From bright blue to cool white, to sunny yellow, to pink or black, users frequently search for products in all colours. The theory of colours states that every colour has its own particular qualities and emotional effects: each colour evokes certain feelings, affects our moods, and may even influence our decision-making. Every sixth search query for dresses includes specifications on colour*. In this article, I will explain to you how efficiently adjusted search queries involving colours impact performance.
On average, search queries involving colours generate a 19% lower cost-turnover ratio
We generally like to show our colours and we frequently search for them online. There’s no limit to our imagination when it comes to unusual descriptions such as “handbag wet asphalt” or, to keep it more simple, “handbag dark grey”. Colour-oriented searchers are guided by emotions, and users will most likely not appreciate ads and landing pages that don’t conform to their expectations. But how is this reflected in the performance? I investigated the following aspects in the context of an online retail company:
- How large is the share of search queries that focus on colours?
- Which performance levels are generated by search queries involving colours?
- Are there any differences between generic and brand-related search queries?
For my investigation, I looked at roughly 15,000 search queries with at least 50 impressions of both generic and brand-related search queries for dresses in the German market, and I gained the following insights:
- Overall, more than 15% of all search queries contain a colour-related search component.
- The performance of search queries involving colours generates a cost-turnover ratio that is on average 19% lower than that of other search queries.
- Generic search queries generate a cost-turnover ratio that is 22% lower. The average of 19% is caused by the slightly negative cost-turnover ratio of +1.5% for brand-related search queries. Users searching for dresses by specific brands therefore tend to focus more strongly on the brand than the colour.
White generates the highest conversion rate; yellow spawns the highest CTR and the lowest cost-turnover ratio
In view of the first eight dress colours generating the highest conversion rates, you will find a number of outliers. Surprisingly, white turned out to be the colour with the most conversions, while blue accounts for the lowest cost-turnover ratio. Although yellow is not among the strongest colours in terms of conversions, it’s the colour that generates the highest click-through rate and the lowest cost-turnover ratio.
Considering that 15% of all search queries contain specifications on colour, we should shift our focus all the more towards the interplay between search queries, ads, and landing pages. Colour-oriented, emotionally guided users will most likely ignore ads that don’t contain their desired colours; the same does, however, not apply to search results on Google Shopping.
We recommend that PPC managers use relevant keywords for search queries involving colours, sort them into carefully structured campaigns, and create ads which contain the required colour in the ad texts. By combining these steps with a strict “negative” strategy on the AdGroup level and on relevant landing pages, you will boost both the relevance and performance of your ads.
On top of that, you might want to take a look at the performance of each colour in order to make informed decisions on measures to optimise your campaign and bid management. For example, you could look into using Sitelinks. Only few advertisers actually use them to touch upon the emotions of potential buyers. For instance, if you place those colours that generate the most revenue in your Sitelinks, you will increase the chance of clicks. If you take the search query “dresses”, for example, it might look something like this:
little black dress – exciting red dress – sunny yellow dress
Those of you who put in that extra effort may, in fact, secure a whole bunch of positive figures in the end.
*Analysis includes 15,000 search queries that involved at least 50 impressions