Product optimization through customer feedback

The analysis of user-generated content requires a structured approach in order to identify and evaluate relevant information. Therefore, in the first step, the object of investigation was modeled by computational linguists using topic-specific keywords and exclusions in order to narrow down the posts from the social web through a search query.

The sources used for this analysis were Facebook, Instagram, Twitter, forums and YouTube. These platforms provide a valid database for gaining a detailed understanding of the topic.

The posts recorded from the keyword query were checked manually to confirm their relevance. Promotional posts, retweets and posts with no content or relevance to the topic were excluded. The remaining posts were carefully examined to extract the relevant statements and coded by topic, sub-topic and sentiment.

Both quantitative and qualitative methods were used to conduct a detailed analysis of the coded data. The quantitative analysis evaluates the posts statistically, for example to look at the development of communication over time or to draw conclusions about the relevance of individual topics from the consumer's perspective based on the volume of communication. As part of the qualitative analysis, the content was investigated in greater depth and a better understanding of the meanings and correlations was gained.

In this analysis, the results were examined and presented by brand. The topics and opinions that were collected in connection with each brand were evaluated and interpreted in order to gain a comprehensive understanding of the strengths and weaknesses of each brand as well as its image and reputation in online discourse. On the other hand, a comprehensive picture of customer feedback on the topic was gained, as both differences between the brands and cross-brand insights on the topic were generated.