Unified Logs and Metrics
Every process and action on every component (e.g., servers, desktops, networks, applications, mobile, IoT, security devices) generates huge volumes of machine data that gets captured as logs and times series metrics. Users can extract valuable intelligence from customer's log data combined with time series metrics. When viewable in real-time, this data provides a deeper insight into application needs and new customer opportunities. Logs and Metrics in a single platform, in real-time provides continuous intelligence, while being predictive by analyzing the past present and being predictive for the future.
Original Metrics logo
I wanted to put a face to the name of Metrics by creating a brand/identity that can represent Metrics from a Product and Marketing perspective.
Early stage design explorations
Medium to hi-fi design mocks for Metrics, focusing on the Query builder tab, and the interactions when constructing a query. The goal here was to make the interface more usable, while enhancing the visual aesthetic. I collaborated with my colleague, Victoria Wong who was responsible for providing the UX wireframes.
Light-color version for multiple cards stack
User Testing / Personas
We conducted user testing session to validate some IxD decisions in the initial design mocks. This also enabled us to narrow our target audience, and define our common users. Most of our participants were located remotely, so we used GoToMeeting and shared a prototype that enabled these users to click through and give us real-time feedback as they clicked through the prototype. We also had prepared a list of questions that we asked the users after each user testing sessions that provided great insight into our product at that specific stage.
Our target user is "Clayton", who is a Dev Ops Engineer and is considered to be a SumoLogic "power user".
Below is a rough user flow for setting up Metrics for a power user, such as "Clayton".
The outlier operator takes numeric time series and figures out if a particular segments deviate from the standard. It detects situations like spikes or step-value changes, and the user is presented with very nice visualizations that show outliers & confidence intervals.
The legend is designed to assist the user in accurately focusing on a specific time series or outlier. The Outliers legend enables the user to switch views from Sensitivity to Ranking, as well as filter the results by showing the number of outliers shown on the chart. In the Metrics legend, the Metrics types and Times Series lines are clearly separated and easily accessible for the user to drill down and see a specific instance.
Tool tip and Scrubber
The tooltip was designed to be fixed on the vertical y-axis and only move horizontally along the x-axis, to prevent the tool tip from jumping up and down as the users scrubs across the chart. Once the user clicks on the chart, the scrubber becomes locked and allows the user to vertically hover along the fixed y-axis to reveal all the points. This locking feature also enables the user to independently hover below to the legend to find the corresponding time series without losing the position on the chart.
Metrics Marketing Comps
I wanted to present metrics from a Marketing perspective to create a feeling that would translate into a story to capture the user/product.
Landing page concept