The intelligent Way To Make Decisions
How Accedo uses data to drive design decisions
The life of a designer is one where we must debate and distinguish between subjective concepts. A challenge for us is to then figure out how to make the subjective, objective. When faced with two seemingly good designs for a product, and a split group of opinionated designers, we had to find a way to come to a decision. Here at Accedo we are striving to make data driven decisions, so we conducted some user research.
The purpose of this study was to compare two different design colour schemes. Both designs were exactly the same except for the colours chosen.
From here on, the design to the left, shall be known as indigo/orange and the design to the right shall be known as purple/teal. Note that the above images are not what we used for the test, but just representations of the design.
We realised for most people, comparing two designs ultimately comes down to a gut feeling, thus we included a general preference question. However, we wanted to know more than this gut feeling, we wanted to know why they preferred it.
Since this design was to represent our product and company, and ultimately, ourselves, we wanted to see if the designs projected the feelings we hoped they did. Thus, we specifically asked our participants which design was more “energetic” and which was more “trustworthy”, two keywords that we hoped the product would embody. We also presented a list of keywords (both positive and negative) for participants to use to describe the designs, as well as an open field for them to input their opinions. Our hypothesis was that indigo/orange would be seen as more energetic, and that to contrast this, the purple/teal would be seen as more trustworthy.
We administered a nine question survey that included demographics questions (age, gender, occupational role, country of origin), and design rating questions (multiple choice and free text). We stressed to the participants that the designs were exactly the same, and the focus should be on analysing the colour schemes. Design questions included rating which design they thought was more energetic, trustworthy, and which one they preferred overall, in addition to clicking which keywords they thought were associated with each design. Participants had the option to describe the designs using plain text. In an attempt not to bias participants towards certain designs and keywords, the presentation order of both the designs and keywords were randomized, with the absolute general preference question always being presented last.
We had 98 respondents complete the survey. When we look at the data aggregated together, we get the following three figures.
As predicted, we see that 78.6% of participants found the indigo/orange design more energetic looking.
Surprisingly, the purple/teal was not seen as more trustworthy. Instead, 45.9% of participants said they thought the indigo/orange was more trustworthy, and the remaining were split between the purple/teal and there being no difference between the designs.
Overall, we were surprised to see that such a strong majority (70.4% of participants!) preferred the indigo/orange design. While we were primarily focused on the feelings that are projected from the two different designs, and the attitudes users had towards each one, such a strong gut feeling cannot be ignored.
The following figure shows a comparison of which keywords participants chose from a list to describe each design. The majority of the keywords did not have a statistically significant difference between designs, however, indigo/orange was called “vibrant” and “modern” significantly more often, whereas purple/teal was called “dull” and “harsh” significantly more often. Ultimately keyword choice helped guide our decision as the indigo/orange had predominantly positive words associated with it, while the purple/teal had more unfavourable words.
User Generated Keywords
Describing feelings about colours can be a hard task, even for the designers. To get a further understanding of participants’ emotions to the two options, we gave participants the option of inputting optional keywords.
The following figure shows the words participants generated to describe both designs.
The user generated keywords generated some interesting results (“Windows 95” and “Easter” being amongst our personal favourites). We put special weight on the results from the user generated keywords because they represent what the user is really thinking, unguided by our designer hands. While both designs were described as “clean”, indigo/orange had the word used many more times. Furthermore, the purple/teal had as many negative words associated with it as it did positive words.
We know from past experience that when we design for different markets, there are different needs and wants. For example, when designing for South Asia, warm, brights colours are favoured, which may be less desirable in other regions. In line with this reasoning, we decided to analyse whether there are differences in colour preferences depending on demographic differences. We first looked to see if region of origin mattered in colour design preferences. We found that those from Asia and Europe fit the overall trend with 90% and 75% respectively preferring indigo/orange, but for those from North America, there is no statistical difference in preference between the two designs. Next, we looked to see if gender mattered in colour design preferences, and we found that regardless of gender, participants preferred the indigo/orange design. Finally, we looked into whether age mattered in preferences. All age demographics preferred the indigo/orange design, except those between the ages of 45-54, who preferred the purple/teal.
When creating products it is essential to keep the user in mind and to design for the people who will be using your products. When the user could be from any age group, and from any region, it makes it especially hard to design. There will never be a perfect design nor will there be one that pleases everyone. We could have spent time, effort and energy debating over an unwinnable argument, but instead we let the data decide for us. We had a distinct winner, and thus, were able to save ourselves from arguments, and focus on designing premium products.
By Ranya Amirthamanoharan and José Somolinos