Improved heat exchanger reliability with AI troubleshooting

Hover over the results to learn more 👀
Higher completion rate
32,8%
Every customer support call carries a cost to the company. By reducing call volume, we saved 11.4 MNOK, ensuring the team could continue investing in product improvements.
Improving the product experience led to a reduction of 110K support calls, allowing the team to dedicate more time to complex user issues (Jan-Aug 2022)
Reducing simple inquiries freed the team to focus on more calls, boosting efficiency and response throughput.
Iterations carried out
6

Alfa Laval asked us to help create an MVP application to evaluate customers’ heat exchangers and suggest maintenance performed by Alfa Laval technicians.

MY ROLE
I led the initial phases of product discovery, including user research, ideation, prototyping, and presenting the MVP application design. I worked closely with stakeholders and the development team throughout the process to ensure a smooth handover for implementation.
End users

Ensuring an intuitive flow in the initial PVA was critical. The application needed to serve both Alfa Laval’s repair technicians, who have deep domain expertise, customer support agents, and account managers who are in discussions with the technical teams operating the heat exchangers.
Discovery process

Aligning stakeholders and defining direction


To kick things off, I facilitated a discovery workshop with Alfa Laval’s key stakeholders. The goal was to align on the app’s objectives, validate assumptions, and uncover hidden challenges.

I began with empathy mapping to encouraging participants to step into the shoes of both customers and technicians. We mapped out how they currently maintain equipment, their frustrations with existing processes, and their expectations from a new service tool.


Seamless flow of a PWA for AI powered diagnostics and preventive maintenance

Our design goal was to create a seamless PWA flow that would guide a technician from scanning the heat exchanger with a third-party thermal camera app to generating a diagnostic report powered by AI.

The report would flag issues early: helping customers prevent costly malfunctions and enable Alfa Laval to offer this as a premium preventive maintenance service.


The unified user flow that connected every step has the following:

1. Entering the Alfa Laval PWA,
2. Scanning or uploading thermal images,
3. Generating an automated health report
4. Contacting customer support when needed.

To make results instantly understandable, we introduced a traffic light system in backend:

🟩 No fault have been found
🟨 Potential faults detected
🟥 Critical faults detected

Iterations

Gathering early insights from user testing with customer support team

For the first round of user testing, we invited five customer support employees to test the new flow, as we were unable to access the repair technicians who would ultimately use the product. Their feedback was highly valuable:

Maintenance notifications were well received.
Navigation and overall logic were smooth.

Technicians, however, needed clearer guidance when using the thermal camera on how to handle connection issues and whether to capture photos within the app or upload them afterward.

Visuals

Bridging PWA and Alfa Laval product guide

Based on these insights, we refined the flow to improve usability and outcomes:

Users can manually enter a serial number if it’s not found in the database.
The products page now displays all owned units, completed checkups, and upcoming maintenance reminders.
A fallback support flow ensures help is always accessible.
Unit verification was simplified, guiding users through key questions and photo uploads to generate more accurate reports.
In the flow, we designed a simple yet effective process for generating a troubleshooting report.
Once the report was created, the customer would receive clear instructions on how to proceed, depending on the condition of the heat exchanger.

The final iteration was very well received by the stakeholders.

The project was then taken over by Alfa Laval's internal engineering team, who continued with the development of this solution with the help of Microsoft consultants.
Learnings
Customer support agents = valuable for research
Initially, the client believed user testing was unnecessary. However, after the discovery workshops, where it was seen very good that certain touch points were missing and they didn't know how exactly to explain it - then they agreed to provide access to the customer service team, who were in daily contact with the repair technicians. This allowed us to gather valuable insights from the people who served as the crucial link between the product and its end users.

And the story didn’t end there…

Beyond the solutions presented here, the project included important dimensions such as cross-functional collaboration, stakeholder alignment, and overcoming team-level challenges. I’m happy to share more about these behind-the-scenes learnings :)
Contact me

CONTACT        
Ready to talk? Get in touch

View case study