Client
A UK local authority's household waste and recycling collection service, delivered jointly with its outsourced collection contractor across a mixed urban and suburban district
Sector
Local Government & Environmental Services
Engagement
Collection-round exception platform linking contractor round telemetry, resident-reported missed collections and street-level access or contamination issues into one shared operational view - multi-quarter programme.
What the client needed
A missed bin collection is one of the most common reasons a resident contacts their council, and before this engagement, resolving one meant reconciling two accounts that rarely matched: the resident's own report, taken by the contact centre, and the contractor's round-completion record, which nearly always showed the street as collected. Neither side was lying - a driver ticking a street as done from memory at the end of a long round and a resident checking their own bin an hour later were both describing something true from where they stood, just not the same thing. Without any independent record of what actually happened at a specific address at a specific time, the contact centre had no way to tell a genuine miss from a bin put out after the lorry had already been, and every case defaulted to a phone call between council and contractor to argue it out. The client's own figures showed a cluster of streets generating repeat missed-collection complaints most weeks, with no way to tell whether that pattern was a persistent operational problem or a persistent misunderstanding.
How we worked
- Integrated with the contractor's existing in-cab telemetry, which already logged a GPS position and a bin-lift sensor event for every collection but had never been made visible to anyone outside the contractor's own operations team.
- Built a shared exception view showing, for any resident-reported miss, exactly what the round telemetry recorded at that address and time - a genuine skip, a recorded lift the resident hadn't yet seen, or no vehicle presence at all.
- Added a same-day return-crew trigger for confirmed genuine misses, replacing a process that previously waited for the next scheduled round regardless of cause.
- Built a separate street-level issue log for access and contamination problems - a car blocking a bin store, a bin presented with the wrong material - that the telemetry could evidence but a phone complaint alone couldn't, routing these to street-level follow-up rather than the return-crew queue.
- Ran the platform alongside the existing phone-based reconciliation process for a full collection cycle across every affected street before treating it as the primary source of truth.
- Trained contact centre agents and contractor supervisors together on reading the same exception view, so a call about a missed bin became a lookup rather than a negotiation.
Measured results
All figures verified with the client. Specific site, personnel and resident detail withheld in line with our standard confidentiality terms.
- Streets previously generating a weekly pattern of repeat missed-collection complaints dropped to occasional, individually explainable exceptions within two full collection cycles.
- Genuine misses now trigger a same-day return-crew visit in the large majority of cases, rather than waiting for the next scheduled round.
- Contact centre call-handling time for a missed-bin enquiry fell substantially, since agents can check the exception view directly instead of raising a query to the contractor and waiting for a reply.
- The shared telemetry surfaced a specific access problem on one estate where a recurring parking pattern was blocking bin-store access on collection day - previously logged as repeated "missed collection" complaints rather than the access issue it actually was.
- The council and contractor now use the same exception data in their own contract performance reviews, replacing a dispute over whose figures were right with a single shared record.
"Every missed-bin call used to start the same way: we'd tell the resident we'd check with the contractor, the contractor would tell us the street was collected, and the resident was left feeling like nobody believed them. Most of the time everyone involved was telling the truth as they understood it. What we didn't have was one record that could actually settle it, so a fair number of genuine misses got written off as a misunderstanding simply because we had no way to prove otherwise. That's the part this fixed."
Working on something similar?
If this engagement looks like the kind of problem you are facing, we would be glad to compare notes by email.
Context and constraints
The brief was explicit that the platform had to work with the contractor's existing in-cab telemetry rather than requiring new hardware on the vehicles, since the contract was mid-term and neither party wanted a procurement exercise before the underlying data problem was even understood. That telemetry already existed and was already reliable - the gap wasn't in the data being collected, it was in nobody outside the contractor's own operations office being able to see it when a resident called in. Getting read access to that feed, and agreeing what a contact centre agent should and shouldn't be able to see about a specific driver's round, took longer than any of the reporting logic that came after it, but it was the piece that made everything else possible.
A second constraint shaped the design just as much: the council did not want a platform that simply took the contractor's telemetry at face value and treated it as ground truth over a resident's report. A bin-lift sensor firing doesn't prove the right bin was emptied, or that it was put out at all - it proves a lift happened at that location and time. The exception view was built to show what the telemetry recorded, not to declare a verdict, leaving the contact centre agent to make the final call with better information rather than handing that judgement to an algorithm.
Why a shared view mattered more than a smarter one
Early design conversations spent real time on whether to build a scoring or prediction layer on top of the telemetry - flagging streets statistically likely to generate complaints before they happened. We deliberately didn't go that direction first. The client's actual problem was: two parties working from separate, unverifiable accounts of the same event, and a resident stuck in the middle of a dispute they had no way to win. A shared, simple, verifiable record of what happened at an address solved that problem directly. A predictive layer might be worth adding later, once the underlying data trust issue was gone, but building it first would have added complexity to a problem that turned out to need transparency, not intelligence.
A full collection cycle before switching over
We ran the exception platform alongside the existing phone-based process for a full collection cycle across every affected street, rather than a shorter pilot on a handful of streets chosen for convenience. That longer, wider comparison is what surfaced the parking-related access problem on one estate - a pattern that a shorter or narrower pilot, covering different streets or a shorter season, might easily have missed, since the parking pressure driving it was tied to a specific school run time that only showed up consistently across a full cycle of data.
Lessons learned
The first lesson was that a dispute over "what actually happened" is usually a data visibility problem wearing the costume of a service failure - the underlying telemetry already existed here, and simply making it visible to the party fielding the complaint resolved most of the argument before it started.
The second lesson was that treating vehicle telemetry as evidence to inform a decision, not as an automatic verdict, kept the platform useful rather than adversarial - contractor staff engaged with a system that showed its working, and would likely have resisted one that simply overruled resident reports by default.
The third lesson was that a longer, full-cycle comparison period against the old process caught a genuine access problem that a shorter pilot, run on a convenient subset of streets, would probably have missed entirely - the same pattern we've seen matter in other operational coordination programmes we've delivered.
If your organisation is caught in a similar dispute between two parties' honest but conflicting accounts of the same operational event, we would be glad to discuss what a programme like this might look like for you. Email sales@halfteck.com.