// Project Goals
We set out to design Resolve, an AI platform that automates the heavy lifting across complex IT ecosystems. The real challenge was making it accessible. We wanted a clean, user-centered interface with a simple drag-and-drop workflow builder that gives users immense power without overwhelming them. Design does not exist in a vacuum. The platform had to be API-first, sliding perfectly into tools like ServiceNow, Jira, and Splunk without breaking existing workflows.
We relied on deep user research to build AI insights that actually mean something, creating a system that learns, predicts, and transparently explains its decisions so IT teams can fix problems before they happen.
// The Result
Good design is measurable. When Resolve launched, it did exactly what we designed it to do, which is giving people their time back. We saw a massive 38.4% bump in efficiency, turning hours of tedious IT work into minutes. By cutting out the busywork, the platform hit a 220% ROI in year one. The users felt the difference immediately. Satisfaction jumped by 52.2% because the workflow builder and real-time analytics simply made sense. We also made onboarding completely frictionless with pre-built templates, and that ease of use drove a 27.5% spike in new subscriptions right out of the gate.
+27.5%
User activation rate
+38.4%
Monthly active users
220%
Transaction completed
+52.2%
Customer satisfaction
Making AI Useful Before Making It Impressive
Resolve was designed around a simple problem: IT teams spend too much time on repetitive operational work, yet automation tools often introduce a different kind of complexity. The opportunity was to create an AI-driven platform that could automate meaningful parts of the IT workflow without asking users to become automation experts first.
The product needed to feel powerful enough for experienced technical teams, but clear enough for users with very different levels of technical confidence. That meant avoiding the temptation to make AI feel magical or mysterious. The focus was on usefulness: helping people understand what was happening, what could be automated and what the system was doing on their behalf.
Turning Complex Automation Into Something People Could Actually Use
The core UX challenge was translating sophisticated backend automation into an interface that felt direct and understandable. We designed the product around a clear dashboard, configurable automation rules and a visual workflow builder that made complex processes easier to create and manage.
Instead of forcing users to think in terms of scripts, dependencies and technical infrastructure, the experience gave them a more visual way to define what should happen and when. Real-time status information, system health indicators and drill-down views helped users understand both the overall picture and the details behind individual processes.
The goal was to give experienced users depth without making everyone else fight through it.
Designing Around Real IT Workflows
The platform was shaped around the tools and processes IT teams were already using rather than asking them to replace everything at once. Resolve needed to work alongside systems such as ServiceNow, Jira and Splunk, so integration became a core part of the product experience. An API-first architecture, pre-built connectors and reusable templates helped reduce the effort required to get started and made automation feel like an extension of existing workflows rather than a separate ecosystem. This was important from a UX perspective as well: the less friction there was between Resolve and the rest of the IT stack, the more natural it became for teams to adopt automation as part of their everyday work.
Designing AI That Explains Itself
One of the more important parts of the experience was making AI-driven recommendations understandable. Resolve could use data from past activity to identify patterns, predict potential issues and suggest actions before problems escalated, but those suggestions needed context.
Users should not be expected to trust a recommendation simply because the system says it came from AI. The interface therefore focused on showing the reasoning behind important suggestions and giving users enough information to decide whether an automated action made sense.
That balance between automation and transparency was central to the design. The aim was not to remove people from the process entirely, but to reduce unnecessary manual work while keeping decision-making visible.
From Automation Tool to Operational Partner
The broader product idea was to move Resolve beyond being a collection of automation features and toward becoming something closer to an operational partner for IT teams. Instead of reacting to every incident manually, users could monitor systems, automate recurring work and use AI-driven insights to identify what needed attention earlier.
The results of the work were very positive and are shown in the metrics section at the beginning of this case study, but the more interesting outcome was the change in how the product could be perceived. When automation becomes easier to configure, easier to understand and easier to trust, it stops feeling like specialist infrastructure and starts becoming part of the everyday operating model.
