SL1 Dashboards

The dashboards guy.

I was sought out by ScienceLogic’s HR and UX teams largely for my experience with dashboards. Since then, I’ve spent five years designing, prioritizing, and refining the data visualization narrative of ScienceLogic’s “New UI” transformation.

TL;DR: I’ve learned:

• A lot about vision, and how to craft a cohesive narrative
• How to prioritize business value alongside user delight
• The value of focus, when so many requests are flying

Background: SL1 does everything.

SL1 is our endlessly extensible platform. It sucks in all sorts of data, collected from servers, networking gear, cloud services, and all types of devices. It produces events and derives status, especially in layers of business-value abstractions. Our biggest customers manage dozens of SL1 deployments, each of which monitors tens of thousands of devices, which produce hundreds of millions of rows of data over time… but that still needs to produce something humanly understandable and manageable.

SL1 has been a complicated and powerful platform since it was started 20 years ago, and the traditional customer base of Managed Service Providers (MSP) like it that way. The more recent goal to take over the Enterprise and Federal markets, though, is an entirely different beast. I was hired as a key part of a tiny skunkworks team that would rethink the entire user experience for modern user needs and modern markets, and we identified dashboards as an immediate target.

Past: Sources and views.

Rewind: 2017. As the only UX Designer at the time, I worked daily with the mercurial but brilliant chief architect on this “New UI” transformation. We systematically broke down the types of important monitoring data that our customers collect and store on the platform. Identifying the primary data types, shapes, relationships, and visualizations enabled us to put together a surprisingly composable set of components into a powerful IT monitoring tool.

Data “sources” closely mirror the customizable data model. The “views” are the visualizations. “Filters” are pluggable, and “selections” connect widgets together. Everything else is just “nerd knobs” - toggles, options, customizations.

My team is responsible for this framework and core features, building blocks that enable users to visualize the answer to their question. I’m proud of the modularity that gets the most flexibility our of our development dollar. But I have also made sure to unify our external dashboard narrative alongside internal “customers”: SMEs who deliver dashboard content for each hybrid-cloud technology, as well as Professional Services who customize and deliver to spec.

Past: Taking the helm.

It seemed as soon as we got Dashboards shipping, the Product Manager called me. “Why don’t you be a PM? That’ll let me focus on these other areas.” I decided I’d prefer the decision-making over the mockup-making. (The diagram-making and whiteboard-sketching never went anywhere.) I had been moonlighting as substitute PO and Scrum Master anyway, so I learned the rest on-the-fly. Oh, I’m immediately taking over two scrum teams building some of the most foundational parts of the monitoring system: Devices, Events, and, yes, Dashboards. Ok. Let’s jump headfirst into 3 of the 5 main nav icons.

Today: Strategically maturing.

Almost four years later, these are some of the most mature areas in our product. The Product organization is also maturing, and I have ceded day-to-day PO management of this scrum team to a newer colleague. I get to keep guiding the high-level direction of Dashboards, and still often make a call, but also get to spend more of my time coordinating cross-team-level features. I would love even more chats with customers to broaden my insight, and fewer miscellaneous fires (with release management, sales, support, customer success, operations, marketing), but my comprehensive knowledge of the product compels me to confront whatever the challenge is today.

Future: Constant evaluation.

There are so many different types of dashboards that could suit different customer needs and use cases. I’m continuously gathering feedback from a diverse set of stakeholders, outlining narratives and potential capabilities. We have precious few scrum teams for the amount of ambitious things to build, so while “maintenance mode” isn’t quite right, there have been a lot of other projects prioritized over making these core features even better.

My latest vision? Recently, I’ve gotten buy-in that the next great dashboard shouldn’t be a dashboard at all.

SL1 Home Page is finally being born after years of unrequited plans and halted half-steps. If the goal for our future growth is Simplify SaaS for the Enterprise, we need something approachable and personal, much more than our past market required. I’m currently working closely day-to-day with our top UX and Engineering teams to build and validate this high-visibility tentpole feature.

Evaluation: Panacea plight.

If I have to be critically self-aware, dashboards may never escape their “flexible” and “powerful” underpinnings. There’s still an organizational imbalance compared to “make it obvious how to build one”, not to mention successful end-to-end delivery of prebuilt content “out of the box”. I do think it would be different without the looming everpresent comparison to the Classic UI Dashboards that we may never fully replace, but there may simply be too many asks to fit in one feature coherently…

• Jump in and learn to build quickly
• Share a really cool insight with colleagues
• Schedule an exported report to be emailed
• Deep-dive into troubleshooting an issue
• Elegantly display CTO-level status on a non-interactive TV
• Heck, develop your own custom code to drive entire workflows

“Dashboards”, I’ve learned, means all things things to all people.

Lessons yet again…

• A lot about vision, and how to craft a cohesive narrative
• How to prioritize business value alongside user delight
• The value of focus, when so many requests are flying

That last one. The big one. Focus. When you have too many projects going on, building a “v1” is often all that’s funded. Maybe investment in engineering R&D grows as quickly as ambitions do, but someone might just have to be disappointed.

As my leadership role grows, I am grateful to have learned how to say “no” in order to make a kick-ass “yes”.