Fleet Dash Cam Purchasing Checklist: A Nextbase Buying Guide for Company Vehicles

If someone in your company hands you 'sort out dash cams for the vehicles' and you're not a mechanic or an IT person, this is the checklist I wish I'd been given in 2022.

I manage purchasing for a 70-person field service company. Our fleet is 14 vehicles: eight Ford Transit vans, four pool cars and two box trucks. That's roughly $180,000 a year across six vendors. I've ordered dash cams for this fleet twice, and the second time I finally stopped repeating the first round's mistakes. Here's the seven-step process I use now.

1. Decide what each vehicle actually has to capture

The first order I placed was based on which model has the most features. That's backwards. Company cameras exist to answer specific questions: who hit the van, whose version of events is right, did that delivery actually happen. Write down the questions for each group of vehicles before you open a price list.

Our four pool cars do short city trips and park overnight in a locked yard. For those, the Nextbase 122 dashcam is enough. It's a Full HD camera with no phone app and no GPS to distract a driver. It records a clear forward view, and if a driver has a dispute at a junction, that footage settles it. To be fair, the 122 is the older entry-level model in the Nextbase range, but it still does the job we ask of it.

The vans are a different story. They sit at customer sites, they park on streets, and some jobs run late, so night-time quality matters. For the vans, we buy the 322GW, which adds GPS tagging for footage, or the 622GW 4K model where drivers work on unlit industrial estates. I'd rather spend more on the vehicles that need it than buy one common spec for the entire fleet and pretend it fits everyone.

2. Confirm what the model numbers on the quote mean

Model numbers get reused, regional versions get suffixes, and old stock sits in warehouses for years. At one point, a distributor quote listed Nextbase dash cam 320RSD next to the Nextbase 122 dashcam. To me, they looked like part of the same range. When I asked for the datasheets, I found real differences in the sensor, mount type and even the power connector. A model number alone doesn't tell you those things.

During our 2024 vendor consolidation, I stopped assuming a bigger number means a newer or better unit. Now I ask for the current datasheet and warranty terms for every SKU. If a supplier cannot explain how the 320RSD listing differs from the 322GW that our installer spec'd, that's a red flag. It's easier to ask before the purchase order goes out than to argue about returns after delivery.

3. Plan the power setup by where each vehicle parks

This is the step I would redo if I could go back. A dash cam matters most when the vehicle is parked, which is exactly when the engine isn't running to power it. If you wire every camera to the ignition, you get no parking coverage. If you leave every camera running off the starter battery all the time, you'll eventually hear from a driver who can't start the van.

We sort our vehicles by parking situation:

• Vehicles in our locked yard overnight don't need parking mode. Ignition-switched power is fine.
• Vehicles that go to drivers' homes or stay at customer sites get a dashcam external battery pack. It charges while the engine runs, then feeds the camera after shutdown without draining the starter battery.
• Vehicles that sit for a week or more get a battery maintainer. We keep a Prologix battery maintainer connected to two spare vans in the yard, which already has outlets for winter block heaters. The maintainer holds the battery at float (a maintenance state, not a continuous charge), so those vans start when they're finally called into service.

One honest warning: a dashcam external battery pack won't rescue a weak alternator or an old battery. The vehicle's charging system has to refill both the starter battery and the pack during a normal drive. That's why step 4 exists.

4. Scan the vehicles before installation

This is the step that no installer suggested to us in our first round. We added it after a van failed to start and the blame landed on the dash cam. The scan is simple: battery voltage with the engine off, charging voltage with it running, and a read of any stored fault codes.

Because our fleet is mostly Ford vans, I get asked for the best OBD2 scanner for Ford at least twice a year. My honest answer is that you don't need a dealer-level tool for this check. A mid-range scanner that reads codes and live voltage is enough. If you need to go deeper into Ford-specific body modules later, FORScan on a laptop covers more than most scanners at any price. The point is to know what existed before the camera went in.

5. Build the order vehicle by vehicle, not model by model

My first camera order was for 18 units. I forgot that each one also needs an SD card, a mount and, in most cases, a power cable. The real order was almost half as big again once accessories were added.

I now write the order as a list per vehicle:

• one front camera at the spec from step 1;
• a rear camera module only where the parking risk justifies it, and only if it's compatible with the front unit;
• one SD card per camera, sized for at least a week of driving;
• a hardwire kit with a fuse tap for every vehicle that needs parking mode;
• one spare mount per five vehicles, because mounts fail at the worst time.

A serialized invoice matters as much as the hardware. The first vendor we used was cheaper, but their invoice didn't list serial numbers, and finance rejected part of the expense. I covered $1,200 from the department budget. Now I verify serialized invoicing before we discuss volume pricing.

6. Set one installation standard and test every unit

If every driver or workshop fits cameras their own way, you end up with random angles and cables running across airbag zones. Our installation rule is deliberately short:

• mount the camera behind the rear-view mirror, in the area the wiper clears;
• take power from a switched fuse using a fuse tap, not from an airbag area or the steering column;
• run the cable along the headliner and down the A-pillar trim, clipped in place.

After installation, each vehicle gets a ten-minute test. Start the engine and confirm the unit records. Switch the ignition off and confirm it enters parking mode or shuts down as designed. Then play one minute of footage back in the office. It's tedious, but it caught two faulty cameras and one fake SD card in our last batch.

7. Keep spares and set the footage policy

Plan for one unit in ten to need replacement within the first year. Drivers drop mounts, cards corrupt, and some defects only appear after a few weeks. I keep one spare camera and two spare mounts for every ten vehicles, configured exactly like the rest of the fleet.

Also decide who owns the footage before the first complaint arrives. Tell drivers what the camera records, why it's fitted, and who can access the footage. In our company, the camera belongs to the fleet, and footage is only reviewed for insurance or a formal incident. That single conversation removed most of the resistance.

What I'd avoid if you're starting from scratch

First, buying one model for the whole fleet. A car in a locked yard and a van parked on a city street are not the same risk, and a single good-enough spec usually means paying for capabilities that half the fleet never uses.

Second, treating power as an afterthought. If you haven't decided which vehicles need parking mode before the installer arrives, the solution will be improvised. A dashcam external battery pack is the right answer for some vehicles and wasted money for others.

Third, skipping the ten-minute test. A dead camera on arrival looks exactly like a bad installation, and a fake SD card looks exactly like a dead camera. The test costs ten minutes; pulling every unit out after a complaint costs days.

There's no single best setup that fits every fleet. After three years of buying these cameras, I've come to believe the right question isn't 'which Nextbase model is best?' It's 'what does this vehicle have to prove, and how long will it sit parked?' Answer those two, and the spec chooses itself.