The Guinness Partnership: predictive insights from ivy to damp and mould
When customers mention plants creeping up a wall, it's easy to dismiss as small talk. Here's how The Guinness Partnership turned comments like that into hard evidence of damp and mould risk — and a step towards predicting problems before they're reported.
The Guinness Partnership held large volumes of customer comments across surveys, complaints and contact centres, but none of it was objectively grouped or scored. Manual review introduced bias, and word-frequency analysis only surfaced the most common terms — not the issues that mattered or the ones no one had thought to look for.
Guinness had extensive, unquantified feedback arriving from every channel — surveys, complaints and contact-centre conversations. The raw comments were rich, but there was no consistent way to group them, score them, or compare them over time.
The methods available didn't help. Reading comments by hand introduced human bias and didn't scale, while simple word-frequency analysis only told the team which words came up most often — not which issues to act on, and certainly not the problems hiding in plain sight that no one had yet named.
The solution
Wordnerds gave Guinness an objective, consistent way to read every comment — and a way to surface issues the team didn't know to search for.
Objective priority lists
Complaint descriptions were analysed for sentiment and for critical topics like vulnerability and health & safety, producing objective priority lists instead of subjective, manually-read ones.
Insight into assets
Customer perspectives on specific building components were extracted and scored for sentiment — giving the team a feedback-led view of how their assets were actually performing for residents.
Topic modelling that finds the unexpected
Rather than forcing comments into predefined buckets, Wordnerds let topics emerge — including a cluster of plant-related comments that turned out to connect to documented damp and mould incidents.
The impact
From a comment about ivy to evidence of damp and mould risk
Informal comments about ivy and creeping plants potentially blocking ventilation became hard evidence: analysis showed a higher incidence of damp and mould in properties where customers mentioned vegetation. Guinness turned an offhand observation into data-driven proof — and is now moving from retrospective reporting towards predicting problems before they're reported.
It started with something easy to overlook: customers mentioning ivy and creeping plants that might be blocking ventilation. Because topics surfaced on their own, those comments clustered together rather than disappearing into the noise. When the team cross-referenced them, properties where customers mentioned vegetation showed a higher incidence of damp and mould — turning an informal observation into documented, data-driven evidence.
That micro-discovery pointed at something bigger. Having shown they could connect what customers say to what's physically happening in a home, Guinness is now shifting from retrospective reporting towards predictive modelling — estimating the likelihood of tenant-reported failures, and surfacing damp and mould that hasn't been reported yet.
About The Guinness Partnership
Established in 1890, The Guinness Partnership is one of England's largest providers of affordable housing and care, managing nearly 70,000 homes and serving over 140,000 customers nationwide.
Its work is focused on improving lives through quality housing — which makes understanding what residents are actually experiencing central to everything the organisation does.
What would Wordnerds surface from your feedback?
Book a diagnostic and we'll look at your customer feedback together — and show you the themes hiding in it.
So you're reading the footer now? Either you ❤️ Wordnerds or you're desperate for something to read. Either way, CX Corner from Wordnerds is the answer. Fortnightly Voice of Customer bombs dropped in your box. Signup 👇 or find out more.
Sovereign · Housing
10 hours a week back in their pockets
How Sovereign freed 10 hours a week from manual tagging and made Wordnerds an integral part of root-cause analysis for 130,000 residents.
Raven · Housing
From data overload to driving change
How Raven Housing Trust stopped drowning in spreadsheets and spent its time driving tangible improvements from tenant feedback instead.
bpha · Housing
Three months behind to one day ahead
How bpha went from quarterly reports that were three months out of date to analysing 100% of feedback on day one — and hit a C1 RSH rating.
