Sense Chat is Heap's AI, powered by machine learning and large language models (LLMs). Think of it as your own personal analyst: it thinks and acts like one, so you can run deep analyses, understand user journeys, intent, and frustration, and know what to do next in minutes, not hours. No analytics experience required.
Sense Chat includes the following capabilities:
Chat: You can now ask analytics questions in natural language. Sense Chat will choose the right chart, events, segments, properties, filters, and groupings for you automatically. You can ask follow-up questions (e.g. "how does this vary by marketing channel?") and Sense Chat will automatically leverage context from the chat history when answering. Our chat is also the first on the market to directly leverage help center content to provide guidance. Ask a question like "What should I look into next?" or "How do I analyze a feature launch?", and Sense Chat will create a personalized response using the context from the conversation and the full power of Heap's help center documentation, best practices guides, and online training materials.
You can also check how Sense arrived at its answer, giving you full visibility into the logic behind each analysis. This makes it easy to trust and validate the results before acting on them.
Follow-ups: Using the auto-generated description on charts, Sense Chat will identify similar questions asked by others in your account and highlight a set of up to 3 follow-up questions to show alongside each chart. These questions are aimed at encouraging iterative analysis (for example: breaking down a time trend in signups by country or user segment to learn where new users are coming from). With Follow-ups, you don't need to have a question in mind – a teammate can share any chart with you, and in one click, you can chat with Heap to dive deeper into the data. When you share a chart with your team, your teammates will see the same follow-up questions too, making it easy for everyone to start exploring the data independently.
Summaries: When creating a new chart, Sense Chat will automatically attempt to create a title for the chart and describe it in the form of a plain English question that answers "what is this chart trying to show?" while also highlighting key choices like events, groupings, or filters that impact the analysis, without the complexity of showing the same query builder interface used to create the chart. This makes for much more approachable, readable charts when sharing insights with your peers. When you share a chart link with colleagues, the plain-English description means they instantly know what they're looking at, without needing to understand the underlying query or data structure.
How does Sense Chat work?
We can think of Sense Chat as consisting of 3 components:
- A "Question Helper" component that identifies relevant metadata to attach to a user's Input question. It sends the question, metadata, and a set of instructions to (2).
- The "OpenAI large language model (LLM)", which takes everything given from (1) and returns a response with a set of instructions for (3).
- An "Answer Helper", which lives inside the Heap app and has access to the same data available to the authenticated end user. It takes the response from (2), generates a chart, and shows it to the end user in the Heap app.
Here's an end-to-end example of how these components work together:
Let's say the user asks Sense Chat: "How many signups have I had in France this week?".
- The "Question Helper" component of Sense Chat attaches a property to the question, such as the "country" property. This question, the metadata, and a set of instructions is sent to the OpenAI LLM.
- The OpenAI LLM returns a response with another set of instructions that tells the Heap user interface what kind of chart to create. In this example, the instructions would be something like: "create a line chart", "use the country property", and "set the date range to 'last 7 days'".
- The "Answer Helper" creates a line chart using the "country" property that looks at data from the last 7 days. This chart is shown to the end user in the Heap app.
Heap owns and develops (1) and (3), and OpenAI owns and develops (2). We carefully control which data we send to OpenAI. See additional FAQs below for more context on these controls.