Seven things to check before you buy a used fleet

Someone sends you a spreadsheet and four photos. Two thousand scooters, one price, "all in working condition". What do you actually check before you wire the money?

This is the part of buying a used fleet nobody writes about, so here it is. Work through it in order - the questions get more expensive to answer wrong as you go down.

1. Can the vehicles be switched on at all?

Start here, because everything else is irrelevant if the answer is no.

Shared vehicles are tied to an operator's IoT platform. The scooter does not belong to whoever has the key - functionally it belongs to whoever controls the backend. If that platform will not release them, you have bought hardware you cannot activate.

Ask, in this order:

• Which IoT vendor is in these vehicles?
• Has the seller actually spoken to that vendor about releasing them, or are they assuming?
• Is there a written confirmation, or at least a named contact who has said yes?
• If the previous operator has folded - who is answering the phone now?

The last one catches people out. A fleet from a company that went under can be perfectly good hardware attached to a backend nobody is administering any more. Sometimes there is a path. Sometimes the honest answer is that the IoT has to be ripped out and replaced, and that is a real cost per unit that belongs in your offer.

2. What is actually left in the batteries?

The pack is the most expensive part of the vehicle and the only one that degrades whether or not anyone rides it.

A fleet at 80% state of health goes back into shared operation. The same fleet at 60% does not - it will not hold a shift, and you have bought a battery replacement programme with some scooters attached.

Two things get confused here:

• A BMS readout is what the battery management system reports. It is an indication, it is free, and it is sometimes optimistic.
• A capacity discharge test is a measurement. You charge a pack, discharge it under load, and find out how many watt-hours actually came out.

They do not always agree. On a large fleet you cannot test every pack, so what matters is the sample: how many were tested, how they were chosen, and whether the serial numbers are recorded so you can check any of them yourself.

Ask for the spread, not the average. A fleet averaging 75% could be uniformly fine or it could be half good packs and half scrap, and those are very different purchases.

3. How many will start?

"Working condition" means nothing. What you want is a count.

• How many units power on?
• How many are complete - deck, stem, both wheels, charger port, no missing panels?
• How many are known non-runners, and are they being sold as such?
• Are there vehicles being counted twice, or a "unit" that is actually a frame with no battery?

Get a physical count against the number on the invoice. It is boring and it is the single most common discrepancy we find.

4. Can you get parts?

This is the one that quietly kills the economics six months in.

An orphaned model whose manufacturer no longer supplies parts has a maintenance cost curve that goes vertical the first time something common breaks. If you cannot buy a controller, a motor or a replacement pack, then every failure retires a vehicle permanently and your fleet shrinks all season.

Check before you buy: is the manufacturer still trading, is the model still supported, and can you actually source packs and motors for it? A well-supported fleet at a higher price per unit is frequently the cheaper fleet.

5. Does the paperwork exist?

Two documents decide whether the fleet can move and whether it can be deployed.

UN 38.3 test summaries for the batteries. Without them the packs default into the damaged-or-defective regime for transport, which closes air freight entirely and pushes packaging costs up on everything else. Most sellers do not have these to hand. Ask early, because chasing them takes time.

Approval documents for the destination market. A Certificate of Conformity for type-approved vehicles, or the national approval where the market requires one. If you are buying L-category vehicles and the seller has lost the CoC, there is a useful provision worth knowing: for ten years after production the manufacturer must issue a duplicate on the owner's request, at no more than cost.

6. Photos are not evidence

Four good photos prove that four vehicles looked fine on the day someone took four photos.

What you want instead:

• Photos of a random sample, not a chosen one.
• Serial numbers visible, so a specific claim attaches to a specific vehicle.
• A video of a unit being powered on and ridden.
• Wide shots of the storage - how a fleet has been stored tells you a lot about what condition it is in.

Ask to pull a specific vehicle on a video call. If a seller will do that without notice, that tells you something. If they will not, that also tells you something.

7. Then go and look, or send someone

On a fleet worth six figures, a site visit is cheap. Either go yourself or have someone do it for you.

That is what a FleetReport is: a technician travels to where the fleet is sitting, counts it, tests packs, checks the backend release with the vendor, photographs what they find, and sends you a PDF with the serial numbers of everything sampled. You do not have to be buying through us to commission one, and we are not paid more if the number comes out higher.

The short version

Ask these seven, in order, and you will avoid nearly every expensive mistake in this market:

• Can it be released from the backend, and who confirmed that?
• What is the measured battery health, on what sample size?
• How many units actually power on?
• Can I still buy parts for this model?
• Do the UN 38.3 summaries exist?
• Are the approval documents there for where I am deploying?
• Will someone let me look, properly, before I pay?

Any seller worth buying from will have answers. If the answers get vague, the price should too.