Amplitude Digital Analytics Platform

Amplitude is an excellent product analytics interface, but what really sets it apart from every other tool is the sophistication in how it collects data. In particular, the 'groups' feature allows you to understand a property at the time of an event easily, e.g. a user's subscription tier.

I couldn't tell you as we can't use it yet.

1. If a significant peak is observed in the chart, you can right-click to quickly access the behavior of the specific user involved and view their full user journey. This feature makes it easy to understand the causes and context behind unusual user actions, which is highly practical for analysis. 2. The ability to create cohorts simplifies the comparison of performance across different groups of users and is one of the features I rely on most often. 3. The platform allows you to easily combine different conditions across multiple events via drag-and-drop to create new custom events for analysis. This greatly simplifies the data analysis process. Even when new features are released and new events are added, there's no need to be overly concerned about event compatibility before and after the release - the tool still enables effective analysis.

Amplitude handles user properties and what they call 'groups' (e.g. account) by enriching events on ingestion. This is an incredible feature as it handles slowly changing dimensions gracefully - making it very easy to know the state of a given property at the time of the event. The catch with this is that you can only query this data if the user has been recently active. Want to find a list of users that currently have 'x' property, regardless of whether they've been active or not? You'll probably need to go to your data warehouse or CDP. Session replay, while a great feature, is less mature and does cause us a number of management headaches. There are some odd requirements around defining session IDs that have caused the feature to break a few times.

Sales, implementation, and account management.

1. The data sync back to the database often takes much longer than expected - sometimes exceeding 2 hours, and in some cases even over 24 hours before the data becomes available. This causes significant issues for timely analysis. Although the issue doesn't occur frequently, it has been raised multiple times but hasn't been effectively resolved. 2. It would be great to have a user interface that supports one-time data uploads (e.g. json.gz files), the ability to create datasets and manage or delete them. This would allow users to leverage Amplitude's charting and analysis capabilities even for standalone or ad-hoc datasets. 3. It would also be helpful to have the ability to create alarms or alerts directly through UI to monitor anomalies in charts more efficiently.