In a six-month pilot project, enercity and findIQ proved that the Hanover-based municipal energy provider can operationally deliver key steps toward the energy transition using digitalized service knowledge: 75% faster diagnostics, zero misdiagnoses, over 200% ROI — and an apprentice with two weeks of experience resolving a fault in 7 minutes. What started as an internal tool is now becoming a business model.
enercity runs district heating infrastructure for roughly one million people across Hanover, Germany. They're in the middle of a major network expansion and they're losing experienced technicians to retirement faster than they can replace them. It's a workforce development problem that every plant manager and operations leader recognizes. The assets keep growing. The expert headcount doesn't.
Their answer: stop letting that knowledge walk out the door. enercity partnered with findIQ — the Industrial Knowledge Intelligence platform — to digitize their best technicians' expertise and put it directly in the hands of the people who need it. Six months in, the results were clear enough to sign a framework agreement and expand across the entire organization.
The problem: more infrastructure, less tribal knowledge
enercity is one of Germany's largest municipal energy companies: the group supplies around one million people with electricity, heat, and drinking water. In addition, the Hanover-based company offers energy-related services including e-mobility, decentralized energy generation, energy efficiency, grid services, and smart infrastructure solutions.
By 2040, enercity's district heating network is slated to nearly double in size. Two coal plants are being decommissioned. Fossil-free alternatives are going in. And the technicians who know how to keep it all running? A significant share of them are within a few years of retirement.
The labor market isn't helping. A 2018 Deloitte survey put average vacancy periods for service technician roles at 185 days. Today, some of those positions simply don't get filled. The result is a pressure that operations leaders know well: more equipment to maintain, fewer experienced people to maintain it, and zero tolerance for downtime in critical infrastructure.
— Marcus Velden, Business Development Manager for Network Services, enercity
To pursue this path, enercity — as a "pioneer of the heating transition" — decided to partner with findIQ.
From use case to pilot project
The two companies came together through a strategic exploration of AI at the management level. After evaluating several solutions, they decided to deploy findIQ specifically to address a bottleneck in their service operations: the diagnosis and restoration of district heating stations. findIQ's methodology starts with the people who actually do the work.
— Sina Volkmann, CEO and co-founder of findIQ
At enercity, that meant structured interviews and workshops with the most experienced field technicians. IT supported the process. A technically credible project lead kept it on track. The pilot focused on compact district heating substations — a recurring asset type that, despite some variation, could be mapped into a structured diagnostic model.
Why generic tools don't cut it in the field
Most knowledge tools weren't built for how field service actually works. findIQ was. Videos work for fixed sequences — fault diagnosis isn't fixed. Prose knowledge articles go stale, and technicians don't maintain documentation; they fix things. Static fault trees break the moment conditions change, and updates never happen. Generic AI and LLM-based tools average around 50% accuracy on precise diagnostic questions. That's not acceptable when a wrong diagnosis means extended downtime or a needlessly replaced component.
findIQ uses a structured symptom-to-cause matrix with probability weighting — purpose-built industrial AI, not improvised generative AI. Content is modular, enriched with images and audio where useful, and mapped in a dynamic heat map. In the field, technicians reach the root cause in an average of six steps, with a diagnostic hit rate of around 80%.
The platform also handles preventive maintenance routines and supports remote maintenance management — meaning senior engineers aren't fielding calls every time a less experienced technician hits a problem they haven't seen before. And it stays current: technicians flag updates from the field, AI-assisted refinements surface based on real usage, and the knowledge base improves with every deployment.
The results: Tested in live operations
enercity didn't take the vendor's word for it. They ran an A/B test in active service operations — real substations, simulated faults, multiple technician groups, repeated runs, with and without findIQ.
The secondary impact matters too. Senior technicians stopped being the bottleneck. Upskilling happened at the point of need, not in a classroom. Institutional knowledge stopped being a person — it became a platform.
What comes next
The pilot led directly to a framework agreement. findIQ is now expanding across enercity's electricity, gas, and water divisions. Heat pumps are the immediate next use case — a fast-scaling technology where most field teams are still building proficiency, and where AI-powered maintenance guidance can close that gap fast.
enercity is also licensing their knowledge content to other utilities facing the same workforce development and upskilling challenges. The internal service tool is becoming a commercial product — and enercity is becoming a technology provider, not just an energy company.
Three things that determine success
findIQ sees this pattern across industrial sectors. The technology is rarely what makes or breaks the outcome. Three factors do:
Start with the people doing the work, not the IT stack. User involvement from day one, a clear mandate, and a project lead who understands both the field and the business. Without that, even the best field service management software stalls.
Industrial AI needs structured knowledge to perform. Generic models without a structured foundation produce guesses. The combination of deterministic AI modeling with a continuously updated knowledge base is what separates predictive maintenance software that works from one that demos well.
If technicians don't use it voluntarily, the ROI never materializes. 4.6 out of 5 from field technicians isn't a vanity metric. It's proof the system works where it counts.
The bottom line
The energy transition doesn't stall for lack of ambition. It stalls when the people maintaining the infrastructure don't have what they need to do their jobs. enercity solved that problem — not by hiring more experts, but by making the experts they already have scale across every technician on their team. That's what Industrial Knowledge Intelligence does. And those results are available to any organization facing the same challenge.
findIQ is the first Industrial Knowledge Intelligence platform — turning expert know-how into precise, scalable maintenance and troubleshooting guidance technicians can trust.
What is findIQ and how does it work?
findIQ is the Industrial Knowledge Intelligence platform that captures the experiential knowledge of expert technicians and transforms it into precise, step-by-step troubleshooting and maintenance guidance. Unlike generic AI or static documentation, findIQ uses purpose-built industrial AI models; deterministic by design to deliver the right answer the first time, directly in the field.
How did enercity use findIQ to address its workforce development challenge?
enercity used findIQ to digitize the expertise of its most experienced district heating technicians before that knowledge retired with them. Through structured interviews and workshops, expert knowledge was captured and mapped into a diagnostic model. The result: any technician — regardless of experience level — can access that expertise on demand and resolve faults with expert-level accuracy.
What results did enercity achieve with findIQ?
In an A/B test conducted in live service operations, enercity saw time-to-resolution drop by more than 75% per dispatch. A trainee with two weeks of experience resolved a fault in 7 minutes — a comparable case took 43 minutes without findIQ. Misdiagnoses were eliminated entirely during the test period. The platform delivered greater than 200% ROI from initial deployment, with full cost recovery within six months.
How is findIQ different from generic AI tools or large language models?
Generic LLMs are generative by design. They produce plausible answers, not necessarily correct ones. In industrial diagnostics, that distinction matters. findIQ uses deterministic, model-based AI engineered to mirror expert reasoning. It consistently delivers the same accurate result across shifts, sites, and technician experience levels. No guesswork and no black box.
Can findIQ handle complex or varied equipment types?
Yes. findIQ is purpose-built for industrial complexity. At enercity, compact district heating substations were mapped into structured diagnostic models that account for asset variation. The platform adapts to symptoms, machine variants, and evolving field conditions — and continuous feedback loops ensure the knowledge base stays current as equipment and operating conditions change.
What industries and use cases is findIQ suited for?
findIQ was built for any industrial environment where experiential knowledge is critical to service performance — manufacturing, utilities, energy, and beyond. It is particularly valuable where expert technicians are retiring, where new hires need to reach proficiency faster, or where consistent diagnostic accuracy across multiple sites or regions is a business requirement.