Driver performance problems: five signals, one conversation

A driver who is late twice a week costs more than two late deliveries. It costs the planner who re-sequences the round around them, the customer service time on the two complaints, the redelivery on the parcel that went back to the depot, and the drivers on either side who quietly stop trusting the schedule. Across a year that one pattern is a five-figure sum, and it is not hidden. It is in the scan data, the complaint log and the fuel card statement, visible to anyone who looks.

Most driver performance problems are like that: visible and unaddressed, not invisible. This post is about the five signals worth acting on, what you can lawfully do with the data behind them, where the line sits when the driver is self-employed, and how the first conversation goes so that it is also the last one you need.

The five driver performance signals worth acting on

Not every dip is a problem. A bad Tuesday is weather. The signals below are patterns, and the test for each is the same: does it repeat, and does it repeat for this driver more than for the drivers on similar rounds?

Repeated lateness on the same window

Late twice a week, or late on the same drop every week, is a routing or a discipline problem and you need to know which. Compare against the drivers on neighbouring rounds. If they hit the window and this driver does not, the round is not the cause.

Incidents and near misses

One incident is an event. Two in a quarter, or a run of minor damage claims on the same van, is a pattern. Reversing is where UK workplace transport deaths cluster: HSE's struck by vehicle guidance puts nearly a quarter of workplace transport deaths during reversing manoeuvres. A driver clipping bollards is telling you something about how they reverse.

Fuel and harsh-event data out of line with peers

Fuel per mile that runs above the fleet average for the same vehicle type, or a harsh braking and acceleration count that is consistently top of the list, is a driving style signal. It costs fuel now and shows up as an incident later.

Customer complaints with a theme

One complaint about a rude driver may be a rude customer. Three complaints in a month that say the same thing are a signal. Read the text, not the count.

Fewer drops per hour than peers on a comparable round

Productivity below the peer group on the same density of round is the softest signal because the reasons vary widely: an unfamiliar area, a vehicle problem, a driver who is unwell, or a driver who has checked out. It is the one that most needs the conversation before the conclusion.

What telematics data you can lawfully act on

Telematics gives you three of the five signals in real time, and it is easy to build a disciplinary case from it without noticing you have crossed a line. The line is drawn by the ICO.

The ICO's guidance on monitoring workers is clear on four points. Telematics data that records what a driver does is personal data and falls under data protection law. You must tell drivers and passengers that the vehicle is monitored and what for. Monitoring driver behaviour, as opposed to tracking the vehicle's location, is high-risk processing and needs a data protection impact assessment before you switch it on. And you must not use the data for a purpose other than the one you stated, unless the new purpose is compatible or you have consent: data collected to plan routes cannot quietly become the basis of a disciplinary.

The practical version:

• Write down the purposes before you install anything: route planning, fuel management, safety coaching. If disciplinary use is a purpose, say so.
• Define the events you will act on (speeding above a threshold, harsh braking above a count) and share the definitions with drivers.
• If drivers can use the vehicle privately, the monitoring stops when the shift does. The ICO says you will rarely be able to justify monitoring private use.
• Use the data first for coaching. A driver who has seen their own harsh-braking count and been offered help before any formal step has been treated fairly, and the record shows it.

Self-employed drivers: where performance management becomes control

If some or all of your drivers are self-employed, the way you handle performance is also an employment status question, and the Supreme Court has already answered the easy version of it.

In Uber BV v Aslam, decided on 19 February 2021, the Court held unanimously that Uber's drivers were workers, not independent contractors, whatever the contract said. Among the findings it relied on: Uber monitored drivers' acceptance rate and logged them off for ten minutes if it fell below a threshold after warnings, and Uber ran a passenger ratings system where failure to maintain a specified average led to warnings and then termination. Lord Leggatt called that ratings mechanism a classic form of subordination characteristic of an employment relationship. Worker status brought the National Living Wage and paid holiday with it.

The lesson for a fleet is not that you cannot hold self-employed drivers to a standard. It is that the standard has to be about outcomes the contract specifies, not about how they do the job:

• Outcomes are fine. Delivery windows, damage rates, scan compliance, the customer standard. Those are the service you contracted for.
• Method is the risk. Telling a self-employed driver how to drive, which route to take or what hours to be available, penalising them for declining work, or running a ratings-based discipline process are all control over the manner of performance.
• Training is a grey area. Mandatory training on how to do the work is one of the indicators of control. Training the driver has asked for, or training that is a condition of the contract and stated as such, sits differently. Take advice on how yours is framed.

None of this is legal advice, and status cases turn on the whole picture. But if your process for a self-employed driver's lateness looks exactly like your process for an employee's lateness, you have a status problem waiting for a tribunal, and the recruitment funnel post covers why that also costs you drivers before it costs you a hearing.

The first conversation: one pattern, one change, one date

The conversation is the part managers put off, and putting it off is what turns a pattern into a habit. Done properly it takes twenty minutes.

Bring one pattern, with dates. Not "you have been late a lot". "You were outside the window on the Tuesday round on 2, 9, 16 and 23 September, and the drivers on the two neighbouring rounds were inside it every week." Data ends the argument about whether there is a problem and moves straight to why.

Ask before you decide. The reason is often something you did not know: a school run, a vehicle fault that was reported and not fixed, a round that was re-cut badly in July. Sometimes it is a driver who no longer wants the job. You cannot tell which until you ask, and the answer decides what you do next.

Agree one change and one review date. One, not five. A different start time, a route change, a module on reversing, a vehicle swap. Set a date two to four weeks out and put it in both diaries.

Write it down the same day. What the pattern was, what the driver said, what you agreed, when you will look again. Two paragraphs. Share it with the driver.

For an employee this is informal, and stays informal if the pattern stops. For a self-employed driver it is a conversation about the contracted outcome, and the note reflects that. Either way, the note is the thing that matters most, and the next section is why.

Why the written record matters to your insurer

After a serious incident, every claims handler and every HSE inspector asks the same question in a different form: what did you know, and what did you do about it?

HSE's guidance for employers on driving and riding for work says you must assess workers' competence and make sure they are adequately trained at no cost to them. If your telematics shows six months of harsh-braking events for one driver and there is no record of anyone acting on it, you knew and did nothing, and the log proves it. If the same telematics sits next to a dated coaching note, a targeted training completion and a review that showed the count falling, you managed a known risk, and the log proves that instead.

On the insurance side, the same evidence decides whether a claim is a routine payout or a fight. Fleet insurers ask at renewal how you manage driver risk, and a fleet that can produce coaching records, licence check dates and training completions per driver is a different proposition from one that cannot. The safety training post covers what a documented programme does for the conversation with your underwriter. The short version: the note you write after a twenty-minute conversation is the cheapest piece of risk management you will ever produce.

A targeted module after an incident beats "regular training"

Most fleets answer performance problems with the same tool: an annual refresher for everyone. It is easy to schedule and it satisfies the policy. It also lands months after the incident and covers nothing specific to the driver in front of you.

The evidence on training decay points the other way. The Department for Transport's rapid evidence assessment on efficient driving found that training produces reductions in fuel use of up to 25% immediately afterwards, falling to up to 6.5% for employee drivers over the long term. The effect is real and it fades. Short, specific, repeated input holds; one long session a year does not.

So the intervention after a reversing incident is a short module on reversing, assigned that week, with a practical check in the yard. After a run of speeding events, a module on speed and following distance. After a complaint theme, a module on the doorstep. Ten to fifteen minutes on the driver's phone, logged against their profile, followed by the review date you already set. That is what Service Club ® built the Academy for: modules an operator can assign to one driver after one specific incident, with the completion recorded where the coaching note lives. It is reached through the employer, not by the driver signing up alone.

Then measure. The harsh-braking count, the lateness on the window, the complaints with that theme. If the number moved, the module worked. If it did not, the problem was never a skills problem, and you have learned that for the cost of fifteen minutes.

Frequently asked questions

Look at the log, then have the conversation

The pattern is already in your data. The cost of it is already on your P&L. What is missing is usually twenty minutes and a written note. If you want the follow-up to be a module you can assign to one driver the same week rather than a refresher for everyone next spring, Service Club Academy is built for that.