When VP of Engineering James Baxley joined Homeaglow, he inherited a massive ten-year-old Django monolith built by the company’s founders. It was profitable but opaque. “Before Honeycomb, it was like poking around in the dark. You’d bump into things, break glass, fix something, and hope it didn’t break again,” said James.


At the time, the engineering team consisted of only three engineers plus the founders. They were deploying changes into a system they didn’t fully understand, and they spent valuable time triaging mysterious outages.

These outages were subtle but harmful:

• A load-bearing nightly job silently stopped running for a month, skewing marketing spend
• A rogue N+1 query locked up their database
• A slow, memory-heavy code path caused a ten second delay in the core booking flow
• A debug log created a giant N+1 that added two hours to a membership billing job

On average, the team was dealing with frequent incidents, each breaking engineering focus and eroding confidence.

Alerting was equally rough. “Our alert system was literally a Twilio integration that texted stack traces to team members,” said James.


The team needed visibility and a cultural reset.

As Homeaglow expanded from three engineers to eight, the lack of visibility became a blocker to growth. “Every engineer at Homeaglow can make large-scale changes. But with so few of us, every minute not building is a problem,” said James.

Engineers spent too much time investigating fires and not enough time building the platform, innovating, or delivering new features. Honeycomb immediately changed that balance.

Homeaglow took a bold approach and instrumented everything at once. “We opened the firehose. We did not know what volume we would have; we did not know anything. That was the whole problem,” said James. Honeycomb surfaced hidden issues and revealed how the system truly behaved.


40x faster booking flow

Honeycomb uncovered a memory allocation issue that resulted in a 40x improvement in the core booking API.


Two-hour billing jobs reduced to minutes

Removing a debug log eliminated a multi-hour delay in a core membership billing process.


10x savings through caching

Redis spikes and caching opportunities were easily identified.


Zero internal incidents for over a year

After initial cleanup, Homeaglow went more than twelve months without a single internally caused outage.

Observability as second nature

When building any new projects, the team naturally thinks about how they will observe this with Honeycomb. This includes custom span additions / counters and using things like alerts helping them improve their opex on infrastructure usage.

The founders’ codebase transformed into a platform fully owned by the engineering team.

James emphasized a mindset shift: “It’s okay to not know, but it’s not okay to stay not knowing.” Honeycomb created a culture of clarity and confidence.

• New engineers begin by exploring traces
• High-cardinality attributes are added freely
• Triggers and SLOs provide proactive guardrails
• Historically fragile systems, including financial processes, are now improved with confidence
• The team investigates product ideas by checking real user behavior first

GraphQL visibility achieved

Homeaglow, a GraphQL-first API company, achieved deep observability with Honeycomb. “We now have full tracing of every level of our GraphQL execution. It has saved us many times,” said James.

Honeycomb began informing business decisions. The Homeaglow team was able to identify three profound opportunities:

Opportunity: Spanish localization

In-house data showed them that Spanish-speaking cleaners had higher review scores. Pairing that data with telemetry data in Honeycomb, particularly accept-language headers, helped them size the opportunity.

Result: Spanish localization became a prioritized initiative.


Opportunity: A customer-native app

Traces revealed that many customers were using the cleaner mobile app.

Result: Homeaglow kicked off a multimonth project to build a native app.

Opportunity: Web improvements

Honeycomb for Frontend Observability allowed them to debug Core Web Vitals, following user journeys to spot what’s slow or broken and fix it quickly.

Result: Over million dollar a year annual recurring revenue (ARR) improvement.

Observability drove both engineering improvements and product innovation.

Homeaglow now integrates Honeycomb’s MCP with internal AI tools, allowing new engineers to query system behavior conversationally. “Since we started using MCP, we have been able to improve performance of our slowest jobs in hours instead of days,” said James. “Engineers pair slow traces with our backend codebase to quickly identify n+1 queries, slow operations, or other bottlenecks and then work with tools such as Claude to build tests, then make improvements. This has fixed several key jobs that were failing and reduced the size of our event volume even!”

Observability is now a strategic capability that improves engineering velocity, product development, customer experiences, and business reliability.

1. Instrument broadly without fear. “You will find scary things. They are already happening.”

2. Embrace high cardinality. This was essential for Homeaglow’s complex workflows.
3. Use observability as an onboarding tool. Honeycomb helped new engineers ramp up quickly.
4. Build team ownership through visibility. Observability enabled the shift from founder owned code to team owned systems.
5. Use data to guide where to invest. Honeycomb surfaced unexpected behavior that led to new product investments.