The example described in this chapter is based on experiences working with the product development centre of a major international fashion brand. The talent pool is team of 4 to 6 students who work closely with one of the creative directors for a period of 6 months, with the objective of building a small visionary collection. To become member of the talent pool you have to apply and undergo a careful selection process.
The objective of the talent pool is to attract, position and retain gifted designers. Over the course of almost 2 years we managed to increase retention of talent pool members from 40% to 85%.
The initial challenge is to on-board and ramp up the talent pool and develop the members understanding of the brands. Obviously, they need to have a good feel for the brand they are going to work for (‘their’ brand), but they also need to understand how their brand relates to the full portfolio of products. To ensure effective collaboration with the different design teams and success of the talent pool project, it is fundamental that members of the talent pool understand the boundaries between the brands, and that they able to discuss the brands freely with the brand director. This requires developing a vocabulary specific to each brand.
On-boarding, i.e. understanding the different brands, is a perceptual challenge requiring the designer to develop the perception of the differences and similarities between the brands. In fact, perception is mostly about differences [see e.g. Gibson, 1979]. In this case the learning focuses on two specific differences: whether or not something belongs to the talent pool’s brand, and whether it belongs to one of the different yet closely related brands.
The on-boarding process was accelerated by having each member of the talent pool develop a moodboard. The talent pool was given two specific tasks with the purpose of training their perception of differences, First, they were asked to create 3 moodboards; one for ‘their’ brand, and two additional for the brands most closely linked to ‘their’ brand. Second, they were asked to apply a specific order to the images and items collected for the moodboard, and ‘not just ‘place’ the images anywhere. Instead, they were asked to a) make sure all of the images or parts of images on the moodboard belong to the brand depicted, and b) arrange the images from ‘fitting well’ to ‘fitting perfect’ starting from the lower left to the upper right.
This assignment had a particular effect on the team dynamics. Where, with a normal moodboard, the discussion is about, first, whether an image belongs on the board (does it fit to the brand) and second, where should it best be placed on the board to support the overall layout and composition. This assignment motivated the students to debate every image to find out, compared to other images, whether it belonged ‘more’ or ‘less’ to the moodboard’s brand. The results were 3 moodboards on which the upper left area fit the brand well, and the lower left fit less well. See Fig. The talent pool was given two full weeks to work on their moodboards, receiving occasional feedback from designers from the various brands. At the end of the two weeks, they delivered 3 moodboards they all felt best described the assigned brands.
The first time we applied this approach we were exploring new territory; would it work? Were the students able to capture the different brands? Were they able to differentiate between what fit well and what fit best? In the same period, the company’s innovation-lab was preparing a presentation day to show results and materials collected over the course of the previous months. We took this opportunity to show the moodboards to a wider audience and organized a corner of the presentation room to perform a small experiment.
People entering the area were presented with a print-out showing the 3 moodboards, and posing two questions.:
- The first question was to indicate for each of the moodboards the most likely brand from amongst all of the brands. The expected result was that the moodboards were assigned to the corresponding brands.
- The second question was, for each of the moodboards and for the brand they indicated the moodboard represented, to circle in red the area least belonging to the brand, and to circle in green the area best belonging to the brand selected. Assuming the brand representations in the different areas of the moodboard applied generally, the expected result here was that the majority of green circles will be found in the upper right corner while the majority of red circles are found in the lower left corner.
In total 50 people participated in the test. Each participant received a A4 with a print of the moodboards. Below each moodboard there was an option to check one of the 3 brands. The participants also received a red and green pen, to indicate the area best fitting for each brand (a green circle) and the area least fitting (a red circle). There were no constraints for circle dimensions. At the end of the day all returned forms were scanned and both red and green circles were aggregated on a single page for evaluation (shown in Fig).
It was not a rigorously controlled experiment; more exploitative (‘fishing‘ as my father would call it) with the purpose of checking whether there is reason to believe opinion is unanimous. A proper follow-up experiment would show the moodboards in random order, under controlled conditions and with minimal distractions, allowing the subjects to form their opinion in isolation. As it was, at times groups of 2 or 3 subjects were evaluating the moodboards at the same time, clearly also discussing the moodboards as they went along. In spite of this, the results satisfied our curiosity and gave some confidence about the validity of the approach and its value for developing the talent pool.
Concerning the first question, labelling the moodboards, the answers were unanimous. All participants assigned the moodboards to the appropriate brand. This is not completely trivial. I often show this example during workshops and, after having introduced the brands involved, I ask the participants to assign the moodboards, and especially between the fashion and the casual moodboard there usually is some discussion.
Concerning the second question, indicating best and worst fitting elements, there appeared to little consensus (see Fig). A cluttering of green and red circles was expected in line with the assignment to structure the moodboard from best fitting elements in the upper right to the least fitting elements in the lower left area. Instead we found that both red and green circles are found all over the moodboards. During the day one could observe different behaviours when assigning best/worst fit areas. Some participants would indicate an area, while others singled out one specific element of the moodboard, but all a different one.
Reflecting on the talent pool activity and the results of the validation we made the following observations:
- Creating moodboards for the brand in the context of two related brands in combination with the task of structuring the moodboards according to best/worst fitting, demonstrated a fast-track towards obtaining a thorough understanding of the brand portfolio. The same effect was observed in two subsequent talent pools where we have taken the same approach.
- The moodboards may reflect only broadly the content they were designed to present, and this high level may be difficult to translate later into single elements.
Based on the experience and results of the verification, we adopted the approach into the standard on-boarding of talent pool. Experience with two subsequent teams showed similar results; accelerated learning about the brands and their differences, moodboards that expressed overall the individual brand although at a detailed level they can be disputed. It also has been the standard approach in the workshops, although in most cases the time for moodboards was limited and therefore the effect less clear compared to the observations with the talent pool.
See workshop handout chapter 4.