user’s perspective it’s one continuous journey. In Dynatrace RUM, “applications” are primarily separated via application detection rules (URL/host/path), so splitting MFEs into multiple apps often makes analysis more fragmented.
IMO instead of multiple Web Applications, do logical segmentation per MFE.
Try use or add custom user action / session properties like mfeName, team, route, set by each MFE (on route change / MFE bootstrap).
Define user action and user session properties for web applications 
Moreover, you can also standardize user action naming (e.g., prefix with [MFE:Payments]) via the JavaScript API so teams can filter and build views easily.
Customize Real User Monitoring with the JavaScript API for web applications