Vercel Speed Insights and Web Analytics now attribute data to custom environments

Vercel's Speed Insights and Web Analytics products now fully support custom environments, the company announced in a changelog post dated 10 November 2026. Deployments to a custom environment — a staging, QA or canary setup, for example — now have their performance and traffic data credited to that environment rather than being folded into the Preview bucket.

What changed

According to the Vercel blog, the previous behaviour was quietly misleading: any data captured on a deployment targeting a custom environment was attributed to the Preview environment instead. That meant a staging stack deliberately built to mirror production, or a canary slice serving a fraction of live traffic, produced numbers that were indistinguishable from throwaway preview deployments.

With the update, attribution follows the environment a deployment actually targeted. Teams can now measure how a staging, QA or canary environment performs and what traffic it receives, without those numbers being merged into Production or Preview statistics.

Vercel is keen to frame this as an attribution fix rather than a data collection change. The same data points are gathered as before, and customers do not need to change any configuration to benefit — the separation happens on Vercel's side.

Where you can see the data

The changelog lists four surfaces where environment-level data is now available:

• Dashboard: an Environment dropdown in Speed Insights and Web Analytics lists custom environments alongside the standard Production and Preview options.
• CLI: the vc metrics command can query the data from the terminal.
• Query builder: environment can be used both as a filter and as a breakdown dimension.
• Aggregation API: the data can be queried programmatically, though this route applies to Web Analytics only.

The asymmetry on the API side is worth noting: Speed Insights users are limited to the dashboard, CLI and query builder, while Web Analytics customers get the full programmatic option.

Availability

Custom environments are a Pro and Enterprise feature, so the improved attribution is effectively limited to paying teams on those plans. Vercel directs readers to its environments documentation for further detail on setting environments up.

Why it matters

Teams that take deployment seriously rarely run just two environments. A production-like staging stack exists precisely to catch problems before release, but its analytics value collapses if its data is mixed with hundreds of ephemeral preview deployments from pull requests. Load tests, internal QA traffic and synthetic monitoring all pollute the picture further.

The consequences of the old behaviour were practical. Performance regression checks run against staging could not be isolated cleanly in dashboards. Traffic anomalies from a canary rollout were invisible as a distinct segment. Anyone building automated reporting on top of the APIs had no reliable way to slice by environment.

Fixing attribution makes Vercel's built-in analytics viable as a primary signal for teams with multi-environment workflows, rather than something to be replaced with a third-party analytics tool the moment environments get complicated. It is a small change in mechanism — relabelling existing data — but a meaningful one for anyone whose staging environment is more than a disposable preview.