ScaleBun

I’m Tired of Opening Four Tools Just to Figure Out One Bug

AnalyticsScaleBunPublished August 13, 20265 min
I’m Tired of Opening Four Tools Just to Figure Out One Bug — ScaleBun blog coverI’m Tired of Opening Four Tools Just to Figure Out One Bug — ScaleBun blog cover

I ran into this the other day and it reminded me how weird our debugging setups have gotten.

We had an error showing up in production. Nothing crazy at first. I opened the error, checked the stack trace, and wanted to see what the user was doing right before it happened.

So I went to the replay tool.

Couldn't find the session right away.

Went back, grabbed the user ID, searched again. Found something close to the same timestamp.

Then I wanted to know if this person was using the new version of the feature.

Different dashboard.

Then someone asked if the bug was actually affecting conversion or if users were just retrying and moving on.

Back to analytics.

At this point I had four tabs open just to understand one person's bad checkout.

And none of the tools were really broken. They were all doing exactly what they were supposed to do.

That's kind of the problem.

Four tools, one broken checkout#

We've split one user experience across a bunch of different products.

The error tool knows an exception happened.

The replay tool knows what the user saw.

Analytics knows they didn't finish checkout.

The feature flag tool knows they were in the new rollout.

But I'm the one sitting there trying to glue all of that together.

What I'd rather see#

Something close to this, in one place:

Text
User opened checkout, new_checkout: trueClicked PayPOST /payment → 500Frontend exception firedUser retried — same resultUser left(replay attached)

That's pretty much everything I need to start debugging.

This isn't a web vs. mobile problem#

On the web maybe it's a JavaScript error.

On mobile maybe the app crashes.

Either way, I'm trying to figure out what the user did, what the software did, and where things went wrong.

Where this is going#

That's actually a big part of what we're trying to do with ScaleBun.

Instead of treating errors, sessions, product events, flags, and release data like completely separate worlds, they can share the same session and user context.

I'm not really interested in having another dashboard with 40 charts.

I mostly want to stop copying IDs from one tool into another just to understand what happened.

One timeline instead of four tools

See how ScaleBun connects errors, sessions, and analytics into one debugging context.

See connected monitoring→