Address
7 Bell Yard, London, WC2A 2JR
Work Hours
Monday to Friday: 8AM - 6PM
1NCE, one of the fastest growing providers of IoT software and connectivity, operates the world’s largest privately run IoT network in the cloud. Offering a software platform for connected products, 1NCE has more than 24,000 customers, managing 30 million devices across 173 countries.
With a growing number of new customers each month, 1NCE recognised that maintaining personalised, high-quality support at scale could quickly become a challenge.
We’re a truly digital business,” explains Jan Sulaiman, VP of Global IoT Solutions at 1NCE. “And our customers come from more than 100 different countries across the world. We can’t possibly provide the level of high-touch service to everyone but at the same time, we want to deliver a consistent experience.
Jan Sulaiman, VP Global Solutions
Challenge
Initially, the pain was felt internally. Jan and his Solution Architecture team were fielding customer queries at all hours of the day, and many of them were repeated across markets and customer segments.
What started as a search for a way to “augment our solution architects and reduce manual workload” evolved into a more ambitious vision.
“At first, we just wanted to make it easier to give consistent answers internally,” says Jan. “But then we said hang on, why limit this to internal use? Why not offer the same to everyone and extend the service to our customers?”
That realisation reframed the opportunity.
With 10,000 visitors per month accessing their Developer Hub. With limited ways to interact with the content dynamically, 1NCE saw the potential to improve the overall customer experience.
One that could engage both existing customers and prospective users directly on their terms.
People love our Developer Hub, it gives them access to real insight even before becoming a customer, something relatively rare in our space. But it was all static documentation. Customers had to figure things out for themselves. That’s what led us to start thinking differently about the opportunity.
Jan Sulaiman, VP Global Solutions
Then came another challenge, servicing customers in the Brazilian market. “The use of WhatsApp in Brazil has changed customer expectations, requiring faster responses from support agents rather than the slower email and ticket systems we use elsewhere.” states Jan.
In Brazil very few people use English, even in technology.
This meant the 1NCE Developer Hub was not solving the usual problems as it does in other regions. To satisfy 1NCE’s growing global user base, a multilingual service quickly became a non-negotiable.
With Brazil, we’d expanded into a market with very different conditions and unexpected customer interactions compared to other regions. We required a tailored strategy that specifically addresses the needs of Brazilian customers.
Jan Sulaiman, VP Global Solutions
Solution
From Passive Documentation to Dynamic, Consistent Conversations From the outset, 1NCE didn’t just want ‘another AI widget’ they wanted a strategic capability.
That meant choosing a partner to go beyond the hype and building something real.
“We always look for companies with a disruption mindset,” says Jan. “We’re not always drawn to big names or brands. We’re drawn to expertise. Disruption only happens when you work with people who think differently. That’s what we saw with Shipshape. A team that’s not trying to do AI for the sake of it but one that genuinely focuses on expertise and continuous innovation. The team at Shipshape brought technical skill and the ability to help us think in the context of the business outcomes we wanted to achieve.”
Shipshape delivered a multilingual AI assistant deployed on 1NCE’s Developer Hub, in English and Portuguese.
Built using 1NCE’s certified documentation and institutional knowledge, the solution transforms static content into a conversational experience, capable of providing nuanced, on-brand answers to technical and support-related queries across the globe.
Shipshape gave us a way to understand the data in plain language, which makes it actionable across the business, it’s not just scary AI black box stuff. It’s insights we can work with, identifying common questions, seeing where customers are stuck, and fixing those issues at the root.
Jan Sulaiman, VP Global Solutions
That continuous improvement cycle is critical to the solution at 1NCE. Every query that the AI cannot answer confidently becomes fuel for enhancement.
The Shipshape solution captures those gaps, analyses them in natural language, and then flags them to 1NCE to create, refine, and integrate answers into the AI knowledge base.
“It helps us see where we’re falling short” says Jan. “If our customers keep asking something we can’t answer, that’s telling us we haven’t been talking about something they clearly care about. That’s insight and it helps us close the loop.”
Importantly, the solution isn’t just designed to help after a support ticket has been raised. It serves as a first line of defence, reducing inbound load while offering clear, reliable answers to users before they even hit the support team.
“This isn’t just about support tickets” Jan says. “It’s about allowing us to support more customers without increasing costs. That’s the real driver, cost avoidance at scale.”
Outcome
The results from the first implementation cycle were compelling, even before optimisation or deeper tuning.
In the first test scenario, 20% of support queries were answered perfectly, indistinguishable from responses by first-line support agents. A further 40% were “almost perfect”, requiring only light refinement.
“Even in the first test against historical support tickets, 60% of queries could be handled to a human-equivalent standard,” Jan notes. “And that’s without any tuning or iteration, just our first attempt at loading some initial knowledge.”
Shipshape built an analytics engine that provided the key to unlocking the remaining 40%.
Due to the way the solution is designed, the knowledge gap closes on a weekly basis. All unanswered questions are logged and reviewed, along with the reasons why. The analytics engine is used to pinpoint where new knowledge needs to be added to the platform, continuously improving, week by week to eradicate any knowledge gaps over time.
The implications are significant.
With even early-stage performance already removing friction, 1NCE can support more users globally without increasing the headcount of its support team.
With annual support costs in the multiple millions dollars every year, the business can protect its margins as it scales; a critical strategic advantage for 1NCE’s long-tail customer economics.
And because the solution captures, structures, and transforms conversational data into natural language analytics, it adds insight beyond support.
Product, marketing, and operational teams can now mine questions for emerging themes, unmet needs, and knowledge gaps all from unstructured user queries.
We’re not just getting answers, we’re getting intelligence, we’re learning what our customers aren’t finding in our documentation. That helps every part of the business.
Jan Sulaiman, VP Global Solutions
What began as a front-end support layer has now evolved into something much broader.
The vision is for the solution to become the centralised AI knowledge base across 1NCE, a single source of truth that internal and external platforms can access via API.
“We’re building the foundation for a universal knowledge layer,” says Jan. “Not just for our Developer Hub, but for every application in the company that needs trusted information.”