Amplitude launches AI Assistant for in-product support

Amplitude has launched AI Assistant, an embedded support agent for use inside digital products. The tool is designed to help users without sending them to separate support channels.

The product sits within Amplitude's analytics platform and uses data on what a user has done inside an application to shape its responses. It can guide users through tasks with step-by-step walkthroughs or carry out actions on a user's behalf to complete workflows.

The launch reflects an effort to move customer support closer to the point where problems occur. Instead of relying on a separate chatbot or ticketing system, the assistant is designed to respond within the product experience and draw on a user's in-product history.

This links support interactions to behavioural data, product journeys and guidance tools already held in the platform. For product teams, that creates a view of where users get stuck and how those moments relate to activation, conversion and retention.

"When a user gets stuck in a product, they shouldn't have to file a ticket or go hunting through help docs," said Spenser Skates, co-founder and chief executive officer of Amplitude.

"Amplitude AI Assistant lives inside the product, understands what users are trying to do from their behavior, and helps them finish the job on the spot. This is the start of a world where products can take care of their own users, and getting help is part of the experience, not a separate, frustrating process," Skates said.

How it works

The assistant is built to do more than answer direct questions. According to Amplitude, it can monitor for signs that a user is struggling and intervene with contextual help before the user disengages or abandons a task.

Teams will also be able to review the sources, reasoning and decision path behind each response. That visibility is intended to give organisations a clearer record of how the system arrived at an answer.

The assistant connects to several existing products in the Amplitude suite. Guides and Surveys can help determine when it should step in and what information it should present, while Session Replay can show what happened before and after a support interaction.

If a case still requires human support, the handoff can include a full case file for the support team. AI Feedback, another tool in the platform, is used to turn support conversations into product data that teams can analyse.

Market shift

Amplitude framed the launch against rising demand for in-product support as businesses look for ways to reduce friction in digital services. The market is also seeing broader interest in software agents that can handle more complex tasks rather than simply answer straightforward questions.

The company cited Gartner's view of that shift in describing the backdrop to the launch.

"Agentic AI systems will act as intermediaries between chatbots and complex processes - revolutionising how users access and use information, handling intricate tasks, executing multistep workflows and providing hyperpersonalised recommendations."

Early users of AI Assistant are applying it to onboarding, reducing support demand and identifying points of friction inside products, according to Amplitude. It did not provide customer numbers or financial terms linked to the launch.

One customer named by the company is Advanced Solutions International, where the software is being evaluated as a way to support users without moving them away from the application.

"Our users don't want to stop what they're doing to look for help or file a support ticket," said Michelle Esquivel, senior manager of global documentation at Advanced Solutions International.

"Amplitude AI Assistant gives us a way to meet users where they are, inside the product, with guidance that feels personal and immediate. We see this as the future of how we can better support our customers," Esquivel said.

Amplitude says it serves more than 4,700 customers, including Atlassian, Burger King, NBCUniversal and Square, and is positioning AI Assistant as part of a broader strategy to tie support interactions more closely to product analytics.