What is our primary use case?
The primary use cases for Unomaly involve all kinds of things. It's a rate anomaly, error anomaly, it could be anything. Any kind of anomalous pattern can be detected.
Unomaly's anomaly detection capabilities contribute to maintaining system reliability. We cannot find all errors humanly. We cannot configure every possible threshold. In the new world of intelligence and AI, we need to have this intelligent way of finding out the anomalies. It is helping us. It's a combination of things. Not everything will be done through the anomalies. A lot of things are still configuration-driven, threshold-driven. This is a layer on top of that.
What is most valuable?
Unomaly's anomaly detection capabilities contribute to maintaining system reliability. We cannot find all errors humanly. We cannot configure every possible threshold. In the new world of intelligence and AI, we need to have this intelligent way of finding out the anomalies. It is helping us. It's a combination of things. Not everything will be done through the anomalies. A lot of things are still configuration-driven, threshold-driven. This is a layer on top of that.
What needs improvement?
We do not use its Contextual Insight feature. We haven't explored the LLM side. That part wasn't GA. They've recently launched it. The agentic AI feature has not been tried yet.
I would need to check with the team about what could be improved in Unomaly as I don't work with it directly.
For how long have I used the solution?
I have been working with LogicMonitor for almost a year now.
How are customer service and support?
I have contacted their technical support quite often, and they are good.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
We are internally using DataDog and LogicMonitor primarily instead of anything else.
What was our ROI?
I have had experience with the pricing as we have done all the negotiations before coming to a conclusion. We are satisfied with it.
I find it reasonable compared to other products in the market, which is how we've come to this conclusion.
Which other solutions did I evaluate?
I might be exploring it for some customer requirements.
I have experience with Dynatrace from the last decade, not in recent times.
We are internally using DataDog and LogicMonitor primarily instead of anything else.
What other advice do I have?
I have been using Unomaly or LM Envision by LogicMonitor for a year for internal purposes.
I personally don't use metrics to evaluate Unomaly's performance as I have a team who handles that aspect.
The endgame has moved towards agentic AI. Two years back, it was supposed to be the endgame with ML and prediction anomaly. The world has moved on. Having Unomaly, even the best anomaly doesn't make too much of a difference. The endgame is now about the metrics of autonomy rather than anomaly. What is the degree of autonomy? What is the return on autonomy? Those are the metrics I'm more interested in than just having the anomaly. The world order has shifted, and the KPIs have shifted.
They already have Gen AI and agentic AI features, but we haven't used them so far.
I will continue to use it in the future for now as it's only been a year. We don't want to change anything internally for now.
I would recommend Unomaly to other customers because anybody using observability can and should use Unomaly in the new world.
I can't think of any types of companies I would not recommend it to because observability cannot exist without Unomaly nowadays.
On a scale of 1 to 10, I rate Unomaly a seven.
