Last week, I used LinkedIn for another experiment, seeking the wisdom of the crowd. And as it can happen, it went quite differently from what I intended, even though the setup seemed quite simple to me: can I trust the products in the picture or not? BUT: it did have some very interesting results…
CONTENTS: A new Oreo product? | It was NOT about print quality | But it WAS about color… and nobody noticed or cared enough to comment… | BONUS: a conversation about ‘color’ | Why is this important?
Originally published: 28/02/2026 – Last update: 28/02/2026
If you are a regular reader of my blog, you will have noticed that I’ve discussed the Oreo logo and brand colors already a few times. For a very specific reason: they changed their logo about 1,5 years ago. And it’s not only the ‘looks’ of the logo, from a more 3-dimensional to a flat design, but also the brand colors changed: a >10 dE00 change. The blue is now much darker.
Since I noticed that change, I’ve paid attention to the Oreo shelves to check out which logo was used. And even today – some 1,5 years after the change – I’m still seeing packages with old, and therefore: wrong, logos and brand colors.
But last week, when shopping in a well known retail shop, I also noticed a new – well, at least for me – product line. But even though it seems like a new product line, it features the old logo. When I got back home, I checked the local Oreo website (Belgium/Flemish), and it does NOT show that product! Which seemed an interesting experiment to me: ask the wisdom of the crowd whether I could trust a product that is not featured on the local Oreo website, AND which shows a ‘wrong’ logo and ‘wrong’ brand color (i.e., the old one).
So, I took a picture and worked on a LinkedIn post. This always takes multiple iterations to make sure I do not reveal what my intention is, and to make sure I’m not pushing for a certain conclusion, to avoid ‘framing’. That is a very important concept from behavioral economics, which should always be taken into account when doing surveys, experiments, and tests.
The text I eventually published:
“𝗪𝗵𝗮𝘁 𝗱𝗼 𝘆𝗼𝘂 𝘁𝗵𝗶𝗻𝗸? Can I trust this Oreo?
I noticed it just last week, but can’t remember seeing this before.
And this product line is not on the local Oreo website…
Maybe it is parallel import from another EU country?
BTW: these were not in the cookies aisle, these were in another spot in the supermarket.”
As you can see, I did not point to the logo or the brand color. If I had, the results would have been completely different. Just check out this research, where participants were asked if they noticed a color difference between two samples. And guess what: about one out of three print professionals noticed a color difference between two identical samples! Why? Because the question was if they noticed a color difference, and being color professionals, they had to see a difference… That’s why I made several iterations of the text before posting it, to make sure I wasn’t framing anyone.
But it seems I have a certain reputation, and everyone assumes that when I post something, it MUST be about print quality… Which was not the case. The fact that people seem to link my posts automatically with print quality, is related to another concept from behavioral economics: ‘priming’.
Both framing and priming can – or will – influence test results, and both often happen. E.g. if you would like to set tolerances for brand colors and you show different variations and ask participants in the test whether the shown deviation is acceptable or not, you are already clearly ‘framing’ people to look very carefully at those colors and make a deliberate choice. Which is quite different from how we look at colors during shopping: in the test ‘System 2’ (the slow but accurate part of our brain) will be activated, while during shopping our brain only uses System 1 (the fast but inaccurate one). During shopping we also only need color as a rough indicator to find a product, not an exact match. Next to that, people from the printing industry are also primed to look more critically at color, that’s part of their job. E.g. when a print buyer enters a printing company: he/she is primed to look extremely critical at the reproduction of brand colors, System 2 will ba activated. Regular consumers have a different attitude towards color differences, check out the results of Q8 in this test.
Here’s a brief article about framing and priming. If you want to know more about ‘behavioral economics’, I can highly recomment ‘Thinking, Fast and Slow‘ by the late Daniel Kahneman. He has shown that humans are absolutely not rational beigns.
It was NOT about print quality
Already the first comment I got was about how bad the print quality was and that the converter would give the brand owner a discount on the next job… I even wondered why that person came to this conclusion, I didn’t really see a (significant) print quality issue.
But one comment later, it seemed people thought it was just one product, with one of the four boxes in a completely different color. And from that point of view, those comments made sense. However, these were two flavors of the same product… So I added that information to my post, two times to be sure that everybody read it, but even then, several other people commented in a way that I’m not sure they got it that these were two flavors, two versions… [sigh]
This is the final version of my LinkedIn post.
But it WAS about color… and nobody noticed or cared enough to comment…
As I mentioned in the introduction, my intention was to see whether people would notice it was the old logo and old brand color. But it seemed that no one noticed… Or cared enough about it to leave a comment…
And that is an important one. Why? Because this is very relevant when it comes to print quality: we are constantly pushed to believe that consumers will not trust a brand when the packages are a few dE00 off. So that was the intention of the test, therefore the wording in my post: “What do you think? Can I trust this Oreo?”
Unfortunately, not that many people reacted to the post the way I had intended: only 10 people, 8 said I could trust it because they know that product, 2 said I couldn’t trust it.
I specifically asked to mention why I couldn’t trust it if people picked that last option. The first one didn’t give a real answer, only that he thought it was a knock-off. For the record: the picture was taken in one of the shops of one of the largest retailers in the BeNeLux (Belgium / The Netherlands / Luxemburg), not in a ‘shady’ shop. The second mentioned a print quality problem, and given the full text of his comment, this person might have thought that it was just one flavor, not two.
Looking at the full statistics for this post, over 1000 individuals saw it, just over half of them from printing or packaging. And, as I mentioned before, nobody noticed or cared enough to comment about the fact that the old Oreo logo, with a >10 dE00 color difference from the real one, was used. Even though we are constantly told that consumers won’t trust products with this kind of deviation, it would negatively impact sales.
As I mentioned before, there is absolutely no evidence for the claim that small color differences will negatively impact sales. All studies on color and packaging, on the need to have ‘the right’ color, are about different color categories, not about a 2, 3, or even 5 dE00 deviation. If you disagree with my statement, please do show me a study that shows I’m wrong! And also the claim that ‘color enhances brand color recognition by 80%’, that’s about the use of color versus no color (black/white) in newspaper advertising. That study is NOT about tiny color differences.
BONUS: a conversation about ‘color’
And then there was this comment: “In colour we trust, if it’s not the brand colour it’s suspect“, clearly, from someone who works with color daily.
But he didn’t answer my original question: Can I trust the products that I showed in the picture? So, I addressed this, and an interesting conversation unfolded, which clearly shows that people can have different views on ‘color’. For experts, this will be a very detailed description, including lighting and measurement modes and even ‘Spectral Absorption–Reflection Balance Point‘. However, for a general human being who is not an expert, ‘color’ will only be a general category, out of more or less a dozen (e.g., ‘red’), with a few variations (light/dark, vibrant/dull). (more on that) And our color memory is very poor, as I’ve shown multiple times, e.g., the studies I mentioned in this article. Even my color memory, as this embarrissing anecdote shows…
But also: sometimes a color is the product (a red Ferrari), sometimes it is not: the Coca-Cola drink is the product, the can is just a container to be able to transport it, to protect it, and to give enough visual clues in a shop to attract the attention of buyers.
Eventually, he never answered the question whether I could trust the Oreo packages I showed, which was the goal of my post… Why not? If the brand color is that important, it should have been easy to answer whether I could trust it, I guess? His answers are clearly from the perspective of someone who is into high end color, as a color expert, which can be quite different from the view of a regular consumer, especially when it comes to packaging of a (cheap) FMCG product.
Why is this important?
This little experiment shows a few interesting things. The first is that people in printing always immediately focus on print quality. Even when this was not the intention, even when I did not mention print quality or color, the comments immediately focused on those aspects. I just told the story that I hadn’t seen this one before, it isn’t on the Oreo website, so I asked if I could trust it.
The second, and most important, is that NOBODY noticed or cared enough to comment on the use of an old logo with a significantly different color: over 10 dE00 from the current and therefore ‘real’ brand color… So how can you reconcile this lack of comments with the urban legend that a tiny color difference would negatively impact sales because people would not trust this slightly different color?
Oh, another interesting one to conclude this article: check out the screenshot below, from the local Oreo website at the time of writing. Even on their website, Oreo is showing packages with the old logo… (in case you don’t see it immediately: the Oreo Crunchies have the old logo). And also take a look at the second screenshot: there is a 2 dE00 color difference between the logo on top of the page and the larger logo a bit lower… If this is tolerated by brand owners, please stop hurting printers by demanding tiny tolerances and rejecting jobs, or demanding discounts, when these are just outside those tiny tolerances. Print buyers shouldn’t be setting the similar quality requirements for a red can of Coca-Cola as for a red Ferrari. These are not in the same league.
I am sure I was the one you are referring to that commented that I would not trust it.
I think in general people – normal consumers, – not colour nerds, are sensitive to shifts and changes in appearance – colour being one of the more important ones. If they see shifts within a few packages on the same shelf, I suspect that will discourage them to buy the product, even if they know it and have tried it and liked it before.
So once again I advocate for as much consistency in appearance and colour as possible. I do not advocate for for instance changing the substrate – even if you can safe a few euro cents on the paper – and I recommend that brands work with print shops that print to ISO 12647 standards AND are capable of committing to a dE00 of no more than 3 when printing the packaging – not just once per year during audit, but always and every time they print the job. This is just a simple safety precaution and I don’t think it is an unreasonable demand with all the technology printers have today – icc colour management being at the heart of it when it comes to process colour printing. There is simply no excuse for professional printers that are getting paid for their services to print above a dE00 of 3 – but again, the customers have to also make sure that the same substrate is used every time. If they ask the printer to use a cheaper or different substrate/paper, they also have to accept a higher dE00.
Consistency is the name of the game, so try not to change ANYTHING if you are a brand owner concerned about your brand colours. That is the simple advice I have, – the same one I have given for the past 25 years.
So, I just showed you an experiment with over 1000 people, over half of them (500+) from the printing and packaging industry, meaning people who deal with color every day, in which NOT ONE noticed or cared enough to comment on the fact that the packages had a very wrong logo and brand color (> 10 dE00), and you want me and all of us to believe that your ‘suspect’ that consumers, who have zero experience with critical color evaluation, 1) would notice a tiny difference in a 3 dimensional setting where the lighting will have a big impact on color perception, and 2) they will not buy that product when there is a slight color difference, even though they need it… And you really want me and the readers of my blog to believe that? With zero evidence or any study to support this claim? While I just have proven that people from the printing and packaging industry didn’t notice a 10+ dE00 difference, or cared enought to comment on it? That they didn’t notice it was a medium blue instead of a dark blue? I think your argument is a bit flawed.
BTW: check out these packages of Kellogg’s:
This was taken in a shop that I used to frequent every other day, where I used to shop when driving home from work. See those different reds? From magenta to vibrant red? Well, after a few days, all of these were sold. Why? Because only people in the print bubble make a fuzz about this. Regular consumers don’t even see this, when their ‘System 1’ is activated. And when they need a certain product, they will buy it, otherwise they might get hungry.
And even I don’t notice a 10 dE00 difference when I’m just relying on my System 1 when shopping, that’s why I bought the wrong breakfast cereals some time ago.
I am actually primarily interested in the opinion of people who have had no special training or experience in judging colour. Any fluctuations and differences in product appearance are bad for any brand just like typos in text are not flattering. Normal average people don’t need special training to spot if something is off with what they see. They may not even realize what it is but our brain is an amazing instrument when it comes to picking up imperfections so my advice to brands is to stay as consistent as they can – including keeping their brand colours as correct as possible wherever they are presented.
I don’t have to proof this. This is just my own opinion based on that I prefer perfection to imperfection, I prefer a dE00 of 3 to a dE00 of 4 or 5 if I have the choice.
Perhaps I am the only one who feels this way and that is ok too.
I’ve already shared several studies – scientific studies, not just my experiments – showing that ordinary people are NOT good at recognizing color differences,nto even when their slow but accurate System 2 is activated, which is NOT the case during shopping, that’s System 1 (the fast but inaccurate one). E.g. this one and this one. And check the little experiment I did with my then girlfriend, when I showed her two packages with a different color. She only noticed the colors were different when I told her to look at the colors.
Being experienced in color has a significant effect on the ability to see minor color differences, as shown in this scientific study. Plus: that test was done consciously, so when System 2 was activated, which is quite different from when System 1 (the fast but inaccurate one) is activated, like during shopping.
You don’t have proof, because the studies that are out there show the opposite of your claims.