How field service data becomes more valuable over time: Lessons from Google, OpenAI and Anthropic

While the majority of us remember a time without AI, AI itself doesn’t remember a time without us — without our data. What we share with AI tools like ChatGPT and Gemini about our problems, our predicaments and the world we navigate, gives them the context to help with the task in front of us. We give these tools this information freely because we expect value in the AI tool’s output, whether it’s collating meeting notes into clear next steps or making sense of an excessively large spreadsheet without losing an afternoon to it.

‍

In the built world, where HVAC equipment is installed, maintained and repaired, the stakes of this relationship are higher. We feed the tools supporting that work oceans of information:: jobsite details, performance metrics, equipment data and service history spanning across whole portfolios. In order for an AI tool to become integral in an organization’s tech stack, for their users to continuously input information on the job, the tool has to provide something useful in return. 

‍

By taking a look at how the companies behind these major AI models build their products, we can better understand how AI tools are expected to grow in the field service space: by making the case for daily interaction with the tool, and by discovering practical ways to make the collected data useful in the future. 

‍

What Google, OpenAI and Anthropic show about data’s value

‍

Google has spent years giving us a reason to tell it what we’re looking for. In its Q2 2025 earnings, the company reported that AI Overviews were driving more than 10% additional searches globally for the types of searches where they appeared. People were searching more as they discovered that Google could help with more of their needs. Giving people something useful was giving them a reason to ask another question.

‍

For OpenAI and Anthropic, the conversations themselves can also help improve future versions of their tools when consumer users allow that use. OpenAI, for example, can use the thumbs-up or thumbs-down feedback ChatGPT users leave on an answer, and the conversation behind it, to help train future models. Even telling the tool that it missed the mark can contribute to making it more useful down the road. Anthropic gives a straightforward example: a developer working through a coding problem with Claude can provide information that helps a future model handle similar problems. Someone gets help with the work in front of them, and that interaction can contribute to a tool they’ll use down the road.

‍

These examples show two sides of the exchange: useful tools give people a reason to come back, and the information people choose to share can help improve future tools. In field service, the long-term value can come from building a more complete history of the equipment and the work done on it. But first, you have to start by asking: what would make a field service stakeholder see that capturing information to feed an AI tool as something that helps them in the long run, and is therefore worth continuing? 

‍

What makes an AI tool worth using every day?

‍

If you dropped an AI tool into a field service business, and asked each level of stakeholder to engage with the tool daily, to feed it jobsite or operational data, what would it take for the stakeholders to want to keep using it? At the technician level, if they’re going to switch up how they work, it better result in a little less strain on their back, and an earlier exit from the jobsite. For salespeople, the tool is going to need to measurably increase their commission. At the business owner level, it should communicate the information gathered by your technicians and sales teams to better sell your service and grow revenue. If the AI tool isn’t resulting in relief or revenue at every level, it won’t be used daily, it won’t grow in sophistication, and it’s going to end up being about as useful as a digital paperweight.

‍

‍

How Veregy turned jobsite data into business value

‍

Veregy gives us an example of what that exchange looks like in practice. The company paired XOi’s AI-enabled tools with changes to its quoting and billing processes. At the technician level, documenting work came with access to equipment information they could use in the field. On the sales side, repair opportunities could become quotes before the technician left the jobsite, and Veregy reported a 73% increase in quote acceptance. At the business level, average invoice turnaround dropped by at least 55 days. And while that information was helping get today’s work approved and billed, each documented visit was also building a history of the equipment; information that could support the next service call, renewal or service agreement. The technician had a reason to capture it today, and the business was building something it could keep using tomorrow.

‍

What started as data captured at the jobsite level was enhanced, organized, and communicated by the XOi AI tools, and became something Veregy could use to move quotes and invoices forward. That’s a practical reason to keep the information coming. And as more visits are documented, the business builds a clearer equipment history to draw on for future service, renewals, and customer planning.

‍

Making service data useful beyond the next repair

‍

That expectation — that the information we put in should keep delivering more value over time — is at the heart of how XOi sees field service growing beyond the “break-fix” service model. The service event that happens today can solve today's problem. The information it produces can help prevent the next one, strengthen the next recommendation, and inform an investment decision years down the road. This new model provides field service providers with a way to move past reactive work by revealing where revenue is at risk and where the next opportunity is emerging. The examples from Google, OpenAI and Anthropic point toward the same long play: give people a reason to contribute today, and keep finding ways to make that information useful down the road.

‍

Your team’s time and information should keep paying off after the job is done. Download ”The Changing World of HVAC: How to operate beyond break-fix” to explore how the service history you’re building today can help you identify future work, guide customer investments, and build revenue that doesn’t depend on waiting for the next breakdown.