I once worked with a housing provider that had a mystery problem. The number of tenant complaints it received suddenly shot up. Every week more people grumbled about their home. Some reported wet walls and window condensation, others talked about damaged furniture and a few tenants said they were having breathing difficulties.
Operatives were dispatched, repairs made, damp-proofing done, windows fixed and walls re-painted. But the wave of complaints kept coming. After many months of tenant dissatisfaction soaring and its repairs spending growing, asset managers discovered the cause. A design defect in over 100 semi-detached houses meant that penetrating damp was affecting the smallest bedroom, where a hidden gutter had begun to leak. Each home was built at the same time, to the same design, and the gutters had all begun to fail within a few months of each other.
This problem could have been avoided through better data analysis and visualisation. Every repair was logged in the housing provider’s asset management system. But what if repair requests, and tenant complaints, were automatically fused with GPS data so that asset managers could map them out? This would have indicated that the influx of new jobs was all within a one mile radius of each other.
If complaint and repairs data had been combined with property type and age records then asset managers would have also seen that it was houses constructed at a similar time and to a similar specification that were affected. Spotting these trends sooner and putting a planned programme of rectification in place could have prevented much of the damage.
Combining asset management data with new streams of information isn’t a new idea. The technology that enables landlords to join these data dots has been available for a number of years. But uptake in the sector is slow because, in reality, blending data sets in the social housing sector isn’t a straightforward process.
With only the minority of housing providers exploiting repairs information to its full potential, I think it’s time for the sector to reappraise its approach and overcome the barriers that are holding them back. Here are five reasons why.
Tenant perceptions can help to predict product age
Traditionally, planned maintenance is organised around component lifespans. There is a standard view of the predicted lifecycle of an item and as works are undertaken, the clock is re-set. This relatively rudimentary way of understanding stock condition can be supplemented through housing surveys, but for large landlords, surveying each property isn’t always possible.
So, rather than purely using a mechanical lifespan calculation to determine when to invest, some landlords are now linking this data with results from tenant satisfaction surveys. Residents’ views about the appearance and condition of their properties could flag up patterns around certain components. For example, a batch of boilers that were expected to last 15 years but which many tenants think are performing poorly could indicate a shorter life span.
Avoid tenant dissatisfaction & emergency repairs
Linking repairs trends to installation dates could also identify issues with a particular model. If 50 electric showers, all the same model and installed at a similar time, are breaking down within a few months of each other, this might enable the landlord to claim against the manufacturer. It could also indicate that asset managers need to change their specification for electric showers or introduce a replacement programme to avoid tenants’ complaints and responsive repairs costs.
Reduce empty homes
Joining up voids records, property-type information and mapping data could support housing providers to cut the number of empty homes. For example, some streets might have unusually high levels of unoccupied houses. Linking these figures with data on the specification of a house and its repairs history could help asset teams understand why there is a cluster of voids. The answer might be the tiny third bedroom, making it unsuitable for families. It could be the fact that it’s at the top of a hill, far from local amenities but also exposed to prevailing weather which is shortening the lifespan of the roof and encouraging leaks. Or perhaps the reason for the unpopularity of a row of properties, each built to the same design, is that they all have defective radiators and poor insulation, meaning that they never stay warm.
This intelligence can help asset managers tackle ongoing voids. Data patterns might signify that a layout reconfiguration could increase lettings. Another solution might be to improve the insulation or replace the heating system on all houses in the street, or perhaps the best idea is to sell off certain properties and cut all losses.
Cut refurbishment costs
As stock gets sold and records revised, it can be easy to miss properties by simply looking at an address list. One street might need kitchen re-fits, but visually mapping the data could draw your attention to a nearby cul-de-sac which has a single property on it, still owned by the housing provider. Rather than coming back to this house the following year, it might be more cost effective to include it in the flow of current works, keeping the tenant happy and the contractor’s price down.
Understand property value
Fusing asset management numbers with public data sources is another way housing providers can better understand the bigger picture. For example, land-registry information about recent property prices can help asset teams to establish their portfolio valuation. Adding in tenant perception data can also provide another perspective. Residents’ opinions about their homes and neighbourhoods might help asset teams to form a more holistic view about the value of certain properties.
One of the key elements to achieving this holistic, 360-degree view is having enough accurate, granular asset management data in the first place. I regularly work with social landlords to drill down into their low-level transactional records, grouping this into specific programmes of work and then putting it into the context of national data sets using, for example, our spending control software, Valueworks.
The housing sector is notorious for keeping data in silos. This extends to contractors where records are collected in aggregate across spreadsheets or filed away in paper form. But the tide is changing. Housing providers can see the valuable intuition that linking pots of operational data could offer; the key is overcoming barriers around resource and expertise, information blind spots, data collection and quality.
Phil Moss is the chief technology officer at Procurement for Housing.
