Use MCP to connect Claude and schedule Cowork tasks to monitor the business and resolve customer support issues. You can A/B test your features through experimentation, and the user-level event tracking is especially helpful: you can jump into a specific user and see every event they’re doing, with full details, which is great for QA on events. It’s also easy to start for free with a solid monthly event allowance.

On the analytics side, event segmentation is a fast way to build meaningful charts and answer product questions. You can ask Amplitude AI for insights about your product, use conversion funnels to understand how users convert and track conversion over time, and use Pathfinder to see which events trigger other events. Retention reporting also makes it clear how cohorts are retained over time.

Speed up your analysis with fast charts and dashboards. Anyone in the company will have access to product performance. Clear UI and build together with Amplitude AI Agent. Review collected by and hosted on G2.com.

Charts are limited to a 1-year window. If you want to view 2 years of data, you have to create two separate charts. This also makes it hard to properly measure Conversion events over periods longer than a year, which is a limitation when trying to track annual renewals for subscription apps.

The Block events feature is a great tool for limiting event volume, but it would be even better if you could segment it by platform—for example, blocking an event only for Android. Review collected by and hosted on G2.com.

What I like best is how Amplitude brings product analytics and Session Replay together in one place. We track complex quote and bind funnels across web and mobile, and being able to jump from a drop-off in a chart straight into a replay of what the user experienced makes debugging and prioritization much faster than digging through logs alone. Review collected by and hosted on G2.com.

When event properties aren’t structured consistently across client and server, cross-platform funnels can fail silently—you see drop-offs without an obvious reason until someone audits the payload shape. That’s less an Amplitude flaw than a consequence of how much control you have, but it’s the kind of downside that bites growing teams. Review collected by and hosted on G2.com.

I really like using Amplitude because it’s powerful but still easy to work with. I don’t have to export data to Excel since I can do all my analysis in one place. The charts and visualizations make it simple to explore the data and dig deeper when needed. Features like Session Replay and Cohort Analysis are very helpful to understand user behavior. I also find the Event Segmentation, Funnel, and Retention charts especially useful for my day-to-day analysis. Getting started was pretty straightforward with the help of their documentation. Review collected by and hosted on G2.com.

It isn’t very intuitive or easy to use for a non-technical audience. Many non-technical users find it overwhelming and often need ongoing support from the data team to understand their reports and to create their own. Review collected by and hosted on G2.com.

The ability to rapidly answer questions that would be otherwise time-consuming and potentially require specialised knowledge of tools like SQL as well as being able to easily dig deeper, investigating related information in a way that can easily be visualised and shared with others. Review collected by and hosted on G2.com.

Much harder to find things I dislike about it than things I like. The only thing that really comes to mind is that the ease of use sometimes comes at the cost of being able to perform some complex analyses. Review collected by and hosted on G2.com.

I appreciate Amplitude Analytics' easy-to-use UI, which allows non-technical teams like design and product to access data without relying on us. The segment and funnel charts are a lifesaver for monitoring and analysis, and I find the ability to download data from a chart very helpful. I value the real-time connectivity and data flow because it's easy to integrate and ensures that events are tracked in real time, which helps reduce the turnaround time for anomaly detection. The initial setup was very easy, and we didn't face much trouble setting it up. Review collected by and hosted on G2.com.

In the free version, on a dashboard, we cannot change the time range for all visuals at once. We have to adjust each one manually inside each chart one by one. If we have many charts on a dashboard, it can take some time to load. Review collected by and hosted on G2.com.

I like the interface, how easy and intuitive it is to use, it's integration with Slack, and the monitoring features where I can get important alerts straight into my mailbox instantly. One thing I absolutely love though, is the Amplitude's AI - it is too good to be true especially when it comes to analyzing dashboards. Review collected by and hosted on G2.com.

Even though the performance on creating dashboards with AI is pretty strong, it can only manage some of the basic charts like segmentation, funnel and retention. Dashboards are really good overall, but the personalization could be better, in my opinion. Also, for some mostly used charts, there are information bubbles everywhere which I really like, but some of the others -in my opinion, the less intuitive ones- doesn't have that supporting bubbles, I'd really like to have them. Review collected by and hosted on G2.com.

Amplitude works well as a shared source of truth across marketing, data, and engineering. The same event taxonomy that drives Braze campaigns also powers the analytics dashboards the wider business looks at, so when a campaign isn't behaving as expected, everyone can investigate in the same tool. Review collected by and hosted on G2.com.

Amplitude has been shipping a lot of UI changes recently, which is generally good, but it means features sometimes move around or get reorganised between releases. New users in our team have to relearn navigation more than I'd expect, and some older documentation no longer matches what's on screen. Review collected by and hosted on G2.com.

The funnel dashboards are what I rely on most. In the past, answering a question like where users were dropping off in a flow meant submitting a data request and waiting in someone else’s queue—and by the time the answer came back, the context had usually shifted. Now I can build the funnel myself and iterate on it in the same sitting, which turns what used to be a multi-day round trip into a same-day answer. That move from “request the data” to “answer it myself” has been the single biggest win.

I also like that I can segment a funnel and immediately see which step the drop-off is concentrated in, without writing SQL. It lets me pursue my own hypotheses as they come up, instead of batching everything into one big ask. Review collected by and hosted on G2.com.

Event selection and taxonomy are my biggest friction point. In a large org, there’s a lot of duplication in the event list: several similarly named events for what is effectively the same user action. The picker also doesn’t help you tell which event is canonical or even actively instrumented. As a result, building a funnel correctly depends on already knowing the taxonomy, which is a real problem for anyone new to the workspace. Surfacing volume or last-seen data directly in the event picker would resolve most of this confusion.

Onboarding also assumes you already know the tool. Getting access isn’t the same as being productive. The docs explain what each chart type does, but not which one to reach for given a specific question, so early on I spent more time debating chart types than actually analyzing. A guided “you’re trying to answer X, start here” path would shorten ramp-up considerably.

Discoverability of existing work is another challenge. With a large number of projects and lots of previously built charts and dashboards, it’s genuinely hard to know whether the analysis you need already exists. I’ve rebuilt things that were probably already sitting in someone else’s dashboard. Search seems to rely on you already knowing the right title, and the practical result is duplicated effort across teams. Review collected by and hosted on G2.com.

I like Amplitude Analytics for its ability to provide clear and detailed insights into user behavior, which is crucial for tracking user interaction with different app features. The charts feature makes it pretty easy to derive insights, and the ability to track custom events simplifies analyzing feature usage across the app. The dashboards and visualizations save a lot of time and the real-time analytics through live events is very helpful in measuring success and failure metrics. Review collected by and hosted on G2.com.

Real-time data updates may sometimes have delays depending on the volume of events being processed. Better integration visibility and debugging support for developers would make implementation and maintenance easier. Could improve integration visibility by providing more detailed event delivery logs, including clearer information on whether events were successfully sent, delayed or failed and a more developer-friendly debugging console with real-time event inspection would make it easier to validate event payloads. Review collected by and hosted on G2.com.

I find that Amplitude can definitely be a bit of a challenge to pick up at first. The UI is powerful, but navigating through the advanced performance settings and cohort builders can be a steep learning curve for team members just starting out. I also find myself wishing for more intelligent automation around taxonomy management; keeping data clean as we scale requires a lot of manual lifting that I think could be better handled by AI suggestions. From a pricing standpoint, it can also get quite expensive as your data grows, which means we have to stay really hands-on with monitoring to make sure we're seeing the right ROI. A more guided onboarding or deeper native support for these specific scaling pain points would make a big difference in the daily experience. Review collected by and hosted on G2.com.

While the platform is undeniably powerful, it isn’t always the easiest to pick up. The interface can feel overwhelming for new team members, making the learning curve for advanced analysis quite steep. I also find that as our data needs grow, the costs climb fairly quickly, requiring us to be very diligent with monitoring to ensure we’re getting the right value. Additionally, while the integrations are extensive, managing data taxonomy at scale is still a manual effort, and sometimes the performance can feel a bit sluggish when running complex, high-volume queries. It would be a huge help to see more intuitive, AI-driven guidance built into the onboarding process to help new users get up to speed faster, streamline our taxonomy, and optimize overall query performance. Review collected by and hosted on G2.com.