← All posts

Six analytics patterns. Zero SQL. Introducing Kanject Insights.

We shipped a click-to-insight engine that supports the six analytics patterns every product team needs — funnels, cohorts, correlations, and more — without writing a line of SQL.

Most analytics tools force a tradeoff: either your data team writes SQL for every question the product team asks, or the product team gives up and ships blind.

Kanject Insights makes that tradeoff disappear.

The pitch

You stream events into your AWS account. The dashboard ships with six built-in analytics patterns, each rendered as a guided builder. Anyone on the team can answer “why” with clicks instead of SQL.

The six patterns are the analytics literature’s greatest hits:

  • Standard Analysis — deconstruct single events across multiple metrics
  • Funnel Analysis — map conversion through sequential steps
  • Correlation — discover relationships between distinct events
  • Cohort Analysis — track behaviour patterns of user segments over time
  • Comparative — compare metrics across time periods
  • Event Sequence — analyse ordered patterns with temporal constraints

Each one is a pattern your data team would otherwise hand-roll in SQL — every time someone in product asks for it, fresh, from scratch. We pre-built them.

What we didn’t do

We didn’t build a SQL editor with autocomplete. We didn’t build a notebook environment. We didn’t build “natural language to SQL” — that’s a great demo and a bad product.

We built six guided builders, each tuned for the kind of question it’s designed to answer. A funnel builder doesn’t let you ask cohort questions, because that’s not what funnels are for. A cohort builder doesn’t let you ask comparison questions. The constraints are the feature.

Why on AWS, in your account

Every event you stream into Insights lives in your DynamoDB. Every query runs in your Lambda. We never see the data — we ship the engine that runs in your account.

This matters more than most analytics vendors will admit. Your customers’ product behaviour data is sensitive. Sending it to a third-party SaaS turns “marketing-ish” data into a compliance question every quarter.

Insights sidesteps that entirely.

What’s coming

The first cut of Insights ships the six patterns and a hosted dashboard. AI summaries — the natural-language “what changed and why” feature you’ve seen elsewhere — is on the roadmap and marked clearly as coming soon on the product page.

If you’re shipping a product on Kanject and you want analytics your team can actually use, this is the moment to talk to us.

More field notes from the team.

Architecture April 26, 2026

Building a serverless analytics engine that scales to zero

Most analytics products solve the dashboard latency problem with an always-on OLAP cluster. We solved the same problem differently — Lambda + DynamoDB + S3 Express + EventBridge in your AWS account. Here's the architecture brief.

Engineering April 15, 2026

Parallel.ForEachAsync wasn't enough. So we built our own.

The four things every batched-I/O workload needs — scoped DI, retry policy, per-partition timeout, and partial-batch acknowledgement — none of which the BCL gives you in one place. Here's why we shipped Kanject.Core.Parallel.

Engineering March 18, 2026

How we ship source generators that survive refactors

Lessons from two years of building Roslyn generators that don't silently break when downstream code moves.