Everything Was Blue and Nothing Matched
A beat up box of samples, six weeks in Spring, and a color system we created via mail.
Late March and there on my doorstep was the Frankenstein box. A jumble of two boxes taped together to contain its miscellaneous treasures.
Our client had sent us everything they had ever made, and I mean the physical objects, not the files. Business cards. A koozie with an attached carabiner. Front of pump assembly machined bezel off. Another bezel, longer, black anodized, with blue fill. Vinyl decal. Roll-up, rubber-banded, pull-up banner. There was a rack wrap the size of a door in there.
We laid it out on the table and counted FIVE blues.
Not five shades of blue? Five distinct blues. The business card was dull, heavy the way ink is on uncoated stock. The koozie was royal, because on a koozie the brand color is not ink, it’s the fabric. The banner was glossy and saturated and looked like a different company. And the product wordmark, the name of their flagship system, showed up in a soft mint on one section of the machine, and a pale washed-out blue on another. Same word for the same product, but two colors that had never been in same room.
Nobody did anything wrong. This is the part people usually get wrong. There was no negligent vendor, no indifferent designer. Instead, there was a brand guide with too many options allowed and too little direction. Add that to a dozen people making one locally reasonable decision each. The powder coater grabbed a green-blue, because the guide didn’t specify which blue. The vinyl shop chose light blue because the guide showed three and didn’t rank them. Each of those choices was a reasonable choice. They looked like five companies with the same name stacked on a table.
Drift is what a system does when no one writes down the system.
Who was it for?
Client develops liquid cooling for data centers Two phase, direct to the chip. The kind of thing that allows a rack of very hot processors to run without drinking a small town’s worth of water. Real engineering. A product where the buying decision is expensive enough that the brand must look like it belongs in the room.
Mid-raise was a factor too, setting the terms. There was an identity there that we couldn’t afford to throw away, and we had six to eight weeks.
Here’s the other thing, with love: they were lost when it came to color. They could tell you the heat flux in square centimeters their evaporators handled, but they couldn’t tell you why their business card looked muddy. Which is okay. They make thermal systems. Color management is a discipline of its own and there is no reason for it to live in their heads. So, our telephone rang. A lot of the project ended up being teaching rather than designing, and I didn’t quite realize that going in.
Mail chips
Deb and I would have loved to tear the whole place down and start over. We said it to each other forty times. But that version takes nine months, and by the time it lands the company has closed a round of funding on a deck that bears no resemblance to any of it.
No budget for a press check, no draw-down at the coater no standing in a shop at seven in the morning watching a sheet come off the press. We were distant from each other and from the client. Every choice had to pass the scrutiny of a monitor, a webcam, and a shipping envelope.
So… we started with Pantone colors. Not because Pantone is magical, but because it’s the only reference in this business that is physical, standardized and the same in two places at the same time. Screens are not consistent, especially over Zoom. With Pantone, we hold a chip in our hand, and the client holds the same chip, so for the first time in the project we’re looking at the same object. It was what their manufacturer had asked for too. Powder coaters don’t want hex codes; they want a number they can see in a book they already have.
Six or seven shipping rounds, over maybe 6 weeks, which sounds slow, because it was. And then we held chips up to the webcams. There’s something so funny about five grown-ups on a video call saying, “hold it closer, “no, turn it away from the window,” “wait, is that the same chip you had before?”
Necessity is the mother of invention, so we created a protocol. Neutral background, same time of day, chip flat & no tilt.
I also ordered too many chips. I was on the floor one time sorting a fan deck and a stack of loose chips I bought on top of it, trying to find the right blue in a set of blues that were, to be clear, all the right blue, and Deb finally said,
“Dave, you’re overthinking this. “Just make up your mind.”
She was right. We laughed. I made the choice. I still have the remains.
Remove then add
The brief was to expand the palette. The first leg was about removing things.
The old brand guide allowed the logo in gray, in white, and in blue, in two layouts, with and without a tagline. That is a lot of doors. And every door is a chance for somebody downstream to go through a different door than the last person went through. We shut most of the doors. We cut the secondary typeface that was doubling down quietly on every type of decision in the company. And we pinned the logo to exactly one blue, no interpretation possible. One blue version for full color, gray on light, white on dark, that’s all. That’s the entire set. This wasn’t a redesign. It was a return to basics.
We kept their navy, mainly because of what it did when we lightened it. Their mark is a matrix of dots in a gradient, so every dot is a tint. A lot of navy blues have just enough red in them that when you lighten them, they slide toward mauve, and you end up with a logo that’s navy at the bottom and faintly lilac at the top, and nobody can tell you why it looks cheap, they just know that it does. Theirs were blue all the way up. When you have one thing that works you make it the center and let the rest negotiate around it.
A couple of sports fans walk into a color review
We got pushback on our first proposal, and the pushback was right. We’d put up a periwinkle, a teal and a yellow. The neutrals landed; the rest were muted and polite, not a company selling the color language of thermal performance.
Sooo… we returned with two directions, one sleek and one bold.
Then the client had the best idea of the project, which was to hand back our own advice. They said,
“Why are we choosing three secondary colors when you told us there were too many choices in this problem?”
Alright. We’d spent a month arguing that the old guide had failed because it offered them a huge menu, then we offered a huge color menu. So, one secondary color it is. And green is the choice, because the whole story of the product is to use less water, to use less power.
That’s when one stakeholder said they really liked the Seattle Seahawks green. And another said they liked the Austin FC jersey green a lot.
They are different greens. One is an electric warm high voltage lime and the other is a soft cool mint you could paint a nursery with. Suddenly, the conversation wasn’t about the brand.
We didn’t fight it. Instead, we took them both extremely seriously. Seahawks green matched to its exact pantone Matched the Austin FC jersey to its exact Pantone, and since that one falls right on a line, found a second option right next to it. Then we added some bright green of our own, which we had been holding since the bold direction. Four greens, one page, all gradient against navy, in the mark, on a rack, in context.
The argument dissolved of its own accord side by side, because it turned out there weren’t four options. They were two of them. The mints were nearly identical to one another, and the brights were nearly identical to one another. The real question wasn’t about which team, soft or loud, and everybody could see that in about nine seconds. Loud was winning and that green that came through the middle was ours. Both fans took their color seriously, saw it lose fairly, and said goodbye happy. Nobody had to tell them no.
The part that never was mended
What still frustrates me is this.
The green they picked works for the brand, but it’s the hardest color in the set technically. It is saturated to a degree that four-color process printing can’t always achieve, so the brochure may never quite match the powder-coated rack sitting next to it exactly. It’s so bright, it’s blown out against a white. We put all this in the guide, rather than keeping quiet about it.
But we never got to stand at a press. Never seen a draw-down come off the line at the coater. We wrote the specs, then we sat back and watched from a thousand miles away as other people implemented it. A giant leap of faith. But what saved it was the teaching.
We had spent a little over six weeks covering coated vs uncoated, why a spot color exists, and what to ask a manufacturer for. The client could go back to their vendors and push like we were right there next to them. They knew what to ask, and what the right answer was. They’ve done it. I’d be lying if I said that felt exactly right. It was like giving someone a recipe and not being allowed in the kitchen.
It would have been a cleaner process with two more weeks and a plane ticket. It might have looked like a mess, but it still worked.
The other box
There’s an epilogue no one told me about.
When the project finished, I had the original box, as well as everything sent in the six or seven rounds after. The client didn’t want any of it returned. Which is generous but left me sitting in the studio with a pile of neoprene, anodized aluminum, powder-coated steel, laminated vinyl, coated paperboard, and about fifty-five pantone paper and 30 pantone plastic chips.
Try to recycle that. The waste streams are all different and half are laminates and bonded materials that no municipal program can deal with. The koozie is made of neoprene, which is functionally forever. Vinyl Banner. The chips are coated paper, which once it’s had that much ink on it isn’t paper anymore, and the other chips are good old plastic.
I’ve had a ridiculous amount of time to do research on mixed-material recycling for a project about a company that’s entire value prop is reducing environmental load in data centers. I didn’t solve it gracefully. Some of it was recycled correctly. Some of it is still here. The bezels are on a shelf in my office because they’re beautiful and I can’t throw them away.
What was on the table at the tradeshows?
6 colors, Pantone first, and translated from there. Each with a rule about what it can be next to, and where. Two logo states instead of six. One typeface and one blue for the product name.
At the tradeshows, the backdrop matched the server rack, which matched the brochure, which matched the deck on the screen behind it. Which was the whole idea. Not that the colors are beautiful, though they are, but that somebody walking that floor sees one company instead of five.
A brand guide is not a color list. It’s a list of decisions someone has written down so no one has to make them again at 4 o’clock on a Friday with a vendor on hold.
We couldn’t afford to be precious with this one, so we were specific instead.
— Dave & Deb