I get the same email almost every week: someone’s ready to order 100 or 500 taxi-top screens, and they want to know which factory to trust.

The ones who’ve done this before usually ask smart questions. The first-timers almost always start with “What’s your price?”—which is understandable, but it’s also how you end up with a container of screens that start failing six months later.

This guide is the email I wish I could send to everyone before they start talking to suppliers. It won’t tell you which factory to pick, but it will tell you which questions separate the good ones from the ones you’ll regret.

1. Brightness: everyone claims 5000 nits

Walk around any LED trade show, and you’ll see “5000 nits” on half the booths. Some of them are telling the truth. Most aren’t—especially once the screen has been running for an hour in the sun.

Here’s the practical breakdown:

The question to ask isn’t “How bright is it?”—it’s “Can I see the calibration report?” A supplier who can show you third-party test data is already ahead of most. And if they can’t explain whether the brightness holds up when the screen is hot, that’s your answer. (Cheap screens can lose 20–30% brightness once they warm up. You won’t see it on a trade show floor, but you will see it three months later on a taxi roof in July.)

One more thing: get a light sensor built in. It’s not a premium feature—it’s basic common sense. Without it, your screens are either too dim during the day or way too bright at night, and someone will complain either way.

2. IP65: what it does and doesn’t tell you

IP65 means water jets from any direction won’t get in. That’s good. But a taxi roof is not a stationary billboard, and IP ratings are tested on stationary equipment.

The real problems are specific to moving vehicles:

• Rain at 60 km/h hits differently than rain falling straight down
• The cabinet expands and contracts every day, which works, but the seals loosen over time
• Constant vibration shakes screws and connectors that would never move on a wall-mounted screen

So when a supplier tells you “It’s IP65,” the follow-up questions matter more than the number itself:

• Are the connectors waterproof, or just the cabinet?
• Is there a drain path inside the housing if water does get in?
• Can I see photos of the internal layout—specifically whether the PCBs have conformal coating?

If you see bare circuit boards with no coating, that screen isn’t going to last. It might pass a water test in a lab, but it won’t survive a year of real roads.

3. Power: the thing that actually kills screens

This is the one buyers overlook most often, and it’s also the most common cause of failure. I’ve seen more screens die from power problems than from anything else.

A taxi’s electrical system is messy. Voltage fluctuates. A 12V sedan and a 24V truck charge differently. And every time the engine starts, the voltage dips and then spikes.

What you want in the spec:

If a supplier can’t explain their power protection in plain language, they probably bought the cheapest power supply they could find and hoped for the best.

Also, ask about average power consumption, not peak. A screen that averages 120W is realistic for a taxi fleet. One that peaks at 300W will drain batteries and make fleet managers unhappy—and unhappy fleet managers cancel contracts.

4. The control system is the real product

The screen is hardware. The content management platform is what your customer actually interacts with every day. If it’s clunky or unreliable, the screen is useless no matter how good the pixels are.

A few things to check:

4G connectivity — Is the module built into the board, or is it a USB dongle taped to the inside of the cabinet? Built-in is far more reliable. Also, does it support the LTE bands in your country? (This catches people out more often than you’d think.)

The platform itself—who hosts it? If the supplier’s server goes down, can you still push content? And does the platform give you play logs you can show to advertisers as proof of playback? Because if you’re selling ad space on these screens, “trust me” isn’t a report.

GPS — Even if you don’t need geofencing today, get it built in. Adding it later means pulling the screen off the roof, opening the cabinet, and wiring in a module. It’s not worth it. GPS also lets you prove where a taxi was when an ad played, which matters for premium clients.

5. Pixel pitch: bigger numbers aren’t always better

There’s a temptation to go with the highest resolution because it sounds better in a pitch. “P2.5 is sharper than P5” is true, but it’s also irrelevant if nobody’s standing close enough to see the difference.

The framework I use with clients: if the screen is meant to be read by someone stuck in traffic next to the taxi, P3.33 is usually the right call. If the budget is tight and the taxis are mostly on highways, P5 is fine. P2.5 only makes sense if the client is a premium brand and the creative actually uses the extra resolution.

One caveat: P2.5 generates more heat and draws more power. Make sure the cabinet is actually designed for it, not just “spec’d for it.”

6. Weight matters more than you think

A taxi operator has two concerns: fuel cost and whether the thing falls off.

Every extra kilogram adds drag. At motorway speeds, a poorly designed cabinet can add 3–5% to fuel consumption. That sounds small until you’re paying for 200 taxis.

The other half of this is the mounting bracket. A fixed bracket is fine until you need to remove the screen for service—then it’s a problem. Quick-release brackets cost more but save hours of labor. And if you’re deploying in a city where theft is a risk, ask about anti-theft locking. Screens do get stolen off taxi roofs.

7. Certifications: boring but expensive if you skip them

If you’re importing into the EU, US, or Australia, certifications are not a nice-to-have. They’re the difference between your shipment clearing customs and sitting in a warehouse while you burn through demurrage fees.

Here’s the trap: some factories will send you a CE certificate that applies to a completely different product. Always ask: “Is this certificate for this exact model, and can you show me the test report from TÜV or SGS?”

And if the screen has 4G, it also needs radio compliance—FCC Part 15 in the US, CE RED in Europe. A screen that fails RF compliance can get seized at the border, and the factory won’t be the one paying for the loss.

Four mistakes I see repeatedly

Buying on price alone. The gap between a 240 screen is usually about 60 is what determines whether the screen lasts 18 months or 5 years. I know which one I’d rather be quoted on.

Forgetting to ask about the platform cost. You buy 500 screens and then realize the content management platform charges 18,000 a year you didn’t budget for. Ask before you sign.

Not testing in your actual climate. A screen that works fine in Shenzhen may not survive a summer in Riyadh or a winter in Moscow. Ask the supplier for reference installations in a similar climate—and actually call them.

Skipping spare parts. Modules fail. Power supplies fail. If your supplier isn’t including 3–5% spare modules with your order, you’ll be air-freighting replacements at emergency rates within six months. It’s not a question of if—it’s when.

Seven questions to ask before you sign

You don’t need to be a technician to ask these. You just need to listen to whether the answer is confident or vague.

1. Can you show me a brightness calibration report from a recognized lab?
2. What’s the average power consumption—not the peak?
3. Can I see inside the cabinet? Are the PCBs coated?
4. Which content management platform are you using, and what does it cost after year one?
5. Do you have certifications for my market for this specific model?
6. What spare parts ratio do you recommend for a [X]-unit order?
7. Can you put me in touch with a customer who’s had these screens in the field for 12 months?

If they can’t or won’t answer these, keep looking. There are good factories out there—but they’re not the ones that lead with price and go quiet when you ask for a test report.

One last thing

The taxi-top LED market is crowded. Everyone claims to be a “leading manufacturer.” Some are. Most are trading companies with a good website and a WeChat contact at a factory they’ve never visited.

If you’re in the middle of evaluating suppliers and something in this guide sparked a question—or if you’ve got a fleet installation challenge that doesn’t fit a standard answer—feel free to get in touch. We’ve been building taxi-top screens for 10+ years, and the one thing I’ve learned is that the right screen for your project depends on where you’re deploying it, not what looks good on a spec sheet.

Ready to talk specs? Contact us for a quote and a sample unit you can actually test before you commit.