GitLab Manage:Analytics Team Meetings, 14 Sep 2020

Previous Meeting

Next Meeting

⏯

youtube image

►

From YouTube: Manage:Analytics category discussion

Description

Nick sketches out his thoughts on how Manage:Analytics should segment their categories for improving customer value

A

Hey team: this is nick post from the manage analytics group and I want to run and uh get some feedback on these new uh category segmentation that I have that I've put together.

A

So uh we have three categories: two of them are existing already and that's enterprise, devops reports and value stream management, and then we have one new one which I've identified and I'll dive into each one of these: uh to increase the uh specificity or or focus of of the categories. What I've done is I've added some um personas and needs and outcomes that we want for each of these, uh these categories and I'll dive into each one of those in a sec.

A

So, let's start out with devops reports and with devops reports, uh the target persona will be an app development director and the key needs that we want to uh that. We want to satisfy the the key outcomes that we want to help dakota. The app development director accomplish is uh syncing.

A

Developer needs with business goals, so that is helping dakota, uh communicate business goals with the team and then track developers progress against these goals, so we have a better understanding of how the developers are performing and making sure that everyone is driving towards the same goal.

A

Number two is helping dakota track their devops transformation, so that is understanding how dakota's teams are performing uh when it comes when it comes to adopting new tools, processes and- and things like get lab and in turn, also understanding how their work has improved because of that sort of adoption and maturity from their devops transformation and then finally, is helping uh dakota get a better understanding of team performance holistically at an instance level and understanding where there's opportunities for improvement by identifying outline teams, understanding um how we can improve and then find if there are any interesting learnings that we could have from from all the teams.

A

So some of the outcomes that we want to be driving for dakota based on these these needs is improving the businesses uh satisfaction with dakota.

A

So the dakota will liaise with the business quite regularly and and report to uh report to the business, as well as as leadership, so improving their satisfaction with with what uh the app development director dakota is doing, um helping to drive uh value through predominantly cost savings, and then also helping to drive gitlab adoption, which is a proxy a little bit for this cost savings outcome and then, finally, the the focus of the the data that we'll we'll uh be looking at for uh for dakota.

A

The the app development director is uh instance, level, and some of the feedback we've heard before is projects and groups, don't necessarily map to teams within large enterprise organizations, and what we want to do is help to provide a little bit more nuance on identifying who teams are, and this is why we'll introduce something called segments which helps us to get a little bit more focus on on what teams, uh what teams are, what and and what they're working on.

A

So that is, enterprise devops reports.

A

Second is team, devops reports, and this is focused specifically on uh delaney, the team development leader, uh dev manager and, ultimately, the needs of uh delaney are general purpose analytics and database tools around some of the most popular uh workflows within gitlab. So this is things like um issue management, mrs this sort of stuff, so getting a better understanding of how delaney's team is performing and uh helping him to ultimately drive product, help productivity and improve team performance, and then finally, we don't want uh just a bunch of of dashboards that people go go to.

A

We want this to be an essential part of delaney's workflow and his team's workflow. So what we want to do is take some of the the analytics and dashboarding and data visualization and put that in context of some popular user workflows as well.

A

So, for example, talking about putting metrics within issues and mrs providing insight on on how, as a developer, is working on something they can become more productive or how they can improve the quality of that specific, mr or issue.

A

So the key outcomes that delaney wants to drive are productivity and quality, and uh this is what we think uh by focusing on on the two above uh capabilities: we can, we can drive for delaney and then the areas that we'll be focusing on with the gint lab is the project level and group level views.

A

This is focused specifically on where teams teams are working, and I think this was where these these two levels will provide the most value for for delaney and his team, and then finally, we have uh value stream management, and this is focused specifically on the target persona of parker, the product manager and uh the three capabilities or needs that we're trying to satisfy uh with value stream.

A

Management for parker is to uh one map and visualize the different stages of parker's value stream and through this visualization, provide better insight into how the value stream is working and then number two is uh once we have mapped out the different stages of the value stream and started to visualize it. How do we start to identify areas of waste and bottlenecks within this value stream?

A

That parker and team can work on in order to better improve their cycle time and better improve their value stream as a whole and then, finally, once we've mapped and visualized and started to identify bottlenecks, how do we use the insights and analytics that are coming from value stream management now to inform and automate our planning workflows so we'll have to work very closely with the planning team on this one, but value stream management has a lot of synergies with this and it'll help to increase productivity.

A

uh Quite a lot if we, if we use some of these insights to to to automate some planning workflows. So the key outcomes that we want to be impacting on this is is uh cycle time cycle time is uh obviously very very related with uh productivity and quality from uh from the previous from the previous category.

A

But this is the key metric that we're trying to try to improve and again uh we'll be focusing on the uh project level and group level here for parker, uh and this is where we think we'll find the most value from uh from value stream management.

A

So there we go uh three different categories: enterprise, devops reports, team, devops reports and value stream management. Enterprise devops reports is for the large organization specifically focused on a director and team. Devops reports is focused on uh development manager and team lead and value stream management is focused on a product manager. Now um I've segmented them in this particular way, because there's a lot of synergy that we have from from working on these three things simultaneously.

A

For example, we will want to provide workflows where we can take the views from enterprise, devops reports and drill down into things like team devops reports, so we'll see how a specific team is is doing, and then we can drive from enterprise devops reports um into team devops reports and understand, specifically why a team is performing that way in a lot of detail, and that also provides an interesting connection as well for um creating uh some business goals within the enterprise, java's reports and cascading those down to team devops reports and then tracking those as a team and those uh results.

A

Aggregating backer up to uh enterprise devops reports as such and um also there'll, be plenty of insights that we have from value stream management that can be fed into team devops reports um in context, actionable insights and, and also we can aggregate some of the the content that we have from value streams and use that um to inform enterprise devops reports. So there's there's a lot of um interaction that we have between these these three categories, but by segmenting them.

A

In this way, I think, will provide a a better focus overall on the user experience and provide more value for for gitlab's customers as a whole. So I'd love to get your feedback on this. This is, as you can see, still just a rough napkin sketch, um but uh yeah happy to hear lots of feedback and collaborate on this going forward. So thank you very much for listening with me.