“Multi-hour analysis sessions involving three separate tools turned into a single conversational workflow. That's the difference.”

Hookdeck is a leading event infrastructure platform that helps developers reliably receive, manage, and deliver webhooks and asynchronous events. Used by engineering teams building event-driven applications, Hookdeck handles the complexity of event routing, queuing, retries, and observability so developers don’t have to.

As the company scales, understanding platform usage, growth patterns, and revenue trends across multiple data sources is critical. Phil Leggetter, Head of DevRel, needed a way to investigate hypotheses about user behavior and business performance without getting bogged down in tooling.

“Before Basedash, the workflow was Metabase for querying BigQuery, PostHog for product analytics, and Claude to help generate SQL when needed. That meant constantly context-switching between three tools.”

Before Basedash

Hookdeck’s analytics workflow relied on a patchwork of tools:

• Metabase for querying BigQuery
• PostHog for product analytics
• Claude for help generating SQL and analyzing results

Phil used Claude to write SQL queries for BigQuery as well as to write SQL for PostHog when the native PostHog interface wouldn’t cut it. He would then bring those results back to Claude to actually perform the analysis. Every non-trivial question meant bouncing between three tools — writing a query in Claude, running it in Metabase or PostHog, copying the results back to Claude for interpretation, refining, and repeating. Cross-source analysis that touched both BigQuery and PostHog made the loop even longer.

Replacing Metabase

Metabase had been the team’s primary way to query BigQuery, but it was one piece of a fragmented workflow. It couldn’t talk to PostHog, it didn’t help generate or debug SQL, and every question still required manually writing queries in a separate window. Metabase did what it was designed to do, but the overhead of stitching it together with Claude and PostHog was where the time went.

Phil thought that MCP integrations might close the gap, i.e. by connecting Metabase and PostHog directly to Claude. Phil’s team tried that, but he found it lacking. PostHog’s MCP didn’t work reliably with Claude Desktop, and even when Claude could easily run queries, it couldn’t produce the kind of clean visualizations that make results easy to share and act on.

Basedash replaced that entire setup. Hookdeck connected both BigQuery and PostHog to Basedash, and from there Phil had a single conversational interface over all of the data that matters — platform usage, revenue, growth patterns, onboarding funnels — without needing to set up dashboards or write raw SQL.

“Basedash’s conversational interface collapsed that entire loop into one place. Instead of bouncing between a chat window and a query tool, the analytical flow is built in.”

Instead of preparing queries and switching tools, Phil now describes what he’s looking for, Basedash queries across the relevant data sources, and they iterate from there. The querying and the analysis happen in the same place — there’s no need to copy results back to a separate tool for interpretation. Basedash understands the business and growth context naturally, which means conversations stay productive without a lot of setup.

“It also seems to understand the business and growth context naturally, which means the conversation stays productive without a lot of setup.”

Charts that actually look good

One thing Phil noticed right away was the quality of Basedash’s visualizations. Charts and graphs generated from queries are clean, well-designed, and easy to share — a clear step up from the default output of most analytics tools.

“The design of the charts is super nice. It makes a real difference when you’re sharing results with the rest of the team.”

Impact

For a team that needs to move fast on product and growth decisions, every hour spent wrestling with tooling is an hour lost. Since adopting Basedash, Hookdeck has replaced a fragmented three-tool workflow with a single conversational interface:

• Multi-hour analysis sessions across BigQuery, PostHog, and Claude now happen in a single conversation
• Cross-source analysis that used to require constant context-switching is seamless
• Revenue, growth, and usage questions get answered in minutes instead of hours
• Beautifully designed charts make it easy to share insights across the team
• Zero time spent building or maintaining dashboards

“Multi-hour analysis sessions involving three separate tools turned into a single conversational workflow. That’s the difference.”