Get an early look at features in development. Our Early Access experiences let you try alpha-stage features that are still evolving, while our Beta programs offer more polished experiences that are almost ready for general release. Your feedback at both stages helps shape the final product.
Try features in development and share feedback to guide their final release.
As part of our beta program, our users can explore newly developed product capabilities while providing feedback. Beta experiences are stable enough for everyday use but still evolving.
Beta releases may be restricted to a select number of Honeycomb users (private) or open to all Honeycomb users (public). If you would like to participate in a beta, get in touch with your Honeycomb point of contact. If you are not sure who that is, Honeycomb Support can help you connect with them.
Status: Beta
Type: Private
Dates: mid-July–December 2025
Metrics 2.0 is a major update to Honeycomb’s metrics experience, built to align with the OpenTelemetry specification for metric types and behaviors. This upgrade introduces a more expressive query model, advanced temporal functions, and full support for alerting via Triggers.
This update enables:
RATE(), INCREASE(), SUMMARIZE(), and LAST() to get deeper insight into your time series data.To explore the beta in more depth, check out the Metrics 2.0 Beta Enablement Guide.
We’d love your feedback, especially on usability, results, and any rough edges you encounter! To opt in or share feedback:
#discuss-metrics channel. Status: Beta
Type: Public
Dates: In progress
Want to see how your teammates are using Honeycomb? Activity logs give you visibility into changes made to key resources in your environment—whether by a person or an automated process.
With activity logs, you can track updates to:
TriggeredTriggeredYour team can explore this data in a dedicated Activity Log Environment, which includes a set of datasets that capture activity for each resource type. Within this environment, you can use all of Honeycomb’s features—like queries, visualizations, Boards, and Triggers—to investigate activity and uncover patterns or issues.
Try experimental features at the earliest stage of development and help shape them from the ground up.
Our early access program provides alpha-stage features to a small group of users for feedback and validation. Early Access features may be incomplete, change significantly before release, or be removed based on what we learn.
Participation is typically by invitation or request. If you’re interested in joining an early access program, get in touch with your Honeycomb point of contact. If you are not sure who that is, Honeycomb Support can help you connect with them.
Status: Early Access
Dates: In progress
Anomaly Detection represents a fundamentally different philosophy in Honeycomb. Rather than requiring teams to predict what might go wrong and configure alerts accordingly, our system learns your service patterns and proactively notifies you when behavior deviates from normal.
When you combine this with BubbleUp, which shows you the outliers around an anomalous event, you can get a very detailed picture of both the what and why on your way to understanding the root cause.
The system is designed with opinionated defaults that work out of the box, while still providing the tuning capabilities teams need. It is optimized for accuracy, ensuring that when you get an alert, it’s worth investigating.
To learn more about joining the early access program for Anomaly Detection, contact your Honeycomb customer success manager.
Explore which features we’ve tested in the past and how they shaped our product today.
An archive of our past beta programs. These features were nearly production-ready and were refined based on user feedback before graduating to general availability.
Type: Public
Dates: July 16 2025 to September 9 2025
The Model Context Protocol (MCP) enables AI assistants to directly interact with your Honeycomb observability data through natural language. This open standard allows seamless communication between AI models and Honeycomb, transforming how you investigate issues and analyze system behavior.
With MCP, you can:
MCP maintains context across interactions, enabling sophisticated multi-step investigations that would typically require deep Honeycomb expertise.
To explore Honeycomb MCP in more depth, check out our Honeycomb Model Context Protocol (MCP) documentation.
Type: Public
Dates: August 20 2025 to August 28 2025
Text Panels let you add narrative context directly to a board, alongside your queries. Use them to explain what the data means, highlight key insights, or link to runbooks and playbooks so your team knows what action to take.
You can format Text Panels with Markdown, making it easy to structure and style your content with headings, lists, links, code blocks, and more.