Gitlab hero border pattern left svg Gitlab hero border pattern right svg

Data Team


With the migration from the previous BI tool to Periscope, we took the opportunity to review with stakeholders across the business how they were engaging with the data team and the reports it produces.

Further context and notes from each call can be found on the relevant issue

Summary of findings

There is no single universally accepted list of what metrics matter now

While the KPIs Page exists, this is not 100% accurate representation of what the SLT considers their most important metrics at the moment (I'd give it an 80%). Part of the reason behind this is that as focus/projects change so does the importance of certain metrics over others.

The primary use of Periscope for most of the participants was to populate their monthly metrics call

There is(was?) an OKR being defined that all OKR tracking should be done through Periscope, but really this should be aimed towards making the monthly metrics calls run from Periscope.

OKRs at Gitlab extend the primary stream of work instead of documenting it

OKRs are important, but they do not generally represent what people are looking at on a day-to-day basis. Any work focused on reporting/dashboarding for teams should not begin from the premise that OKRs are the place to start.

The Data Team is larger than its official members

As noted by @emilie, people with "Operations" in their title are generally doing Data Analyst functions, people like Francis and Robert come to mind (although there are many others across the organization).

This brought to mind the Spotify Squads/Tribes/Guilds Framework, and that we need to work on how to better integrate with the "Extended" Data Team.

We should consider how to take a step closer to Salesforce

The sales team does not want people getting their data somewhere else, and will not be satisfied with a Single Source of Truth outside of SFDC.

This comes from the need to keep people in the tool so that activities get correctly logged and data of the process is captured how it should be.

The business also has serious problems with Data Quality on Salesforce.

While embedding Periscope into Salesforce is a good first step, we need to consider how to help the business get into a place where SFDC reports work just as well and where there is a Data Quality process.

Everyone wants Product Usage Data related to their metrics

The Product Team needs deeper insight into product usage, but the rest of the organization would be satisfied with feature usage.

I could not find a place that has a full inventory of all Features tied to a Category and to a Stage. Getting to a place where that information exists and is tied to a "usage" definition would be a huge first step in the right direction.

There is a big appetite for Training, but not a lot of clarity on where to start

The interest and willingness to participate in training are there, but for the non-analysts the topics are daunting and there is no clear roadmap on to get from 0 to exploring data on Periscope.

Recommendations from these calls:

Should-haves in the future: