Two tools, two questions, one complete picture

The diagnostic workflow problem

When a conversion rate drops, most merchants face the same diagnostic challenge: Shopify Analytics shows the numbers going down, but nothing in the data tells you why. Is it a seasonal effect? A new competitor? A campaign that stopped performing? Or is something broken in your store?

The right monitoring stack answers the technical half of that question definitively: is something broken? And if so, what, where, for whom, and how much is it costing?

What error tracking captures

Error tracking is your early warning system. It watches for:

• JavaScript runtime exceptions (TypeError, ReferenceError, etc.)
• Unhandled promise rejections
• Failed network requests (4xx, 5xx from your APIs and third-party services)
• Console errors from scripts you may not have written

When any of these occur, error tracking captures the technical context — stack trace, source file, browser, device, page URL — and groups it with other occurrences of the same error into a single actionable issue. It alerts you. It calculates revenue impact. It shows you whether it's getting worse or better over time.

Error tracking answers: “What is broken, on which device and browser, on which page, and how many users is it affecting?”

What session replay captures

Session replay records user interactions and plays them back: mouse movement, clicks, scrolls, form interactions, visible DOM state. It shows you what the user experienced — the visible outcome of whatever technical state the store was in.

Session replay answers: “What did the user see and do when this happened?”

Why you need both for some bugs

Consider a common scenario: Bloodhound alerts you that a JavaScript error is firing on the product page for 3% of mobile sessions. The error is a TypeError: Cannot read property 'price' of undefined in a theme script at line 847. Revenue impact estimate: $1,200/month.

From the error tracking data alone, you can:

• Identify the affected file and line number
• See it only affects mobile devices (primarily iOS Safari)
• Confirm it started four days ago (correlating with a theme update)
• Know it affects product pages with variants

This is often enough to fix the bug directly. But sometimes the line number doesn't make it immediately obvious why the property is undefined. What state was the page in? What variant was selected? What did the user do immediately before the error?

This is where session replay earns its place. Filter for mobile iOS sessions on product pages, look at sessions from the last four days, find ones that ended without an add-to-cart. Watch ten of them. You'll likely see the pattern: a specific variant selection triggers a price update that fails.

The recommended workflow

The optimal workflow for the combined stack:

• Error tracking as the alert layer. Bloodhound tells you something is wrong, with enough technical context to fix it in most cases. Set up Slack or Discord notifications so you know immediately.
• Session replay as the post-mortem tool. When error data alone isn't enough to reproduce a bug, use session replay to understand the user journey and visual state. Filter recordings by the URL, device type, and time window from the error tracking data.
• Performance data for systemic issues. Core Web Vitals trends reveal performance regressions before they become revenue issues. A LCP that's trending from 2.0s to 3.5s over three weeks is a problem that needs addressing, even if it's not producing hard errors.

Tool stack options

For most Shopify stores, the practical combination is:

• Bloodhound ($49/month or free) for error monitoring, performance, and security
• Microsoft Clarity (free) for session replay and heatmaps

This gives you complete coverage at minimal cost. Clarity is genuinely free with no session limits, has reasonable privacy controls, and provides good quality session replay for most diagnostic needs.

If you need more advanced session replay — higher quality recordings, better performance, tighter integration with error events — then Hotjar or LogRocket are reasonable upgrades. The Bloodhound + LogRocket combination specifically gives you the best of both: Shopify-native error monitoring with checkout visibility from Bloodhound, and best-in-class session replay from LogRocket.

What to measure to know your stack is working

With both tools in place, you should be tracking:

• Open issues in error tracking (should trend toward zero over time)
• Time to resolution for new errors (should improve as processes mature)
• Regression rate (errors returning after being fixed)
• Core Web Vitals p75 scores (should stay in “good” range)
• Checkout completion rate (should improve as bugs are caught earlier)

A healthy store is one where error monitoring is boring — alerts fire rarely, issues are resolved quickly, and the metrics trend in the right direction. The monitoring stack exists so that exceptional events get your attention, not so you have to look for them.