Skip to content

2024 Q1 & Q2 Development Update

Q2 2024 Work In Progress

Automated Learning

We're currently testing and refining a new automated unsupervised learning process which identifies normal behaviour. We've always offered unsupervised learning based on user provided trained data. The new automated unsupervised learning process doesn't require trained data sets and autonomously learns without the need for any user input. Just sit back, relax, and let the system learn!

What’s Up?

We’re proud of our service availability and realised we weren’t showing off as much as we could, so we’ve added a new Service Status page which keeps track of how awesome we are (and will also transparently let you know in the unlikely event of something failing). It’s work in progress but already available in its basic form at

What The Flock

We’re adding the Flock platform to our alert integrations.

Q1 2024 Work Completed

Metric Search API

It felt like an obvious feature we’d overlooked, so we added it. The new metric search API method works a little bit differently to the way you could already select metrics to create an alert - it’s more flexible and designed more to help you find things than to select and group them. It’s available now via the new API method /metrics/search and also in the latest release of our Dashboard. If you’re not set up with an account yet, you can still check this out on our live demo via

We plan to expand this functionality in the future to be able to search other items and events in your data as well as metrics.

API documentation for the new method is at:

Super-charged Metric Graphs

Previously our Dashboard was loading interactive graphs by default, when in many cases that interactivity was not required. We’ve changed this to load static graphs by default, with the option to see the interactive version if you want to tweak. The result is that pages containing graphs now load a lot more quickly across the Dashboard.

New Dashboard Progress Indicators

The analysis and processing that happens behind the scenes is complex and sometimes takes a while to complete. As much as we always aim to optimise and refine it, hard work still takes time and until quantum computing is more readily available, there’s no way around that.

However, we’re trying to improve the way we communicate the details about these longer-running tasks so that you’re never confused about whether something is working or not, or how long you might need to wait (or decide to come back later).

We’ve implemented a progress feature where the back-end tasks can maintain an ongoing communication with clients as work progresses, and give estimates about time left, which will be displayed as a progress bar, or an email on completion, or triggering some other activity to run when it’s done. This is implemented in our API and rolling out in the Dashboard and back-end services currently.

General Performance Improvements & Minor Bug Fixes

  • Added the ability to fire off a test alert when you make changes to it. Previously test alerting was only available when you created a new alert.
  • Fixed an issue where previously chosen Slack alert recipients were not being populated in the edit form if you made changes.
  • Lots of loading speed improvements and caching optimization across the board