Beyond the P1/P2 light booth:
What color actually looks like on the shelf

If you work in printing, or if you are a brand owner, print buyer, or designer who cares about the quality of print products, you probably know very well what the press control station or a viewing booth looks like. You spend a lot of time there. But, here’s a genuine question: have you already looked closely at shops, where consumers decide to buy a product, or not? Under which conditions – including time constraints – are consumers making these buying decisions? And did you watch that space with the same rigor as when doing a press check? If you ask me, there’s a huge disconnect between the two.

CONTENTS: ISO 3664 | An expedition in uncharted territory… | Why is this important?
Originally published: 07/06/2026 – Last update: 07/06/2026

As a reminder, there is an ISO standard for visual evaluation of print: ISO 3664, which was updated last year. The standard distinguishes between two types of evaluation: a critical comparison (P1) and a practical appraisal of print (P2). The difference between the two: the amount of light. For P1, this is 2000 lux, for P2, it’s 500 lux (both with a certain margin). For the rest, it is in general the same: D50, high Color Rendering Index (CRI), and gray surroundings. Added 07/06/2026: As Paul Sherfield correctly pointed out in a comment: the 2025 revision also introduced two new viewing conditions: P3 and P4. These are identical to resp. P1 and P2, but cutting the UV part.

In case you’re wondering why there’s more intense light for P1: with more light (up to a certain point, I guess), it’s easier to spot color differences. Color differences you might not see with 500 lux. Please, keep this in mind.

An expedition in uncharted territory…
However, this ‘sterile’ color evaluation condition is quite different from how print is ‘consumed’ in real life. I’ve already discussed that the real world is not flat (unlike the sample you evaluate in a light booth), but there are many more differences that definitely influence how consumers perceive the color of packaging. And how they perceive color differences in the reproduction of a brand color, or not, because of those influences.

Let me take you on an expedition in uncharted territory, where consumers will make their final purchasing decisions!
Over the past week, I documented several shops, where I do my grocery shopping (depending on what I need and the timing). In every shop, I took a picture of the Coca-Cola aisle, plus a measurement with a Lux Meter app… For the record: I had my smartphone about 1m20 above the floor when taking the Lux measurements.

The first is the supermarket near my real job (Delhaize Museumstraat), where I sometimes buy lunch. And it’s immediately interesting: it looks like chaos!

Let’s take a closer look at one part of the image.

This one, with the middle and top shelves, has three different types of packages: regular Coca-Cola on the top shelf, and two versions of Coca-Cola mini on the middle shelf, cans on the left, bottles on the right. These two versions have different substrates and different color perceptions. Do you think anyone but me noticed that the three packages in this picture are a slightly different shade of red?

Also, especially with the folding carton, you can clearly see the effect of shadows: the top one is in the shadows, the bottom one is not. The top one has a darker red due to the shadow than the bottom one.

Also interesting, the bottles on the top shelf: the shrink wrap is highly reflective. Even if the foil were exactly the same, you would still see a gradient due to the curvature. Plus, that foil isn’t 100% opaque; you can clearly see where the label is on the bottle: it’s a different red from the area without a label (which’s darker).

Let’s take a look at another one from another chain: the Jumbo near my home, which looks similar.

The third one I went to is a larger Delhaize shop (Bosuil), with a very large selection of Coca-Cola in even more varieties of packages.

When looking closer at a part of that picture, not only are shadow effects seen, but also the effect of a rotated box and the effect of curved surfaces, which create gradients.

The fourth, a shop that I don’t frequent that often: Colruyt (Deurne Noord). Now this is a special one: Colruyt has been focusing on low prices for a very long time by lowering its operating costs. They were the very first to use a computer-based cash register with punch cards… already in the 1970’s… Their shops look like warehouses; full pallets of products are placed on the racks. As a shopper, you have to take the products you need from that pallet or out of tertiary packaging yourself, which saves Colruyt a lot of hours. And: the amount of light is much lower than in other shops (230 lux), to save energy. I also measured the amount of light in those shadow regions: 2 lux…

Sometimes Colruyt also repackages products; e.g., in the picture above, they combined 2 shrink-wrapped packages of Coca-Cola with some extra foil. Which, of course, also influences the color perception. This one also clearly shows the shadow effect between packages, a factor limiting the ability to detect minor color differences, as I showed in my previous article. Also, you can clearly see how the curvature of cans creates gradients.

The final one is an Albert Heijn shop (Silsburg), and here I did an extra check: the light level at the height of the different shelves…

A last one to conclude this series of real-life pictures: check the color of that 12-pack in the image above. This is not the same color: there is a difference between the regions where there is a can below the shrink foil and those where there is a gap between the cans.

And just for fun, just out of curiosity: the amount of light in my fridge: 100 lux. 

Why is this important?
The real world, where consumers make buying decisions, looks nothing like the sterile P1/P2 light booths used for print quality assessment. The P1 condition was created with a very specific goal in mind: to detect even the smallest color differences. The real world isn’t close to that amount of light, it’s more like P2.

But P1/P2 also avoid all kinds of other influences, e.g., the surroundings must be neutral gray. Which is certainly not the case in a shop: these are a combination of different products, with different materials. And the real world is not flat: take a close look at the red of a can of Coca-Cola; it will be a gradient due to the curvature of the can. If you do see only one color, that’s the trick of your mind, which knows what Coca-Cola red looks like and will always ‘see’ a Coca-Cola can or bottle as that same red… not the one that my camera registered. Think about that.

And then there also the time aspect: you spend a lot of time checking color reproduction in a P1/P2 environment. That’s not the case with consumers: studies with eye tracking in a real world setting have shown that consumers usually look less than 2 seconds at a package. And well known, distinctive brands? Coca-Cola got a 0,1 second ‘total fixation duration’…

The disconnect between the P1/P2 light booth evaluation and the real world is costing printers money. Some brand owners and print buyers are demanding tolerances so tight that they would be extremely difficult to detect in a real-life setting. This shouldn’t happen: there is no evidence that small color differences influence buying decisions; consumers are very loyal to their favorite brands.

For the record: I’m not promoting bad print quality. And quality management tools are useful to keep production under control and reach a higher efficiency (e.g. less setup waste). It’s the never ending race to even tighter color requirements that I’m fighting against. That’s a nasty power game and more about ego than about ensuring that a consumer will recognize a product and decide to buy it.  

Related posts:

My color memory IRL (it’s embarrassing…) Brand colors IRL: the Coca-Cola case Brand colors in real life: the Milka case