Tool Trace
A tracking platform for Sandvik Coromant's regrindable drilling tools — commercially launched within a year, now serving 10 active customers and helping unlock a new business model.
The Problem
Sandvik Coromant needed to encourage customers to reuse regrindable drilling tools instead of buying new ones - but customers had no visibility into tool performance or lifecycle value.
My Role
Lead UX designer creating a tracking platform for three user types: manufacturers, customers, and recon operators who regrind used tools.
The Approach
Through site visits and user interviews, I identified that different users needed different data depth. I designed dashboards that show key metrics at a glance but let users dig deeper when needed and streamlined inspection flows for operators.
3+
Environments
In the product lifecycle, three unique users required three unique workspaces. As the project grew, new needs surfaced, requiring even more workspaces!
(Hover the images to the right to see the different environments)
Tool Production
Tool Recon
Tool Customer
Working in an agile team of six — one designer, a product owner, an architect, and four full-stack developers — I moved between problem definition, sketching, and developer feasibility checks within each sprint.
Complex data visualizations and graph features were common requests — rather than designing speculatively, I'd consult developers first to understand what was achievable before investing time in detailed designs. I also regularly reviewed the live build to catch layout and UX discrepancies before they became habits.
Visual design was built on an inherited design system from the client's web team, which I extended with new tokens as new needs arose — always in Figma, always with auto layout.
The reconditioning process
When the project started, user insights were unclear. Frameworks and practices between various recon centres were many and inconsistent. Together with the product owner, I conducted interviews with end-users from each environment to map out their current flows, pain points and what data was currently being recorded.
Unboxing
Unpacking a shipment and registering each tool into the system, creating a new recon order.
During site visits we discovered operators needed to cross-reference tool IDs with a marking machine placed too far from their workstation. Rather than redesigning the station layout, we designed a printable barcode sheet operators could bring to the machine and scan directly. This eliminated a manual transcription step from their workflow.
Incoming Inspection
Detailed analysis and registration of individual tool wear and condition
Operators reported too many clicks to scan and register each tool. During our site visits the reason became clearer: operators wore gloves and couldn't use a laptop trackpad, making every extra interaction a real friction point. We introduced keyboard shortcuts for all critical steps, making the flow operable without a trackpad entirely. We also analyzed the most common inspection values and set these as dropdown defaults, meaning operators could move through the majority of tools without touching those fields at all.
Final Inspection
Quality assurance after the regrind process before returning to customer
Before shipping, operators needed to manually mark each tool as approved — a necessary step since some tools don't survive the regrind process. With potentially dozens of tools per order, this meant a lot of clicks for what was usually a routine confirmation. We flipped the logic: all tools default to approved, and operators only need to flag the exceptions. What had been a repetitive multi-click process became a quick scan for the few that didn't make it.
Operators reported the changes made their work meaningfully faster — the inspection flow now mirrors their physical workflow closely enough that time-consuming steps, from order creation to tool marking, have been largely eliminated.
The end customer
The end customer is typically a production manager overseeing a manufacturing operation. Before Tool Trace, their biggest pain points were a lack of visibility into tool locations, stock levels, and tool performance — why tools underperform or break prematurely. The customer environment is built around three areas that address each of these directly.
Wear Analysis
Wear type against wear severity across a customer's tool fleet, giving a picture of how tools are being used and stressed. Different wear patterns indicate different handling conditions at site level — the data surfaces the signal, and the customer and Sandvik's technical teams act on it together.
Machine Analysis
The same wear data viewed at machine level. If a specific machine is producing abnormal wear patterns it becomes visible here, pointing to potential errors in that machine's operation.
Retired Tools
Average number of regrinds achieved by tools that have reached end of life, giving customers a concrete picture of tool longevity across a full lifecycle — only accessible once a tool has run its full course.
My Inventory
A complete view of every tool the customer owns, organized by article type down to each individual unit. The inventory list is the entry point, but the design is layered — summary KPIs and charts are available at a glance for routine checks, with the option to drill down into granular lifecycle history for any individual tool when needed. This way the interface serves both the quick daily check-in and the deeper investigation without overwhelming either.
On an article level, we show a breakdown of how many tools have had X number of regrinds, visualized as a chart. Since each regrind increases the risk of a tool not surviving the next reconditioning cycle, this gives customers an intuitive read on inventory risk without needing to understand the underlying mechanics.
We also surface a months to zero stock estimate based on average usage data. But average data only tells part of the story — a customer planning to change operations needs to model different scenarios. An estimation tool lets users input their own expected regrind frequency and usage rate to project stock depletion under different conditions, helping them make proactive ordering decisions rather than reactive ones.
Location
Customers often lack visibility into where their tools are in the cycle at any given moment — at the reconditioning centre, in transit, on site — making it difficult to anticipate shortages before they become a problem. The feature maps each tool to its last registered event, grouped into labeled location bubbles showing tool counts at a glance. Clicking a bubble reveals the individual tools at that location. The list and graph are fully connected — filtering by article type in the list updates the graph in real time, letting users narrow from a full fleet overview down to the location status of a specific tool type in one interaction.
The design started earlier than the data did. An initial prototype used a world map with location pins, but had to be set aside when it became clear we didn't yet have access to the tracking data needed to make it real.Instead, location awareness was built incrementally — as new procedures were introduced for recon operators, service managers, and production staff, each unlocking a new data source.Most recently, integration with DHL's API added another layer of transit visibility.
A dedicated no recent activity tab surfaces tools that have gone silent — likely scrapped or lost. Users can remove these from the system, preventing ghost tools from skewing the rest of the data, like falsely showing available stock that no longer exists. The feature was refined through workshops with an end customer.
The production unit
The production environment was intentionally minimal — operators at this stage needed only to register and tag new tools entering the system, so the interface reflected that narrow scope.
End
Tool Trace commercially launched within a year and has grown into a prioritised showcase product at Sandvik Coromant — unlocking conversations about a new business model. Ten active customers are now using it to detect machining errors, extend tool life, and make data-driven ordering decisions. As the sales team put it: "We need this."
Looking back, earlier access to end customers during the recon design phase would have surfaced the physical environment constraints we only discovered through site visits. It's a reminder that no amount of remote interviews fully replaces being present where the work actually happens.