ING comes in second place in terms of the number of messages. The disruptions that were the result of a number of DDoS attacks have caused ING to rank this high. After ING, Albert Heijn, Samsung and KPN follow. These top 5 brands are each discussed more than 1000 times per day on social media.
Trend in number of messages top 100 brands (top) and number of messages top 20 brands (bottom)
Progression of number of posts from the top 100 brands (top) and number of posts from the top 20 brands (bottom).
Campina is discussed most positively
Based on an advanced language algorithm (from Teezir), each message was determined to be positive, negative or neutral. With this data, a sentiment score was calculated for each brand. The sentiment score is the percentage of positive messages (compared to all messages about a brand) minus the percentage of negative messages. A negative balance therefore means that there are more negative messages than positive messages. See the image below for an overview of the top 5 positive and top 5 negative brands.
Most positive and negative brands, expressed in sentiment scores (% positive posts - % negative posts)
The most positive and negative brands, expressed in sentiment scores (% positive posts – % negative posts).
Due to the many positive reactions to the new lactose-free milk from Campina, this brand is discussed most positively on social media. No less than 30% of the messages about Campina were positive and only 6% of the messages were negative. Garnier, L'Oréal Paris, Nespresso and Andrélon are respectively in second to fifth place.
UPC is discussed most negatively
UPC comes out as the most negatively discussed brand on social media this year. The brand is discussed more negatively than competitors such as Telfort, Vodafone and T-Mobile. These latter telecom companies also have a negative sentiment. UPC ends up in last place mainly because of a peak in the number of negative messages around the end of February. There was an e-mail outage, which was complained about a lot on social media.
ING was also negatively discussed during the research period. The outage at the beginning of April not only resulted in a lot of messages about ING, it also resulted in a predominantly negative sentiment. Analysis showed that sentiment dropped significantly during the outage (from an average of -0.6 to -24). However, sentiment had already returned to its old level two weeks after the outage. Sentiment therefore experienced a significant spike due to the outage, but that did not lead to a more negative sentiment in the longer term.
Industry Analysis: Internet and Telecom Sector Most Negatively Discussed
The internet and telecom sector is the only sector with a negative sentiment, see the image below. This can be explained by the fact that these brands are mainly discussed at the time of a malfunction. The automotive sector comes out as the most positive sector.
The online sentiment must therefore be seen in the light of a total industry. A negative sentiment does not necessarily have to be negative and a positive sentiment does not necessarily have to be positive. For example, KPN has a negative sentiment, but compared to competitors it appears that all brands in the industry are predominantly discussed negatively. As a brand it is therefore useful to mirror the sentiment to competitors and on that basis to specify goals and draw conclusions.
Online sentiment by industry, expressed in sentiment scores (% positive posts - % negative posts)
Online sentiment by industry, expressed in sentiment scores (% positive posts – % negative posts).
What now?
This research can provide a number of insights that can be essential for a (online and offline) brand strategy. Two data from the research are central to this. The first is that the number of messages per brand is not evenly distributed; a number of brands are talked about a lot, while the majority of messages are talked about much less. The second point is that the average sentiment should be given a value judgment mainly in comparison with other competitors.
Roughly three types of brands can be identified from the research:
Brands with few messages;
Brands with lots of messages and negative sentiment and;
Brands with lots of messages and a predominantly positive sentiment.
Based on these three types of brands, we describe a number of matching opportunities and insights.
1. Brands with few messages
The majority of brands are discussed very little.
Firstly, this uneven distribution is increasingly recognised on the internet at all turkey mobile phone number list levels. The internet is a ' scale-free network ', where the growth of the network has led to a situation where choices are made based on previously made and other people's choices. This creates a situation where much-discussed brands are only talked about more. For a less-often-discussed brand, it is therefore very difficult to break through these power relations.
The chance that people will start talking about a brand is small, unless it is already one of the most talked about brands. In order to create more word-of-mouth reach on social media, one should therefore not simply assume that the social network will go viral. Even if a lot of media money is invested, it is no guarantee that people will talk about your brand. Developing relevant action mechanisms and ensuring that controversial topics around the brand are of great importance.
Another possible strategy is to target the people who are already talking about a brand. Even if a brand is discussed less often, there are always a few people who are talking a lot about it. These 'super promoters' can be activated, so that a more positive sentiment is created around a brand. Moreover, this can also lead to more reach and more word-of-mouth in the long run.
2. Brands with lots of posts, but negative sentiment
Brands with many messages and negative sentiment are currently mainly internet and telecom companies. ING also had negative sentiment during the period of this research. An important lesson for much-discussed brands with negative sentiment is that this sentiment must be given meaning in comparison to competitors. Negative sentiment is not necessarily bad.