Welcome back to our monthly series on CMF data! In chapter 3, we learned about the different types of data sources. Now, let’s explore some common misuse when it comes to collecting data by conducting CMF surveys.

There is no shame in past mistakes. In the examples below, I have made some myself in my early days in CMF Design. We are all somewhere on the learning curve. What’s more important is to equip yourself with the knowledge to move forward along the learning curve and advance your CMF skill set.

In many color researches that have been conducted by trusted institutions, the format of online surveys are frequently chosen for the reason of speed and budget efficiency. However, we should always be reminded that CMF is beyond colors, and far more than what screens can represent digitally.

When it comes to CMF matters, online surveys should be avoided due to color inconsistencies across different monitors. On top of that, many CMF details are also lost in translation when presented digitally: tactility, light reflection, micro patterns, and etc.

And therefore, when possible, in-person surveys are always encouraged. Despite of their higher costs and efforts, this format will guarantee a more meaningful outcome.

When conducting surveys, always make sure all samples are equal in shape and size. All samples should look identical and differ only in CMF details to be tested.

For example, when gathering color preference feedback, present samples in consistent shape and size with varied colors. Avoid presenting samples with differing shapes, sizes, patterns, gloss levels, as these unrelated factors can skew participants’ color perceptions and influence how they feel about a color.

Ideally, represent samples in similar shape and size to the actual product. For instance, when surveying CMF preferences for mobile phones, present samples in rectangular shapes and approximate size of a phone. This will help participants to better visualize their preference for such product category.

Lastly, but not least, random sampling is important in research studies because it helps to minimize bias. Circling samples within personal connections and asking for their feedback might seem like an easy way to obtain “survey data,” but how much can we trust these results?

If I conducted a poll on my Instagram asking which CMF Designer people follow, a significant portion might choose me simply because they already follow and know me, making the result irrelevant.

Similarly, if I conducted a CMF poll around the office, my colleagues, who are of similar ages, work in the same industry, and have similar pay grades, do not truly represent the diverse group of consumers we aim to understand.

Random sampling in research studies is essential for minimizing bias by ensuring the sample size, i.e., the surveyed participants, represents the population accurately.

Therefore, the next time a colleague comes to you and shares “survey result” with you, kindly remind them that these are internal feedback and should not be considered as genuine survey results, as they do not accurately reflect the feedback of our target consumers. Of course, internal feedback is just as valuable, but they should not be mistaken for consumer surveys.

Now that we’ve seen common mistakes in collecting CMF data from consumer surveys, let’s endeavor to avoid these pitfalls in the future!