A Data Protection Impact Assessment, or DPIA, sounds like paperwork produced to satisfy a regulator. It is more useful than that, and treating it as a compliance chore is a missed opportunity.
Done properly, it forces an organization to answer questions it would otherwise leave vague: what exactly are we collecting, why, who can see it, for how long, and what will we never use it for. Those are precisely the questions your drivers will ask. Having written answers before anyone asks can be the difference between a smooth rollout and a long, corrosive dispute.
This is general guidance, not legal advice. Vision Telematics does not offer legal advice, and nothing here should be relied on as a substitute for it. Data protection duties depend on your own circumstances and can change, so you should take independent legal advice before making decisions about monitoring your drivers.
When it is required rather than advisable
The UK GDPR requires a DPIA, before the processing starts, wherever a type of processing is likely to result in a high risk to people’s rights and freedoms. The Information Commission’s Office (ICO), the UK’s data protection regulator, applies that to vehicles directly in its guidance on monitoring workers. It says you must carry out a DPIA for the following:
Monitoring driver behavior. Driving style, harsh events, speeding and scoring. The ICO says this type of processing “would be considered high risk”.
In-vehicle cameras or audio. The ICO names “using cameras or audio” alongside behavior monitoring, and says that if you are considering video or audio monitoring of workers you must complete a DPIA. It does not separate forward-facing from driver-facing cameras, and it points out that outward-facing cameras can capture other motorists and pedestrians.
Analytics about drivers. Any monitoring tool that “uses analytics to make inferences, predictions, or decisions about drivers”, which describes most driver scoring and risk rating.
Facial recognition. The ICO says you must carry out a DPIA because it presents a high risk. Facial analysis that does not identify anyone, such as camera-based fatigue detection, is not named in that sentence, but it is still a camera and analytics making inferences about a driver, so the entries above apply to it.
Outward-facing cameras raise one further test. The UK GDPR says a DPIA is required for “a systematic monitoring of a publicly accessible area on a large scale”. The ICO says that whether something is large scale depends on the number of people concerned, the volume and variety of the data, how long it lasts and the area it covers, so a large fleet filming public roads should not assume it falls outside that test.
That leaves basic location tracking of a vehicle used solely for work, with no behavioral analysis. It is the lower-risk case, but it should not be assumed to be exempt. The ICO’s list of processing that requires a DPIA includes tracking an individual’s geolocation when combined with another high-risk criterion, and gives processing the location data of employees as an example. Its guidance also says employees can count as vulnerable because of the power imbalance with their employer, and vulnerability is one of those criteria.
If you screen a system and decide a DPIA is not needed, the ICO says you should document the decision and your reasons. It also says that if you are in any doubt, “we strongly recommend you do a DPIA”. Failing to carry one out when required may leave you open to enforcement action, including a fine. What the law requires of employers is set out in the guide to whether vehicle tracking is legal.
The practical reality: if your system does anything more sophisticated than showing a dot on a map, do the assessment. If it only shows a dot on a map, screen it and write the answer down.
A summary of ICO guidance, not a substitute for it. Read the guidance against what your own system does.
What to include
The UK GDPR sets a minimum: a description of the processing and its purposes, an assessment of whether it is necessary and proportionate, an assessment of the risks to individuals, and the measures to address them. For a fleet, that usually comes down to seven things.
What you are doing. The data collected, the sources, how often, who it identifies, where it goes, who processes it and where it is stored, including any transfers outside the UK.
Why you are doing it. The specific business need, stated concretely. “Improving safety” is too vague to assess. A concrete statement, naming the collision type, how often it happens and what it costs, can be weighed against the intrusion it justifies.
Necessity and proportionality. The core of the document. Could a less intrusive method achieve the same aim? If you are proposing continuous recording, why is event-triggered recording insufficient? If you are proposing a driver-facing camera, what specifically does it address that a forward camera does not? The ICO says you should consider whether less intrusive methods could achieve your purpose and document that assessment in the DPIA, so record the alternatives you considered and why you rejected them. If you have not yet decided whether to fit driver-facing cameras at all, start with the case for and against driver-facing cameras.
The risks to individuals. Be honest. Drivers may feel under surveillance. Data may be repurposed. Managers may act on incomplete context. Private life may be captured incidentally. Third parties are recorded without meaningful choice. Special category information may be captured accidentally: a medical conversation picked up on audio, a health condition visible in footage.
The mitigations. For each risk, what you are doing about it. On-device processing rather than continuous streaming. Audio switched off by default. Event-triggered recording. Short automatic retention. Role-based access with audit logging. A way to stop recording during private use. A written commitment on how footage is used in performance and disciplinary processes. A route for drivers to appeal.
Consultation. Who you spoke to and what changed as a result. The ICO says you should seek and document the views of workers or their representatives, such as trade unions, unless there is a good reason not to. This section is far more persuasive if something actually changed.
The outcome. Whether the remaining risk is acceptable, who approved it, and when it will be reviewed. If you have a data protection officer, you must ask for their advice and record it. If the DPIA identifies a high risk that you cannot reduce, you must consult the ICO before going ahead. The ICO describes a DPIA as a ‘living’ process, so keep it under review and revisit it if the system or the way you use it changes.
The part people get wrong
Two failures recur.
Writing it after the decision. An assessment produced to justify a system already chosen and installed is visible as such. The UK GDPR requires the assessment before the processing, and the ICO says it should begin early in the life of a project and is “not simply a rubber stamp”. Start it while options are still open, including the option of a less intrusive configuration.
Writing it in isolation. An assessment produced by one person in an office, without speaking to drivers, tends to miss the risks that matter and to produce mitigations that do not address real concerns. Involve driver representatives. It improves the document, and it turns the assessment from something done to the workforce into something done with them.
Why it helps commercially, not just legally
When a driver asks “who can see this footage?” or “will this be used to discipline me?”, the difference between a confident, specific, written answer and an improvised reassurance is enormous.
The assessment produces those answers as a by-product, and you will need them anyway: the ICO says that if you go ahead with monitoring after a DPIA, you must tell your workers about it before it begins. How to handle that conversation is set out in how to introduce cameras without losing your drivers.
The assessment is also the document you will want if there is ever a complaint, a grievance, a regulatory question or a serious incident, because it shows that you thought about the risks in advance and did something about them.
That is worth considerably more than the time it takes. The ICO says a DPIA does not have to be complex or time-consuming in every case, as long as there is a level of rigour in proportion to the privacy risks.
Sources: UK GDPR, Articles 35 and 36, as published on legislation.gov.uk. Information Commission’s Office (ICO) — Data Protection Impact Assessments (DPIAs), including When do we need to do a DPIA?, How do we do a DPIA?, Do we need to consult the ICO? and Examples of processing ‘likely to result in high risk’; Employment practices and data protection: monitoring workers; Surveillance in vehicles. All read on 2 November 2026; each carries the ICO’s notice that it is under review following the Data (Use and Access) Act. ICO — ICO welcomes transition to new Information Commission, 30 September 2026 (the Information Commission replaced the Information Commissioner on that date, and the organization is referred to as the Information Commission’s Office, or ICO). General guidance, not legal advice.
