June 2020 APM telemetry update
Hi all, I wanted to provide some updates on the performance indicators (Formerly known as North Star Metric) we are collecting using snowplow for our .com solution.
As you are aware we track several leading indicators in order to understand customer usage and adoption
- The leading indicator for APM AKA Performance Indicator (PI) is the combined page views of metric and log (rationale can be found here),
- We measure only the page views which contained data (we don't count empty results).
- I am filtering out our testing project so if you maintain projects which uses our solution on .com please let me know so I can filter them out, (this will help our numbers be more accurate).
- The numbers you see below are .com only, we know that the majority of our users are on self-managed, we are unable to collect those same metrics using usage ping (Self-managed), ETA for having this available is Q3.
- We measure both page views (by session id) and by domain ID which is as close as we can get to measure unique users.
- Numbers represent the month of June.
❓ So how are we doing?
Performance indicator +75% MoM
An increase of 75% of pages views for metrics and logs from last month.
You can see this metric over time trending upwards.
Let's break this down further
📈 Metrics
- Total number of page views 772.
- total Unique users 413.
Default dashboard views
- Total number of views 298
- total Unique users 199
Custom dashboard views
- Total number of views 484
- total Unique users 260
📝 Logging
- Total number of views 967
- total Unique users 299
Logging is on fire
Insight
- Overall our PI is constantly rising, which means users are using our solution more often
- Metrics page view trending down for the last 2 months.
- We can see that the decrease was mainly driven by the fact that fewer users are using our default dashboard
- We see that users are trending toward using custom dashboards
- Logging is constantly rising, and for the first time it had more views then metrics.
- Metrics has more unique users then logging, however, logging is used more often.
Edited by Dov Hershkovitch