The purpose of signals is to proactively notify you about important changes in user behavior so you don't have to manually dig through reports. By automatically surfacing issues and opportunities, Novus helps you quickly understand what is happening in your product, why it matters, and where you should focus your efforts to improve the user experience.

Signals

Signals are automatic, data-driven alerts about your product's performance. Novus analyzes your data, such as funnel drop-offs and metric trends, to highlight areas that need your attention. You can view overall performance through key metrics, review specific signals linked to pages or features, and let Novus investigate and propose fixes for you.

Key metrics

Key metrics give you a quick summary of your top-level product performance. These include data points like visitor counts or feature adoption rates. Each metric displays its current value, a brief context summary, the trend direction (up, down, or flat), and the percentage change compared to the previous timeframe. This helps you monitor the overall health of your product at a glance.

Supporting Signal data

Signals categorize specific patterns in your product into three types:

Each signal is assigned a priority (high, medium, or low) and is linked to the specific pages, features, or funnels it affects.

Investigate signals

When Novus detects an issue or an opportunity, you can initiate an investigation to find the root cause and generate an actionable plan. Novus analyzes the data on your behalf to propose specific code or configuration changes.

  1. Select a specific signal from your list to view its details.
  2. Select Investigate Signal. Novus will review the associated data and generate an Investigation Plan.
  3. Review the generated plan, which includes a diagnosis, root cause, proposed actions, and estimated impact.
  4. (Optional) Select Edit Plan to make adjustments, then select Save Changes.
  5. Select Approve & Fix to approve the plan. Novus will then implement the changes, which may include creating a pull request for your development team.

Note: Informational insights cannot be investigated.

Source artifacts