Amplitude Explore Hub

An AI evaluation (or eval) is a repeatable test that measures whether an AI system produces output meeting defined quality criteria. Here's what product teams need to know.

An AI trace is the complete record of a single AI interaction: what the user asked, what the model did, and what happened next. Here's what PMs need to know.

An offline eval tests your AI system against a fixed dataset before it ships. Learn what offline evals are, how to build one, and when to use them alongside online evals.

Trace analysis is how product teams inspect AI interactions to find failures and improve quality. Here's what it is, how it works, and why PMs need it.

Add product analytics to any Cursor project in minutes. Three methods: AI prompt, Amplitude Wizard CLI, or the Amplitude MCP plugin.

Compare autocapture and manual event tracking for product analytics. Learn when to use each method and how a hybrid approach gives teams faster insights with trusted data.

Learn how to build a future-proof taxonomy that supports funnels, journeys, cohorts, and experiments.

A data taxonomy allows organizations to manage and classify information. Explore the benefits, examples, and alternatives to data taxonomies below.

Learn what account-based marketing is, how B2B companies use ABM to target high-value accounts, and why it delivers higher ROI than traditional marketing.

Learn how cross-channel marketing connects customer touchpoints, drives conversions, and maximizes ROI. Includes strategies, examples, and best practices.

Discover how cross-device attribution connects fragmented touchpoints, fixes ROI blind spots, and helps marketers allocate budgets with confidence.

Discover how campaign analytics helps you retain users, optimize spend, and drive long-term success by linking acquisition to customer lifetime value

Explore the ultimate resource on experimentation platforms. Grasp the art of data-driven decisions, boost customer satisfaction, and drive business growth.

Test real customer demand before writing a single line of code. Use fake door testing to uncover what users actually want—and focus your team’s time where it matters most.

Incrementality Testing is a controlled experiment that reveals which conversions your marketing actually caused. Learn how to measure real lift and optimize spend with Amplitude.

What is Marketing Experimentation: a structured method to test messages, audiences, and channels. Learn how to design experiments and scale proven strategies with Amplitude.

An offline eval is a pre-deployment test that measures whether an AI system handles a fixed set of known inputs correctly. Here's how product teams use them.

A code-based eval scores AI output using deterministic logic: regex checks, JSON schema validation, expected tool calls, and more. No LLM required to judge it.

An LLM as a judge eval uses a language model to score AI output against a rubric. Here's how LLM judges work, what they get wrong, and when to trust them.

Learn the art of a SaaS product launch with our step-by-step guide. Discover the best strategies, tools, and metrics for success in the competitive market.

Explore the ultimate guide to customer acquisition cost (CAC) to understand the metric’s importance, monetary value, and most impactful channels to utilize.

Dive into the world of website metrics with our comprehensive guide. Uncover valuable insights, optimize user experiences, and boost your digital success.

Learn what Average Revenue Per Customer (ARPU) is and how to calculate and increase it. Dive into the essential metric for business success in our guide.

Explore everything to know about Net Promoter Score (NPS). Learn its impact on customer loyalty and satisfaction and how to leverage it for business growth.