A while back in my Damn Good Design series, we looked at the much-maligned Pontiac Aztek and explained how the poor thing started out as a promising idea with a reasonably snazzy concept, only to completely shit the bed by the time the production version appeared. GM wanted a sexy multi-purpose vehicle that combined the handling of a Camaro and the practicality of a Blazer. Instead of the best of both worlds, they ended up with the worst of all worlds, because the Aztek was doomed to use the platform of a minivan. The design of the version you could buy was compromised from the very beginning.
GMs problem was they were beholden to market research that told them they needed to make more innovative and exciting cars, without really understanding what this meant. Like their cross-town rivals Ford, they were wedded to a management style that attempted to apply quantitative measurements to qualitative characteristics, leading to nonsense statements like the one made by then CEO Rick Wagoner that 40% of GM’s new cars would be “innovative.”
Thus, despite it crashing and burning in early customer clinics, the Aztek was greenlit and progressed into development. GM convinced themselves because the Aztek hit all their internal metrics and there wasn’t anything like it on the market, it would be a smash hit, and they didn’t want to hear otherwise.
The Aztek Was Not a Bad Idea
Identifying a gap in the market or using existing vehicles as an analog for your ideas are both valid ways for product planners and car designers to come up with proposals for new types of vehicles. When I was student, Jaguar was expensively attempting to become a British BMW. The 3-series rivaling Jaguar XE had finally launched, and so penning smaller, hatchback Jaguars along the lines of a Mercedes A-Class or BMW 1-series was a popular project among my less imaginative peers.
I was never entirely convinced of the merits of this idea, but later when J. Mays was my teacher, he instructed me to identify gaps in a manufacturer’s range: Don’t look at what they have, look at what they don’t have. When I worked at Gaydon, I proposed repositioning one of the existing models as a Land Rover version of something well-loved from another manufacturer’s past. It made perfect sense to me on a number of different levels, but obviously such heresy couldn’t be countenanced. But that’s what designers are paid for: Part of the role is to have ideas and then turn them into compelling visuals so others higher up the decision-making food chain can judge their merits from a business point of view.
There was nothing wrong with the Aztek as an idea: a chunky, robust-looking vehicle with a fun-to-drive on-road bias that was rugged enough to handle some light off-roading. The market at the time offered full-fat body-on-frame SUVs like the Blazer or smaller compact unibody crossovers like the CR-V or RAV4, so it’s not hard to understand GM’s thinking: The Aztek would be something to slot in between these two extremes. The easy and typical GM way of doing things would have been to give the Blazer a Pontiac-style glow-up similar to how they crapped out the Oldsmobile Bravada, but the GMT330 platform was on its last legs and the American car-buying public had long since started seeing through such nose and tail badge engineering jobs. Besides such a lash-up wouldn’t have been “innovative” and a traditional SUV like that would not really have fit in with a Pontiac line-up whose tagline at the time was “we are driving excitement.”
Let’s assume then (and we don’t know really if this was true or not) there was no business case for the Aztek without basing it on a minivan platform. The numbers guys punched all the data into the GM-finance-o-tron 5000 and putting a new body on the GMT330 platform wouldn’t have worked. It was too old, too agricultural to give the desired driving characteristics and economy figures, and it wouldn’t have supported the interior room and flexibility the concept promised. If the Aztek was going to be greenlit, it had to be on the existing U platform.
Let’s See If We Can Fix It
So although it would be relatively straightforward to remedy the proportional crimes of the Aztek by simply setting fire to the thing and starting again, that’s not what we’re going to do.
Drawing something fantastical and then complaining that engineers and accountants have screwed up your perfect form is not really what being a car designer is about. Being a designer means being a problem solver: How do we make this thing we are working on as good as it can be within the constraints we’re operating under? In other words, totally redrawing the Aztek to make it closer to the original brief of a Camaro interbred with a Blazer would be an easy cheat.
What I’m going to do is make changes based on what I suggested in my earlier article. To try and improve it while keeping the underlying platform and hard points. Could a few relatively simple tweaks turn the Aztek into something that looks a lot better that wouldn’t have put customers off? Is there something better lurking under there, or was the whole thing a fool’s errand from the start?
Here are the annotated images highlighting the basic problems from the original article:
Although the proportions are not great thanks to the underlying hardpoints not being ideal, there are ways to mitigate the damage. Although getting the underlying shape correct is important, you can get away with a slightly awkward volume if everything else works to help the design out.
The Aztek is basically what designer wankers like me call a monovolume. Although it’s not the sexiest outline to ever leave a designer’s sketchpad, the one-box shape doesn’t automatically mean you end up with an ugly car. The original version of the Renault Espace and Dodge Caravan might not be objects of lust but they have a pleasing functionality and honesty about what they are. Even the GM Dustbusters were interesting and tried to move the category forward. The problem with the Aztek is the details: the glazing, stance, and graphics are all misjudged, which exaggerates the problem with the shape rather than hiding it. So let’s try and fix them.
First of all, here’s an unaltered side view.
Starting at the back, I’ve removed the underbite rear bumper, smoothing it off. This has two benefits: It removes visual weight (and actual physical weight I guess) from the lower part of the car, making it look lighter on its wheels and 2. It increases the departure angle, making it look more capable off-road.
I’ve sized the wheels up slightly, and very slightly moved them up into the wheel arches. This fills the arches out better, improves the stance (the way the car sits on the road), and better balances out the overall height. It no longer looks quite so tippy toes.
Let’s move round to the front:
Have You Had a Nose Job?
Again, this isn’t a great angle. Or rather, it’s a good angle for making the car look better but not for illustrating what I’m trying to show, which is the sogginess of the front lighting and grille (known as the Down the Road Graphic or DRG). The base of the hood is quite a bit higher than the base of the screen (almost certainly one of the hard points that was dictated by the platform). But if you look at where the hood meets the windshield on the far side of the car, you can see there is a big gap there – this means we can lower the height of the rear of the hood, so it doesn’t dive towards the front quite so aggressively.
I’ve flattened the line of the hood out and pulled the lighting and grill up. If you’re struggling to see the difference, look at the gap below the main grill to the black lower bumper, and the gap between the headlights and the indicator units. I’ve also pulled that lower bumper in a bit, so it no longer juts out like a teenager’s lower lip when you tell them they’re grounded.
I normally hate yellow. It works on exactly one car (the 1971 Buick GSX), but it is the Aztek’s “hero” color. Unfortunately, I couldn’t find images without the body-colored cladding we could use. So I’ve added grey trim to the strake on the lower doors and removed the paint from the door handles. This helps break up the vast expanse of yellow on the bodyside, saves a bit of money (which I’ll be spending later), and looks more robust. I’ve also helped the seller of this car out by removing the scratch in the front bumper (below the headlight in the original image if you can’t see it).
Yes. And a Butt Lift.
My main problem with the back of the car apart from that horrible bumper, is the position of the split in the tailgate. I think it’s just too low, adding the feeling of bottom heaviness affecting the whole car. I’m going to pull that up so it meets the feature line in the bodyside and get rid of the bumper.
The split on the vertical section between glass and yellow bodywork is now closer to thirds, rather than the 50-50 split it was originally. Thirds are a more pleasing visual composition, which is why the rule of thirds exists in photography. Moving this split up also has the additional benefit of lining the bottom of the tail lights up with the feature line on the bodyside, making for a cleaner and more harmonious look. Finally, the money I saved on not painting the door handles and lower strake I’ve spent on giving the Aztek a proper fuel filler cap.
In all of the three altered images, I’ve also tweaked the lower line of the third side window (the one between the C and D pillar) to make it higher and more sympathetic to the line of the rest of the Daylight Opening. I kept going over and over my changes here, each time thinking I hadn’t moved it up far enough. And I’m still not sure I’ve got it right. It’s one of those things in the studio you would iterate on a full-size model over and over, probably printing out tens of different full-size versions to stick on the clay. I was tempted to angle it slightly, but that wouldn’t really have been the fashion at the time.
Given the constraints, I think these (relatively) minor changes make a big difference. Whether they would have made enough difference to overcome the Aztek’s stiff purchase price and cringey marketing is another matter entirely, but to repeat a line I recently made in the comments: a good design can save a bad car, but a bad design will certainly sink a good one.
And anyone in the comments who says “It’s the same image” will instantly earn my scorn. Don’t make me get on a paddle steamer again.
Authors note: It will not have escaped your attention we had a monumental publishing omnishambles gentle editorial imbroglio last week. Now I’ve removed the knives from my back and scrubbed the bus tire tracks off my torso here’s what happened: David was out doing Important Business Stuff. Matt was in London trying out a fallback career as a London Cab driver. That left Jason and Peter at the controls. Jason as usual was out of his mind on bathtub bourbon and illegal Mexican painkillers he still insists are for medical reasons. I know it helps to be wasted to work here but some of us are trying to be professionals. And Peter was … where was Peter? Who even is Peter? I don’t think that guy actually exists outside of being an AI chatbot in Slack.
I wrote something that was published (I just upload my content into the mainframe. I don’t have control over making my stuff appear on the site – for the moment my power is limited). David returned, saw the article, and promptly gave birth to several new kittens. Faced with having to alter the piece as he saw fit or rousing me from my coffin at 3 am UK time to make a few tweaks, he broke out the editorial chainsaw (it’s kept next to the battery removal chainsaw in the Autopian chainsaw cupboard).
Needless to say, when I awoke and saw what had transpired I immediately bought a first-class ticket on the next transatlantic paddle steamer so I could let my views be known in person. In the spirit of full transparency then, here’s the footage of David and I discussing the matter in the Autopian war room:
I’ll leave it to your imaginations to figure who is who.
I liked the old gas fill, the more the car looked like a certain tool brand, the better I liked it. As much as I mocked the Aztec, I can’t deny that my Element owes some royalties to it. Although I waited till 09 to get less fading black plastic.
Funny, I hated this thing when it came out but hate it less so now, particularly without all of the grey cladding. I do like the tweaks.
That said, I have some fond memories of the Aztek as that was what I was driving when I launched a wildly successful artisanal, locally sourced recreational pharmaceutical company in Albuquerque some years back.
It just hit me, thinking about the stupid two grills thing it has going on (and that GM/Chevrolet seem determined to outdo, with 5 or 6 strata of grills smashed into one fascia).
Pontiac themselves made a substantially defucked Aztek. It was called the Vibe.
I am shocked at how just removing the buck-tooth overhangs in the front and especially rear in the first image set make it look ~50% of the way to acceptable already.
But I am one of the very, very few who doesn’t actually hate the Aztek as-is, and frankly I think it’s unapologetic homliness is superior to the intentionally-and-purposely-designed-yet-still-accidentally-ugly that makes up most of the modern CUV market.
I’m surprised you didn’t do something about the clunky shapes of the fender bulges, particularly the front one. Oh well, at least the monochrome versions of the car draw less attention to them than the 2-tone versions.
My biggest takeaway is that The Autopian be has TWO chainsaws. That’s a prime example of all that’s wrong in this late capitalist world of excess and the gulf between the haves and the have nots.
The redesign looks much better, much less droopy. The front still doesn’t look good, but it’s a look BMW picked up and ran with 20years later.
I’d love to know how much branding can overcome bad design with respect to sales, it seems to me to be (depressingly) a massive factor. Would the Aztec have done better, and been perceived to be less ugly, had it been a BMW X2 for example?..
BMW couldn’t have done something like this at the time. The link between branding, design and sales is something I’m thinking about for a future piece.
I’d be interested to read it. It seems ‘premium’ brands are far less affected sales wise by the quality of their designs, BMWs range is the ugliest/worst it’s ever been and yet last year was their best ever for sales numbers or profit (I can’t remember which of the two it was). Where as mainstream brands seem very dependent on design as a way of maintaining/gaining sales.
I hope this will become a regular column from Adrian. Next you could do the last generation Prius (it reminded me of a catfish somehow) or the Marai. In fact, a lot of Toyota’s recent output is a bit fugly.
The last generation Prius convinced me of two things: The Japanese can out ugly anyone when they want to, and: The Japanese apparently *do* use psychedelic drugs. At least, the Prius designers did. Then they went and ruined it all by producing the perfect Gen 3 Volt. Huh.
All I see is a shorter rear bumper and plastic handles/door guards. We need a slider with both cars so we can see subtle changes. I’m not saying the changes are too subtle, but they’re too subtle when separated by text, 4″ and blurry beer vision.
Sorry I completely forgot what you were saying after visualising the image of you and David fighting.
I’m a lover not a fighter [presses ice pack to nose].
It looks like the Tuesday 11 o’clock TV reenactment, where names have been changed to protect the identity of the individuals involved
Adrian being Adrian, we would not expect to get an animated before and after photo lest incur the wrath of Adrian
Those are usually done by Peter, who doesn’t actually exist.
The improvements all work except the nose. While the nosejob does make a difference, as it stands it needs to be dumpstered. Way too busy for my palette.
Didn’t we just discuss how it was un-cool for the Cherokee (people) to have something named after them?
Dear Adrian, when something is so, so, truly broken, it’s beyond trying to fix. Just throw it out with the rubbish and save your energy The Aztec, along with the original Chrysler Pacifica, were actually trendsetting segment creators. They were considered odd and off at the time they came out ,that said, now most people drive a modern version of them. The Pacifica and Aztec pretty much started the cross over craze. Damm both of them to Hades forever LOL
The front looks like it stacked 2 cars together. Sorry, before and after. You can do double lights or double grills. You don’t go full doubles.
It’s a double double..
The hood and grill openings were one piece on the concept car, and the nose was less blunt. But a side effect was the four openings created a cross bar shape that looked very Dodge-like. https://www.chicagoautoshow.com/assets/1/24/TimelineMainDimension/PontiacAztekConcept@1999Web22.jpg
Also, the hood cut line on the show car created a line that extended along the bottoms of the windows, creating a slightly arched flowing line — which explains why the bottom of the rear side window is lower. The leading edge of the rear fender bulge was originally more angular, like the front bulge. But when the arched line was lost I think they tried to make up for it with a more curved bulge.
But it could be it was decided the concept looked too sleek for an adventure-ish vehicle. “Make it boxier, clunkier, more utilitarian, more butch.” “You mean uglier?” “Yeah.”
This might have helped a bit if you’d been called in at the 11th hour with all the tooling money already spent.
But I think the only real fix would’ve been to swallow their fears of it being perceived as an old-fashioned station wagon and jettisoning the minivan platform in favor of the W-body sedan one as soon as a model-specific cowl height was removed from the question.
Again it’s possible they considered it but for one reason or another it wasn’t suitable.
Great idea. 11th Hour Design should be an occasional feature.
I would love to see a series like this, where there are obviously severe limitations about what can be changed, yet Adrian shows how much better a little applied knowledge, taste, and skill can make an existing design.
We’ve seen his design and redesign skills, but I really like the idea of isolating it like this. Especially the limits. It ups the challenge if you can’t just throw out the parts you don’t like, or add in stuff that would cost hundreds of thousands in retooling.
I know we already had “Altered by Adrian”, but his changes to the E-type included some major structural alterations that could only happen at the beginning, or during a major rework. I still like those. There’s space for both here.
Maybe we can make a version of the Bat Signal to call Adrian to the job, except with two crossed pens or something like that…
I never though the Aztek was THAT ugly, and the later versions like this model look so much better with just a bigger/ more interesting wheel/ tire package. Your small tweaks make a big difference and emphasizes the importance of proportion and nuance in design. Good job!
Oh good, do the Civic Type-R next because that sure could use some fixin.
The current one?
Every car would benefit from changing wheel color from black to silver.
I’m just glad to see you back.
I never went away.
Then we need more! More Adrian! Daily posts. More Adrian! More. More! MORE!
What am I, some kind of performing monkey for you people?
No!
You’re a rock star and we treat you like such. (Just, please don’t smash any guitars or Morgans.)
Oh, and please don’t do a tour with the Spice Girls.
More like a dancing goat.
That is much improved, and definitely not the same picture dot gif the way past do-overs were.
It still leaves almost intact the inexcusable front end. The bottom lights and grille are Pontiac, but too short and too low (to make room for the upper lights). The upper lights are floating disembodied incomplete lighting units. The gap between the top and bottom lights isn’t justified. It just makes both sets of lights ugly.
I’m 83.4% certain the front end alone killed 91.6% of sales.
when I do a re-design, 60% of the time it works every time.
You’re being too hard on yourself. Based on these articles I’d say at least 74%.
I would love to see the original designs before bean counting and autoesthetics.
I liked the show car at the time and was shocked and disgusted when i saw the production version.
Sorry they are both ugly. I can’t imagine the rear view with your fixes. And the front of both are hideous. Doesn’t some design rule require quad headlights to match? Shouldn’t the grill be separate and not continue in the hood? BTW a Proton Yellow Isuzu Vehicross is the very best Yellow car ever and what you should have used to design a better Aztec, which is now so ugly that I want one.
The changes shift the Aztek’s appearance from weirdly ugly to weirdly boring with greater susceptibility to suffer tailgate and headlight damage from minor scrapes. Look at its front end. Most of it is plastic bumper cover. Lowering its stance an inch or two and maybe a mild refresh of the front end. would’ve mitigated 90% of the ugliness.
The biggest problem was that Toyota Highlander and Honda Pilot came out about the same time and did everything the Aztec was supposed to.
They don’t have the tent attachment nor cameo on Breaking Bad.
Like many GM products the Aztek was a poor execution of a good design. The Honda Element was a successful execution of the Aztek concept.
While I think the best fix for the Aztek is the Monty Python 16 ton weight, Adrian’s sketches are an excellent lipstick on a pig job. It looks much better but it’s still a GM minivan underneath
I like it. I like other redesigns more, but I think this is the first one I’ve seen that tries to work around hard limitations of the structure instead of just throwing it all out, coming up with something off the rails, and saying “SEE GM I MADE IT GOOD”
There’s a hack with a YouTube channel and no car design training or experience who does something like that. I’m sure I mentioned him recently…..
Your mention in the original article was spot on.