The Observability Cloud

What do you like best about the product?

I like that Observe by Snowflake brings logs, metrics, and traces together in one place. The dashboard makes it super easy to monitor everything, and the worksheets and datasets let me dig deeper when I need to investigate an issue. I also really like being able to correlate different types of data, which makes troubleshooting and identifying the root cause of a problem much faster. It takes away a lot of the time I usually spend looking for information to solve problems with my systems. The initial setup was very simple and straightforward.

What do you dislike about the product?

I think Observe by Snowflake can be a little complex. The learning curve is quite steep, and I think everything could be easier to navigate.

What problems is the product solving and how is that benefiting you?

Observe by Snowflake takes away a lot of the time I usually spend looking for information to solve problems with my systems.

What do you like best about the product?

Observe by Snowflake is an AI-powered observability platform built natively on the Snowflake AI Data Cloud to unify logs, metrics, and traces.

Key Capabilities

• AI SRE: Automates troubleshooting, root-cause analysis, and agentic workflows using natural language via the UI, CLI, and MCP Server.
• Observability Context Graph: Maps service semantics and relationships to reason about complex distributed systems and AI agents.

What do you dislike about the product?

While Observe by Snowflake provides cost savings and deep data integration, users and industry analysis highlight several drawbacks, limitations, and growing pains with the platform:

1. High Tech Debt and Internal Scaling Pains

Following its acquisition by Snowflake, the platform has faced standard post-acquisition integration hurdles. According to employee reviews on Blind, internal scaling has led to significant technical debt and legacy design issues that engineering teams are actively trying to resolve.

2. Rigid UX Focused Strictly on Observability

A common critique revolves around the scope of its interface. On Reddit, opinions are mixed regarding its core architecture.

What problems is the product solving and how is that benefiting you?

This platform unifies your system logs, metrics, and traces into a single telemetry lakehouse.

The problems it solves

• The “data silo” tax: Traditional monitoring often forces you to bounce between separate tools for logs, dashboard metrics, and APM traces, which makes root-cause analysis slower and more fragmented.
• Massive ingestion costs: Many traditional observability tools charge exorbitant rates to ingest and store large data volumes, pushing teams to aggressively sample data or delete historical records just to control spend.
• Slow incident resolution: When a microservice or an AI agent breaks in production, manually hunting through unclear or unmapped dependencies to understand why it failed can take hours.

What do you like best about the product?

The clean interface and having logs, metrics, and traces in one place. Queries stay fast on large data volumes, and the AI-assisted troubleshooting surfaces likely causes quickly, so we spend less time digging during incidents.

What do you dislike about the product?

The learning curve for its query language can be steep at first, and some integrations take extra setup to get the data shaped the way you want. Pricing can also be hard to predict as data volumes grow, so it takes some monitoring to keep costs in check.

What problems is the product solving and how is that benefiting you?

It solves the problem of scattered telemetry. Instead of piecing together logs, metrics, and traces from different tools, we get one connected view, so we find root causes faster and resolve incidents sooner. It plugs into our existing stack without much hassle, and the onboarding and support helped us get up and running quickly. Overall, less time firefighting and a better return on what we spend.

What do you like best about the product?

Instead of forcing you to hunt through disconnected silos of logs, metrics, and traces, Observe links them into a relational model. It maps technical telemetry to actual business entities like shopping carts, user sessions, or customer accounts. This turns troubleshooting from a frustrating text-search problem into an instant graph-traversal exercise

What do you dislike about the product?

While the UI relies heavily on a streamlined visual layout, users note that unlocking the platform's advanced workflows, deep customizations, and unique metrics still requires writing SQL and manual data engineering. Teams transitioning from traditional, point-and-click tools like Datadog often face a steep learning curve to grasp how to manually format and tune the platform's dataset models.

What problems is the product solving and how is that benefiting you?

• The "Three Silos" Problem: Traditional monitoring treats logs, metrics, and traces as three separate buckets of data. When an outage occurs, engineers waste critical minutes manually matching timestamps across different tabs.
• The Observability Tax: As applications generate more data, incumbent legacy platforms charge skyrocketing, volume-based ingestion fees. This forces companies to throw away valuable logs to stay under budget.
• Context Disconnect: Engineers often see that a server's CPU is spiking, but they cannot easily tell which specific customer account or shopping cart session is causing the traffic.

What do you like best about the product?

What I like about Observe is the potential to make troubleshooting easier while keeping costs down. The AI capabilities are interesting, but I think the real value is how quickly teams can identify and resolve issues without having to jump between multiple tools. Of course, that depends on how well it performs in practice.

What do you dislike about the product?

One thing I don't particularly like about Observe is the learning curve. I understand that observability is inherently complex, but if you're building a tool that's supposed to make troubleshooting easier, why does it take so much effort to get comfortable using it? The whole point should be to reduce complexity, not introduce another layer of it. I shouldn't need to spend hours understanding how the platform organizes and connects data before I can start getting meaningful answers. AI is great sure, but I question how useful they really are if users still need significant technical knowledge to validate what the AI is telling them. For me, a good observability platform shouldn't just be powerful; it should make complicated information easier to understand and act on. Otherwise, we're just moving the complexity from one place to another and calling it innovation.

What problems is the product solving and how is that benefiting you?

Observe helps address the challenge of having operational data scattered across multiple systems, making it difficult and time-consuming to identify the root cause of issues. By bringing logs, metrics, and traces into one platform, it makes troubleshooting more centralized and gives teams better visibility into what's actually happening. The biggest benefit for me is the potential to reduce the time spent investigating problems. Instead of jumping between different tools and trying to piece everything together, I can get a clearer picture of where things are going wrong. I also appreciate the potential cost savings, especially as the volume of data grows. That said, I think there's still room to improve the learning curve and make the platform more intuitive. A tool designed to simplify troubleshooting shouldn't require so much effort to learn in the first place.