Two engineers. Same street. Two weeks apart.
The first one arrives on a Tuesday morning with a full day blocked for six installations. He rings the first doorbell at 8:30. By 11:00 he’s on his third job. But each installation is taking longer than it should, because at every property he’s running cable from scratch — routing from the external connection point through the building, drilling where needed, finding the right entry point, running the internal cable to where the customer wants the router. Each job that should take forty-five minutes is taking ninety. He gets through four installations by the end of the day. Two are rescheduled.
The second operator’s engineer comes through the same street two weeks before the installations are booked. He pre-cables every property where a connection order is pending — runs the external drop cable to each property, makes the connection at the external point, leaves the internal cable terminated and labelled at the entry point. No customers need to be in. He does twelve properties in a day without ringing a single doorbell.
Two weeks later, an installation engineer follows behind. Every property is pre-cabled. His job at each address is to connect the pre-run cable to the ONT, position the router, test the service and leave. Average job time: thirty-five minutes. He completes eight installations in a day, every customer is connected and he’s home by 4pm.
Same street. Same number of properties. Completely different economics.
What pre-cabling actually does to the numbers
The cost difference between a pre-cabled and a non-pre-cabled installation comes from a few specific factors, each of which compounds the others.
Job time. The most visible factor. An installation engineer doing both the external cable run and the internal connection in a single visit is doing two distinct jobs. Pre-cabling separates them and separation means specialisation — a pre-cable engineer moves fast along a street because they’re doing one thing repeatedly and an installation engineer moves fast because the time-consuming external work is already done.
Customer availability. Pre-cabling the external connection doesn’t require the customer to be home. The drop cable can be run, the external connection point sealed and the cable left ready at the entry point without anyone unlocking a door. This decoupling is significant. The installation engineer’s visit — which does require the customer to be present — becomes shorter and more predictable. The abortive visit rate for installation jobs drops because the only reason for an abortive visit is customer availability, not cable routing problems.
Geographic efficiency. When the scheduling system groups pre-cable jobs by area rather than by individual order date, a single engineer can work a street systematically — property to property, without backtracking, without travel time between jobs. The geographic efficiency of a pre-cable run is much higher than a mixed-task installation schedule, where engineers might be going from one side of town to another between jobs.
The cost differential — roughly £40–£70 per connection — adds up quickly at any meaningful installation volume. An operator doing 500 installations a month saves somewhere between £20,000 and £35,000 a month by running a systematic pre-cable programme versus a single-visit approach. Annualised, that’s a quarter of a million pounds in field efficiency, without touching engineering headcount.
How the scheduling has to work
Pre-cabling only delivers its efficiency benefit if the scheduling system treats it as a distinct job type with its own geographic logic. This is where most operators who try to run pre-cable programmes run into problems — they’re trying to manage it manually, or within a scheduling system that treats all jobs as equivalent and schedules them by date rather than by area.
The right approach is for the OSS to identify, typically a week to ten days ahead of the installation date, all pending installation orders within a given geographic area. Those addresses are grouped into a pre-cable batch and scheduled as a single area-based run — one engineer, one area, one day. The pre-cable job completes. The installation jobs that follow inherit the pre-cabled status and are scheduled knowing that the external work is done.
The scheduling system needs to understand the relationship between pre-cable and installation jobs. A pre-cable job isn’t just an earlier task — it’s a dependency. If the pre-cable hasn’t happened, the installation job needs to either be rescheduled or converted to a single-visit job with the appropriate time allocation. Getting this logic right in the scheduler is the difference between a pre-cable programme that actually delivers the efficiency benefit and one that creates confusion and missed dependencies.
One engineer, one area — the geographic grouping that makes it work
The efficiency of a pre-cable programme is almost entirely a function of geographic grouping. An engineer pre-cabling twelve properties on the same street is extraordinarily efficient — park once, walk to each property, minimal travel time between jobs. An engineer pre-cabling twelve properties scattered across a ten-mile radius is barely more efficient than doing twelve individual jobs.
The ML scheduling engine in Confideo OSS looks at pending installation orders and clusters them geographically — identifying streets or small areas where enough installs are pending in the coming week to justify a pre-cable run. The clustering algorithm balances two competing pressures: tight geographic grouping (which maximises pre-cable efficiency) against order lead times (which push toward scheduling individual installations sooner rather than waiting to batch them with nearby properties).
In practice, a threshold approach works well: if three or more installation orders are pending within a defined geographic radius within a ten-day window, generate a pre-cable job. Below that threshold, the travel efficiency gain doesn’t outweigh the scheduling complexity. Above it, the pre-cable run is almost always worth doing.
When pre-cabling fails — and what happens next
Pre-cabling doesn’t always go smoothly. The engineer arrives at a property and finds access is blocked — a locked gate, an inaccessible duct entry, an external wall that requires specialist equipment to drill. Or the cable run requires a different route than the standard approach because of a non-standard property layout. Or the connection point is missing or damaged. The pre-cable cannot be completed as planned.
This is where the workflow matters enormously. A failed pre-cable that just becomes a note on a job sheet — “couldn’t complete, access issue” — and sits waiting for someone to pick it up is a problem that drifts. The installation engineer arrives a week later, finds the property isn’t pre-cabled and either aborts the job or does the full single-visit installation without the benefit of the pre-cable preparation. The efficiency of the pre-cable programme is undermined by every failure that isn’t properly handled.
The right handling of a failed pre-cable is automatic escalation to the Network Service Planning (NSP) team. The moment the engineer marks the pre-cable as failed and records the reason, the OSS creates an NSP job — a planning task that puts the property back in the hands of the planning team with all the context needed to resolve it properly.
The key feature of this workflow is that nothing is lost. The failed pre-cable doesn’t sit as a note on a job waiting for someone to notice it. It immediately becomes a planning task with all the context attached — the failure reason, the engineer’s notes, any photos of the access issue — and the planning team can act on it the same day. The installation job is suspended cleanly until the pre-cable dependency is resolved, so the installation scheduling queue doesn’t have jobs sitting in it that can’t be executed.
What the planning team does with an NSP job
When a failed pre-cable lands with the NSP team as an NSP job, they’re not starting from scratch. The engineer’s failure record tells them the specific problem — blocked access, non-standard route, missing connection point, specialist equipment needed. That context shapes what the planning response needs to be.
For access issues — a locked gate, a property where the standard external route is blocked — the planning team may simply arrange contact with the customer or property manager to resolve the access question before re-dispatching the pre-cable engineer. No survey needed. Just coordination.
For non-standard routes — properties where the cable can’t follow the standard external path because of the building layout, obstacles, or an unusual connection point — planning needs to produce a connection list: a documented alternative cable route with specific instructions for the engineer who will complete the pre-cable. This is the kind of detailed specification that a standard pre-cable engineer isn’t expected to produce on the fly at the property, but that a planning team can produce from a desk using survey photographs and building records.
For properties requiring specialist equipment — high-reach access, confined space entry, specialist drilling — the planning team arranges the right contractor rather than sending a standard installation engineer who doesn’t have the equipment or certification for the job.
In each case, the output is a resolved pre-cable job with a complete specification that allows the subsequent installation to proceed normally. The NSP team’s role is to absorb the complexity that the standard pre-cable workflow can’t handle and produce a clean path through it.
The programme as a whole
The operators who run pre-cable programmes well treat it as a continuous operational rhythm, not a project. The scheduling system is always looking one to two weeks ahead, clustering pending installations geographically, generating pre-cable batches and managing the dependency between pre-cable and installation jobs. Failed pre-cables are automatically escalated and tracked. The planning team has a live queue of NSP jobs, not an occasional inbox full of emails.
The cost-per-connection benefit compounds over time. As the programme matures and the planning team builds up a record of non-standard properties in each area — the ones with locked gates, the ones with unusual building layouts, the ones that always need specialist equipment — the failure rate on pre-cable jobs drops. The street-level knowledge accumulates in the OSS and pre-cable engineers arrive with better specifications and fewer surprises.
None of this happens by accident. It requires the scheduling system to understand pre-cabling as a distinct job type, the OSS to manage the dependency between pre-cable and installation jobs and the NSP workflow to automatically catch and route failures without human intervention. Get those three things right and the economics of last-mile delivery shift permanently in the right direction.