Here’s How Automakers Design Test-Car Camouflage To Hide Precious Secrets


If you want to build the world’s newest and most exciting washing machine you can test it in a lab until it’s time to release it. People do not treat cars like they treat washing machines. People do way worse things to their cars which means you have to test a new car out in the real world where everyone has a camera connected to the Internet in their pocket. Adding camouflage to disguise a test car is not a life-and-death matter in the same way hiding a tank is, but as a car designer I took the task just as seriously. Here’s how my team did it.

Cars are by some margin the most complex consumer grade products you can buy. Think about it; at least 3000 individual parts made of different materials; various alloys, plastics, rubbers and fabrics make up intricate systems that all have to work together seamlessly to make sure the thing doesn’t rattle itself apart, keeps its occupants safe and comfortable and fulfill its intended function of fitting into your life and enabling you to get shit done with a minimum of fuss. It has to get by on minimal maintenance, hopefully be nice to look at and easy to operate, affordable to purchase and be stamped out by the thousand on tiny profit margins with no variation in quality. Oh, and if it will start every morning that’d be peachy. The modern passenger vehicle is an engineering miracle.

It’s Hard To Camouflage A Car Without Affecting Its Functionality

Lr Def 110 20my Gaydon 100919 02

Successfully achieving all of this is why it takes four or five years to get a car from swishy sketch to something you can walk into a showroom and plonk down a check for. Getting there takes an enormous engineering effort and countless hours of testing both real and virtual, in labs and workshops and on the private test track. But, at some point, the rubber quite literally has to hit the road.

With a washing machine, so long as it doesn’t leak, can complete a prescribed number of wash cycles over its lifetime and doesn’t explode when you load it with knickers and detergent you’re probably good to ship. This can all be tested in a secure lab away from prying eyes. Prototype cars need to be tested, tweaked, calibrated and generally flogged to death in extreme climate environments that would have sane people booking the next flight back to civilization. Test drivers try to kill a car to make sure it doesn’t kill you.

Lr Def 90 20my Arjeplog 100919 03 Lr Def 90 110 20my Arjeplog 100919 01

Initial road testing and component validation will be done on private test tracks, secluded from the gaze of people who want to splash your new model over their front-page years before you’re ready to show it. Hacked up development mules are built to test the engineering components, usually under whatever existing body can be butchered to fit, but eventually you need to start driving the actual car on real roads. In the past, scoop photographers needed insider knowledge, patience and a very long lens; these days smartphones and the internet have turned watchful enthusiasts into new model paparazzi.

Everything concerned with the appearance of a car, whether interior or exterior, falls within the purview of the design studio, that highly secure sanctum buried deep within an OEM’s R&D center. Designers are even responsible for the camouflage used to disguise a new model’s appearance until it’s time for the grand reveal as a climax to a long marketing strip tease. The problem is the designers and marketing team want to hide as much as possible; the engineering and test teams want to hide as little as possible, because anything attached to the sheet metal affects its performance and test results. Want to cover up your trick new air vents? Yeah, you’re impacting the performance of the cooling system. Hiding that spoiler? Now the aero is fouled up. See the problem?

Here’s What Goes Into Prototype Camo

Lr Def 90 20my Dolomites 100919 01

Prototype camo consists of three main parts. Vacuum molded hard parts riveted to body work to hide the overall form, foam pieces stuck to the bodywork to confuse the surfacing, and the wrap which covers the whole thing in a psychedelic pattern meant to fool eyeballs and cameras into misreading the shape. Metal mesh and ABS plastic sheets will be slapped over vents and light graphics to cover their shape without sacrificing too much functionality.

When the job of designing camo lands on the junior designer’s desk (this was one of my first jobs upon getting hired), the first thing to understand is exactly what parts of the car need to be hidden. Our new car was extremely distinctive and was also probably one of the most feverishly anticipated new releases in years. Everybody and their dog wanted to know what we had up our sleeves – we knew it would be snapped the moment the first prototypes rolled out of the gates.

What We Tried To Hide

By this stage you have the final production data (there’s always last-minute tweaks for tiny detail stuff), so you drop that into rendering software like Autodesk VRED and once you have your standard three views (front and rear three quarter, side view), you can simply Photoshop over the top some ideas for the hard parts. We needed to hide our distinctive shoulder, hood shape and roof line. So I mocked up some long pieces that would sit below the window line and square off the shape of the top of the body side, a big clamshell to cover the entire hood, and a wedge shape part that would sit at the top of the windshield to hide the dipping roof line.


Once these have been reviewed by managers and approved, it’s off to meet the prototype team to see if the disguise is feasible, and doable within the budget they have. The number of actual development vehicles on the road will depend on the size and scope of the test program. Our new car needed to be tested on and off road extensively, sometimes to destruction. The test fleet was about 100 cars, deployed all over the world. And they all need their appearance hidden. These cars were what is known as the pilot builds – assembled in a special dedicated facility to allow production line managers to figure out how to actually bolt the thing together and work out the procedures the line workers would follow to assemble customer vehicles. Because these cars are probably something like 95% what customers can buy and prototypes unable to be sold, using them as test vehicles gives them a perfect second lease of life.

The light graphics on our car were unique and unlike anything else on the road, and as a design team we were rightly proud of them and didn’t want to show them until launch. Luckily, you can hide a lot of the actual lit element and still remain legal, as long as you have all the functions visible. The problem was our headlights were not flush to the body work – they were inset with quite a lot of depth and a complicated shape. We covered them with clear plastic and then extended the wrap to cover the top and bottom, meaning only a thin lateral strip of headlight was visible. A similar approach was used at the back with the taillights. Because the indicators and brake lights were contained in the outer elements, we were able to cover most of the main lamp with a sheet of ABS.

Some manufacturers have a bit of fun with the wrap and use humorous imagery or try to make their car look like someone else’s. As the car heads towards reveal the wrap might be changed (or have stickers added over the top) at the behest of marketing to start building social media buzz. We just used a generic swirl in black and white, probably because it had been used before and there was still several miles of the stuff still lying around. There are still further opportunities for deception, though. Once the hard parts and the wrap pattern are decided you can play around with blocking out graphics in black spray paint – to trick people into thinking your design has shapes and visual features it doesn’t actually have. I spent a few days Photoshopping various wheel arch shapes, grille patterns and vent openings in the wrong place onto a render of the wrapped and disguised car for approval. One alternative I mocked up had a seven bar grille as I thought this would show we had a sense of humor about one of our main competitors and be good PR. The studio chief had a good laugh and agreed with me before saying “no.”

Usually wheels will be painted black as well to hide the design. We were horrified inside the studio when our car was papped with one of its signature alloys unpainted. But this is the problem – people involved outside of the studio are not as concerned with keeping a new car’s appearance a secret as the design team are. Engineers generally consider designers a nuisance at best and one step up from the lobby flower arrangers at worst. One old hand on the test drive team thought he could bully the new designer (me) into leaving all the car’s glazing unwrapped, quite shirtily suggesting I didn’t know my ass from my elbow when I said the windows needed to be covered as much as was possible. Wrapping the glazing not only hides the DLO shape, but also keeps the interior hidden. Obviously you can’t cover the windshield, but rear side glass and rear windshield are fair game, as long as the drivers can see out safely.

The Camo Really Does Help Keep The Design Secret

Lr Def 110 20my Tusk 100919 04

So the first cars emerge for hundreds of thousands of miles of punishment like a wheeled Scooby-Doo villain, their true nature shrouded in mystery. These days it generally doesn’t take the forum detectives too long to figure out exactly what they are looking at, but even as testing progresses and the hard parts slowly have to be removed (to get more accurate results) the wrap is enough to confuse even the most dedicated Photoshop amateur. Our design remained secret right up until a contract painter snapped a model in a paint booth one weekend, just days before launch.

I became the go-to guy for camo in our studio, and the next car I worked on had a sister car that was about a year ahead in development, which was already up and running around covered up on public roads. Because our version was a type of car we had never done before, we were REALLY keen to keep people guessing. The sister car was a BEV like ours, but the press had been rumoring ICE versions were being considered. A quick meeting with the studio on the other side of the building confirmed yes, we could use their hard parts on our car. The game was afoot. Not only would those parts make our car look like one of theirs, but we could also mockup some fake exhaust tips to make it look like the non-existent ICE version the press was speculating about.

The best part was because we were tight with the DVLA (the UK equivalent of the DMV) we could register and plate our cars as those of our sister brand as well, meaning that anyone looking up the registration number would be looking at factually incorrect information from an official government website. Ultimately all this fun and games was for nothing as both were unceremoniously cancelled at the 11th hour before they’d even got out into the open, which for any designer is the real kick in the teeth.

Share on facebook
Share on whatsapp
Share on twitter
Share on linkedin
Share on reddit

62 Responses

  1. This was a great writeup and is tied into something I’ve always wondered. How are these vehicles legally allowed to drive on public streets? I’m assuming you can’t register and insure a car that technically doesn’t exist yet, so is there some sort of special licensing for prototype vehicles?

    1. Yes, they’re registered as prototypes, they have to have additional safety systems (you’ve seen the big red ‘kill switch’ in interior spy shots) and only specially trained and licensed drivers are allowed to operate them.

      1. In the US? No special license required to operate a prototype or self-built vehicle. As long as you’re rolling on manufacturer plates, you can drive a completely bare tube chassis that would never even come close to passing crash testing – forget lighting – around.

        And nearly every state issues manufacturer plates. I honestly have no idea why the scammers at Lordstown have been getting MI manufacturer plates. Ohio is a permissive at-request state; two page form, $29.75 made out to the Ohio Treasurer of State, and as long as it’s untitled or registered to you, you’re good to go. Don’t even need a dealer license.

      1. I saw on Twitter recently a picture of the Weinermobile tackling the corkscrew at Laguna Seca. I’m guessing the driver was sitting on a warm soft brown cushion.

  2. I like those stickers they use in some stores and restaurants, with the holes you can see out of while inside the building, but you can’t see in from outside. They should use those for camo on the windows.

    And also, I still think that square behind the back doors of the new Defender is stupid.

    1. It’s a bit of a visual flourish, and on the 90 it’s optional. But it does give you additional storage on the inside.

      On the 110 there is body in white that would be visible if it wasn’t there. But on both cars it does visually link the rear wheel arch and the roof, which is one of the critical things in defining how a cars stance.

  3. When will BMW remove the camouflage from their current generation of cars so we can see them without their hideous faces? Presumably that is camouflage, because who would design something that ugly?

  4. This was fascinating – since this interwebs thing took off, I’d wondered how this is done. Thank you Adrian!

    Question: sometimes we see “spy” shots where the wrapping seems more like a velcro-ed on car cover with cutouts for the windows.

    Is it bad photography, poor-quality wrapping, or are there additional covers for things like last minute testing, using on multiple vehicles, or ??

    1. Yeah, that’s another way of doing it – it seems to be more of a thing US OEMs do I think. They use a kind of ‘bag’ cover that is loose fitting and has mesh over the lights.

      1. It does a great job of hiding the shape of the body, that’s for sure. I thought that automakers tended to use that stuff earlier in the testing, and then switch to a more form-fitting wrap near the end when the car needed to be as close as possible to what customers would actually buy.

      1. That makes sense, but I’ve always found it hard to believe that the attention-grabbing effect of the dazzle patterns doesn’t largely cancel out their usefulness. It’s like if you created a chemical that attracts mosquitoes, but also messes up their sense of direction. Would it work to prevent mosquito bites? Maybe …

        I am skeptical of camouflage strategies in general ever since that minor scandal in the US military over the use of pixellated “digital camo”. It’s pretty hard to actually PROVE that a camouflage design works well.

        1. The pattern itself is kind of irrelevant. It’s more that it’s repeated mostly seamlessly and non-glossy (so it doesn’t create highlights) that gives it it’s effectiveness.

          We did look at doing dazzle type camo, and I researched it and you’re correct in that in trials on warships it’s effectiveness in hiding ship size and range was not proven.

          1. Basically they spent three billion dollars developing One Camo To Rule Them All, issued it to their troops, used it for years, and then discovered that it didn’t fucking work worth shit. In a sane world it would have been a much bigger scandal, but alas we do not live in such a world so it was quickly drowned out by other, even bigger ones.

      2. nah, wrapping never fooled me. The only things that work is when they glue inserts on the car and still use some wrapping on top. I took pictures with my 5dM3 of prototype cars, wrapped in matte wrapping and, if you zoom in or use big screens you can see the details perfectly… Even on low definition pics like you see on the ‘net is not too hard to see the details.
        I believe the industry overestimates the impact of dazzle camo wrapping, it’s quite ridiculous.

  5. And sometimes an automaker really wants the paparazzi to pay attention to their camouflage – like when MG Rover wanted to make everyone think development work on the RDX60 program was continuing on track, during a loll in negotiations with SAIC, so they taped a bunch of garbage bags to a Rover 45 and drove it around where they knew it would be noticed by photographers. Got published in Auto Express as a hot scoop on the new RD60, so it worked, as long as you don’t define “worked” as stopping the collapse of the company, because it didn’t do that.

  6. I had a buddy that was working on the Ford Flex program. They were thinking of putting a woodgrain wrap along the sides of some of the late prototypes as a joke and see what the press would say. Unfortunately it got nixed by some manager before they could do it.

    1. The number of easy cheap wins that get killed by corporate sense of humor failure never ceases to amaze me. That would have been brilliant, and the sort of thing I would have suggested.

  7. This leaves me longing for an article about pre-production mules. Have always found them fascinating. I remember the Bronco mules being a bobbed Ford Ranger and a single cab Raptor. People were crying about how ugly the Bronco was going to be (people with half a brain knew what was going on there).

    1. I think THIS should be a series. “Weird Production Mules” We could start with the ridiculous stretched Ferrari 355 thing they used as a testbed for the Enzo.

  8. “We needed to hide our distinctive shoulder, hood shape and roof line.”

    …. this is a perverse definition of the word “needed”.

    What would have happened if you didn’t? You would have lost control of the message? Horrors.

  9. good stuff. Always good stuff from you Adrian. Thanks. I once had in our circle of friends the chief designer of the Hyundai Genesis 2015(?) and it was fascinating talking to him. Designers are their own breed of people aren’t they, almost like ascets of a religion, aren’t they?

    1. Well, I’m a confirmed atheist, but yes it does take a certain set of skills and sensibilities to succeed as a designer (in any discipline).
      It’s not enough to to just like something, you have to understand why you like, and be able to describe what works and what doesn’t in terms of what that design is trying to achieve. This is in addition to the vocational abilities needed; sketching, rendering, modelling and so on.
      The unfortunate side effect of this is it places the designer as having better taste and aesthetic judgement that someone who is not a designer – and to a degree this is true. If I wasn’t able to create a good looking vehicle or component I wouldn’t be able to do my job. This is slightly what irks me when people say all design is subjective; it’s not. You might not like something I or any other designer does, but that doesn’t mean it’s not an attractive piece of work.

      1. “This is slightly what irks me when people say all design is subjective; it’s not. You might not like something I or any other designer does, but that doesn’t mean it’s not an attractive piece of work.”

        This raises a two-part question for me: have you ever had a situation where you looked at a design — yours or someone else’s — loved it, other designers loved it, and the target market said “it’s OK….I guess….”? Or conversely, did you look at a design, see a bunch of flaws, but have it end up really good looking?

        And I guess that raises another question: are you able to turn the designer in you off when looking at a car?

        1. And, because I can’t edit, thanks once again for a fascinating read, and for actively discussing it with us Autopians. I find your content on this site very enlightening.

        2. Yes, any designer should be able to give and received considered notes and criticism. At studio level, you know it’s professional advice and not arguing with uninformed fucknuggets on the internet.

          The BMW i3 & i8 are my go-to examples of cars that win design awards, are critically acclaimed and held up by every designer as fantastic pieces of work. And yet the market was ambivalent, because they were probably a step to far in terms of appearance for mass acceptance.

          Yes I am able to remove the black turtle neck and enjoy cars on a purely subjective level, but nearly everything I like has something interesting about it in it’s ideas or appearance.

          1. Do you really think it was the design holding the BMWs back? I always thought it was the actual performance. The i3 had terrible range, and i8 didn’t have the performance to match its incredible looks.

            1. I particularly agree with you on the i8. I think it is a wonderfully space-age looking exotic with an exotic price to match. But the performance and electric only range were just really disappointing for that price point.

              The i3 (which I just happed to see on the road this morning) was mostly let down by looks. Its electric only range was descent for the the range-extender version. But it’s looks were just a little too bulbous for the average BWM buyer. At that price, you really have a lot of people that are simply buying the roundel and they want it on a pretty package. Will be interesting to see how the beaver-teeth grille impacts long-term sales of their new cars.

          2. Just like I can go to a car show, see an exquisitely designed and built custom car and appreciate the effort that went into details and the car, and yet still not like it overall.

            1. Yes, the level of craftsmanship and love that goes into customs is sometimes staggering, but somehow the build doesn’t quite hang together. I think this is for a couple of reasons. When you’re using aftermarket parts they’re quite often garish and not designed for an individual car. Following on from that, custom builders have to what the person paying the check wants, and the builder is unlikely to push back against that. Thirdly, builders are not designers (despite what they might say). It’s a different skill set (Foose is the exception because he is a trained car designer, but I don’t think he has any OEM experience).. Being a designer means you look at the whole car and find a theme that works throughout the whole car, which when you are using off the shelf parts is difficult.
              A lot of restomods I find to be overwheeled, because the assumption is big wheels better, but cars from the fifties, sixties and seventies were not designed around 20” rims and lower ride heights so they look wrong. Also with a lot of these cars builders never attempt to fix the bad shut lines and panel gaps they had from the factory.
              For my money someone like Ring Brothers are the ones doing it right, but I appreciate those are super high end.

      2. Adrian, building upon the “design is subjective” discussion. I’m curious about your take on design and how it is accepted by the public in relation to traditional architypes. My curiosity is based on discussions I’ve had with other designers in a couple of fields, and I haven’t met any industrial designers with an automotive background beyond the college level.

        As an architect I’ve found that some of the more exciting buildings are what I would classify as architecture for architects. It tends to be too bold and is misunderstood by most outside of the profession. Trying to gain a better understanding of what is generally accepted by the most people. It boiled down to 2 things.
        1) A persons need to compartmentalize or label things. This falls into building types keeping traditional elements and easily understood to fit into it’s architype.
        2) Having a design that holds a timeless quality. I see this as also being relatable to designs of the past, yet also holding a quality that the design will age well. This is hard to predict and often can’t be judged until many years later. For building material choice can play a big part in this.

        Back to how I see this relating to automotive design…. If you take out of the equation the capability of the vehicle, and just consider the form and styling of the vehicle. What’s you take on automotive design in relation to architypes and the general publics perception of what a car should look like? Are there any car designs that you can relate as being more exciting to other car designers, but not generally accepted by the public? Do you feel any constraints to meet expectations of architypes or traditional design?

        This might be tough to sum up in comments. Maybe something that leads to a future article.

        1. Short answer; pick up any credible book about car design, something like ‘Fifty Cars that Changed the World’ from the Design Museum or Stephen Bayley’s ‘Cars – Freedom, Style, Sex, Power, Motion, Colour, Everything’ and probably half of the cars mentioned will have been abject flops (or at least didn’t make any money). I have tangentially covered some of this in my column.

          Long answer: Lemme think on it and I’ll probably write it up here.

  10. The only reason any of this is necessary is because of the late Jim Dunne, who used to stalk Detroit and always got the scoop photo on new models. They started doing camo to foil him.

    1. In Europe new models certainly were getting scooped in the early sixties. Car had the new Hillman Imp on their cover yers before it was due for release, but yes even in the late fifties Detroit wasn’t disguising new models. Chuck Jordan saw the ‘57 Mopars lined up outside the factory and was so shocked he went back to the Tech Center and grabbed Mitchell. The result was all the ‘59 GM cars were redone in Earl’s absence.

  11. Not that you’d be at liberty to confirm, but I assume the canceled sister cars you’re talking about are the Jaguar J-Pace and Land Rover Road Rover, which were apparently full BEV vehicles and not ICE/hybrid vehicles, as first suspected.

    This is a great article. I enjoy seeing pictures of the hacked-up mules. My favorite was the artful mule Rolls-Royce did for testing the Cullinan’s running gear and powertrain, which was a Phantom VII with a considerable amount of length taken out of its wheelbase, a jacked-up ride height, and a giant wing for aero. Rather than make the Cullinan mule a shameful thing, Rolls-Royce got rid of what I’m sure were ungainly seam and spot welds, worked it over in satin black paint, and gave it a renegade persona all its own, christening it “Project Cullinan.”

    It doesn’t seem that anyone ever camouflages the mules, since they’re based on existing production cars.

Leave a Reply