How Numan scaled safe, AI-first support with Fin AI Agent

Numan is on a mission to help people live happier, healthier and longer lives - and that starts by making healthcare more accessible. They offer diagnostics, medication, and holistic support from health coaches and clinicians – all through a seamless mobile platform. These involve personalised treatment for sensitive, often stigmatised health issues, from weight management to erectile dysfunction. In this high-trust, highly regulated space, customer support plays a critical role in helping patients use the platform and get the care they need – ensuring every step is safe, compliant, and accurate.

As support volume surged to nearly 70,000 conversations a month, Numan needed a more scalable solution that could deliver fast, accurate answers and ensure that all medical queries were assigned to trained clinicians. Most importantly, it had to do all of this without compromising on safety or quality. That’s when they turned to Fin.

In doing so, they didn’t just manage rising volume – they elevated the entire patient experience, maintaining a 90% CSAT, and allowing the team to focus on resolving more complex queries and providing the highest level of customer service to their patients.  

Let’s see how they did it.

Numan’s rapid growth brought a sharp increase in support demand, with queries ranging from delivery updates to sensitive questions about treatment plans. At the time, Numan was using another customer service platform – but as volumes surged to 70,000 conversations a month, it became clear it wasn’t built to scale with intelligence. "Our support team had to manage every type of query manually," shares Stephen Senez, Customer Service Platform Manager at Numan. "It became clear we needed something more scalable and intelligent that could help us handle growing volumes and enable our team to focus on higher-value work."

We tested a few options, but they were miles behind. Fin, on the other hand, was already a fully fleshed-out tool at the time, ready to go.

Numan knew AI would be the answer, so they began exploring different solutions, including their previous platform’s own AI. “We tested a few options, but they were miles behind,” says Stephen. “Fin, on the other hand, was already a fully fleshed-out tool at the time, ready to go.”

But that alone wasn’t enough. In healthtech, the bar is high. “Speed matters – but accuracy matters more,” Stephen says. “If we were going to introduce AI, it couldn’t just sound good. It was essential it could assign medical queries to a member of the clinical team.” They tested numerous AI solutions rigorously. “A lot of the tools performed well in demos, but when we tested them ourselves, they broke easily,” Stephen recalls. “Fin was the only one that gave high-quality responses based on our actual help center and blog content – with almost no setup.”

Fin offered more than automation. It could resolve a large share of conversations instantly, while ensuring that all medical queries were routed straight to the Clinical team.  "With Fin, we saw an AI agent that could give customers accurate answers quickly – and crucially, escalate to a teammate when needed, whether that’s a member of the customer team, a clinician, or a health coach," says Stephen. "That was key for us." Intercom also met the compliance requirements of operating in healthcare. With HIPAA, GDPR, ISO 27001, and other key certifications, Fin gave the team confidence that patient data would be safe.

Rolling out Fin wasn’t just about adding automation – it was about making sure AI could operate safely within the boundaries of a regulated industry.

We were cautious from day one, we wanted to be sure that Fin could operate effectively without compromising on standards.

That caution shaped every decision Numan made as they rolled out Fin:

1. Revamping their knowledge base to train Fin 

From the start, the team understood that Fin’s effectiveness would depend on the quality of their help center content. They refined their internal knowledge database to ensure Fin was drawing from accurate, well-structured information.

2. Reviewing conversations to build trust

The team began by manually reviewing 100% of Fin interactions to ensure quality, accuracy, and safety. “We were cautious from day one,” says Rhidian Boobier, Head of Customer Operations at Numan. “We wanted to be sure that Fin could operate effectively without compromising on standards.” Once confidence grew, they shifted to a 5% monthly audit led by their team.

3. Giving Fin guidance to assign clinical queries to humans

The next step was to make sure anything clinical was excluded from Fin entirely. So the team implemented a set of clear rules to instantly hand over any conversation involving medication, treatment decisions, or follow ups regarding side effects to a clinician. These queries would go straight to licensed clinicians or medically trained teammates. "We wanted to move fast with AI, but we needed guardrails to do it safely. Guidance gave us the confidence to scale Fin and stay compliant," shares Stephen.

Numan’s results with Fin were immediate and grounded in what mattered most – safety and a better customer experience.

By deploying Fin and leaning into features like Fin Guidance, Numan transformed the way their support team operates. Here are just some of the outcomes they’ve seen:

• 19,000 hours saved annually, equivalent to the workload of nine full-time agents. “Each time Fin resolves a query, we save around four pounds,” says Stephen. By resolving tens of thousands of conversations automatically, Fin gave Numan the ability to scale sustainably without additional hiring.
• 47% Fin resolution rate. This level of automation has unlocked greater efficiency, leaving more time and space to focus on the clinical and sensitive issues that truly need a human touch.
• 90% CSAT for Fin, on par with the team’s human agents. Customers are getting faster answers and a consistent experience, without sacrificing the quality they expect. “Our customers are really impressed – they get the answer they need instantly,” Rhidian shares.

With Fin, Numan now delivers fast, high-quality support at scale – without compromising on the safety and standards the healthtech industry demands. By handling thousands of operational queries each month, Fin has freed up bandwidth for their team to focus on what matters most: supporting patients with empathy and care on clinical conversations. “Fin takes care of the volume. Our team focuses on care,” Rhidian explains.

Fin has become a core part of their support team, working alongside their human agents to deliver safe, high-quality care at scale. For Numan, scaling with AI was never about replacing people – it’s about letting humans do what they do best. “Fin handles the repetitive work so our agents can bring empathy, context, and insight to the conversations that really need it," says Rhidian.

There’s still skepticism in healthcare around AI, and that’s understandable as healthcare is held to a higher standard. But with the right safeguards and the right partner, it’s absolutely possible to scale safely and deliver better experiences.

With Fin deeply embedded in their support operations, Numan is already focused on what’s next, and smarter AI-powered reporting is top of mind. The team has been exploring Intercom’s Customer Experience Score and Topics Explorer to surface real-time insights from every conversation, helping them improve the support experience, close content gaps, and better inform how the business prioritizes and builds.

That broader impact is exactly what excites Numan about this next phase. By combining automation with control and visibility, Numan is not only scaling support safely, but also unlocking insights that help improve both patient experience and internal operations. “There’s still skepticism in healthcare around AI, and that’s understandable as healthcare is held to a higher standard,” says Rhidian. “But with the right safeguards and the right partner, it’s absolutely possible to scale safely and deliver better experiences.”