Skip to content

The open setup race

Each wizard – Xovris’s, and Sentry’s and PostHog’s own where they set the stack up – is run on the same small app for each stack, from the command to the first real error stored in its account. Every run is counted, and every run that did not get there is listed with its cause.

Measured 14 November 2026 in Xovris’s own test setting, 5 runs per wizard per stack, 4 stacks so far. Written by the race’s own runs, never typed.

Go (net/http)

No other set-up reached a stored error here, so there is nothing yet to compare Xovris with.Set up Go (net/http)

Set-upReached the first real errorMedian time90th percentileAnswers or pressesRuns that did not reach it
Xovris5 of 515.3 s22.9 s2none
PostHog0 of 1––1the wizard had not ended within 15 min (1); lost by the test setting, not counted (4)

Next.js

Xovris is ahead of PostHog and Sentry here: sooner to the first real error, or with fewer answers from the person.Set up Next.js

Set-upReached the first real errorMedian time90th percentileAnswers or pressesRuns that did not reach it
Xovris5 of 546.9 s59.8 s2none
PostHog5 of 5390.1 s422.4 s10none
Sentry4 of 570.9 s148.3 s12the set-up stopped with an error (exit code 1) (1)

Node.js (Express, Fastify)

No other set-up reached a stored error here, so there is nothing yet to compare Xovris with.Set up Node.js (Express, Fastify)

Set-upReached the first real errorMedian time90th percentileAnswers or pressesRuns that did not reach it
Xovris5 of 516.6 s20.8 s2none
PostHog0 of 5––15no event in its account within 15 min of the command (3); the set-up stopped with an error (exit code 1) (2)

Python (Django, FastAPI, Flask)

Xovris is ahead of PostHog here: sooner to the first real error, or with fewer answers from the person.Set up Python (Django, FastAPI, Flask)

Set-upReached the first real errorMedian time90th percentileAnswers or pressesRuns that did not reach it
Xovris5 of 512.2 s18.9 s2none
PostHog2 of 5340 s473.1 s13no event in its account within 15 min of the command (3)

How the race is run

  • Each stack has one small app with a route that fails on request; every run starts from a fresh copy of it, its own packages already installed.
  • Each wizard runs as its own instructions say, signed in to its own account, and the clock stops when the first real error from the app is stored there.
  • Answers or presses are every question the person was asked, a confirmation or a sign-in in the browser included.
  • A run that has not reached a stored error within 15 minutes counts as not reaching it.
  • A run the test setting itself lost is shown as that, and never counted as the wizard’s own failure.

Ask Xovris

Open the full page

Your data is never used to train our AI. How your data is kept

A problem is called fixed only once it works again. How each fix is checked

XOVRIS LTD, registered in England and Wales, company number 17499910. Registered office: 66 Paul Street, London EC2A 4NA.