VIVID was the Winner of the Artificial Intelligence category of the Housing Technology Awards 2026.
At VIVID, we’re focused on giving our customers the best experience possible so that’s why we’re transforming how we use AI to support customers and improve our services.
Instead of waiting for complaints, we now predict where problems might occur and step in early to fix them where we can. Using AI-driven insights, we can spot patterns in repairs and customer interactions, helping us to prevent problems before they become bigger. This shift from reactive to proactive means fewer serious problems, faster resolutions and, most importantly, a better experience for customers.
What we did
Complaints were previously something that we reacted to and while we were confident in our ability to resolve them, we knew we could do better. Now, we don’t just respond to complaints, we work to prevent them happening.
We introduced predictive analysis and AI-driven early interventions, which helps us spot cases that might become more serious problems. We also created a cross‑functional ‘complex cases team’, combining experts from repairs, customer service and data. They take full ownership of these cases, making sure that customers receive the right support and faster, more effective resolutions.
How it works
To support this approach, we built a two-layer AI system:
- Repeat-repairs pipeline – This uses natural language processing (NLP) to analyse repair descriptions, engineers’ notes and boiler error-codes. It identifies recurrent repair problems such as failed first-time-fixes, repeat visits and suspected workmanship problems.
- AI complaint-risk model – Built using a gradient boosting model, this predicts the likelihood of a complaint by looking at behavioural insights, interaction patterns, repair histories and complex-case flags.
Together, these tools help us to spot high‑risk cases early and give our teams the information they need to step in quickly. They provide a full picture of what’s going on, including the root causes, so we can resolve problems faster.
Home-built & just £1,000 per year
One of the best bits? We built the whole system ourselves. It’s designed specifically around how we work, uses our existing platforms and only costs around £1,000 per year to run.
Since its launch, our predictive AI system has flagged 500 high-risk cases, and of these, fewer than one per cent became complaints.
But this wasn’t just about preventing complaints, it was also about improving customer experience. Customer satisfaction has been exceptionally strong, with an average score of 9.9 across these cases.
The system also gives us greater visibility of workmanship problems and recurrent faults, such as boiler problems. These insights mean that we can make targeted improvements, such as reducing repeat gas repairs from 17 per cent to nine per cent.
AI supporting our teams
Our work with AI is all about empowering our teams so they can deliver the best possible experience for our customers. Our AI isn’t here to replace people; it’s a tool that helps our colleagues to focus on the important stuff, with clearer insights and fewer barriers in the way. Furthermore, explainable AI has played a big part in boosting people’s confidence because the reasoning behind each case is explained.
Teamwork has been just as important. Repair specialists, customer-service colleagues and data scientists are now working more closely together, using shared insights to guide their decisions and take action quickly. This joined‑up approach means our teams can solve problems sooner, support each other better and provide faster, more personalised resolutions for customers.
Where next?
We’ve proven AI can make a real difference to customer experience at VIVID. By combining predictive modelling with repair-pattern NLP, we’ve built a system that doesn’t just spot when things might go wrong, it helps us to understand why.
It’s changed how we work, giving our teams the insights to act earlier and put things right faster. And because we built it in-house and made it part of our daily way of working, it’s practical, low-cost and ready to grow. We’ll continue to embed the system into our customer experience team and keep exploring the ways we can use AI and smart-technology for the benefit of our customers and their homes.
Fiona Harris is the chief information and data officer at VIVID. The housing provider was the Winner of the Artificial Intelligence category of the Housing Technology Awards 2026.
