Select Page
Sentiment analysis in online reviews. Why does it matter for business?

Sentiment analysis in online reviews. Why does it matter for business?

Analyzing sentiment in the comments and opinions that swarm social media helps uncover what is essential for marketing departments and beyond. It helps track business performance based on customer and product satisfaction information and manage brand reputation online.

Sentiment. Do online comments matter?

The skill is in knowing how to take advantage of customer reviews and comments, which are abundant online. However, it is difficult to extract the ones that are relevant from all the online chatter. Every day there is a whole lot of conversation on social media about services, companies and products.

Social media is a very important space in business development today. It is one of the most cost-effective ways to conduct market research by analyzing customer opinions, and on top of that, it is time-efficient. It allows you to engage with customers – both current and potential – in a valuable way.

This is essential in the process of branding and resolving any negative or sensitive issues as quickly as possible. When we analyze sentiment in social media comments, we get express insights into various aspects of the business that ultimately affect profits.

These include the pros and cons of a product or service, changing market preferences, after-sales service and many other parameters. Today, up to 70% of companies use insights and knowledge from social media reviews to formulate brand strategy, and more than 60% find sentiment analysis data in reviews extremely useful for effective ongoing customer service and sales.

Sentiment analysis is a stepping stone to customer experience analysis and is done through machine learning (ML) models. These include neural networks, natural language processing (NLP), thematic categorization and finally sentiment and emotion analysis.

What does the knowledge coming from the network allow?

Here are some of the most important ways for companies to use sentiment analysis on comments. These include extracting in-depth information about customers and products, discovering emerging market trends, making sales or increasing market share. These include:

  • Monitoring overall customer satisfaction;
  • Improving the customer experience;
  • Getting real-time consumer insights;
  • Identifying emotional triggers in customers;
  • Improving products and services;
  • Building brand loyalty;
  • Reducing customer churn;
  • Training customer service employees;
  • Training chatbots;
  • Managing brand reputation;
  • Monitoring changes in sentiment over time;
  • Improving marketing content published on social media;
  • Gaining competitive intelligence;
  • Preparing for changes in market trends.

Branding is the focal point of all marketing efforts when it comes to products and services, so studying its image is about understanding how it is perceived by those who use those products or services. It’s also a way to learn the opinions of those who value the competition more.

Branding is basically a psychological phenomena created and designed by marketing to influence consumer preferences and buying behavior. It is the face of a company and the outer shell that the world sees.

Sentiment analysis. Examine your image!

Sentiment analysis helps companies capture what their customers are really saying about them by analyzing the feedback consumers use to express their feelings and emotions in published reviews. Whether these sentiments are positive, neutral, or negative, in the digital age it is important to use this to align your business with market trends

Consumer opinions are essential to understanding a brand. Companies can learn from consumer opinions to improve customer experience and create or change operating strategies.

Sentiment analysis can help a company examine its brand by:

  • Discovering what motivates customers and stimulates purchase intentions, or fosters dissatisfaction;
  • assessing the impact of marketing campaigns and creative strategies on consumer perception;
  • understanding consumer opinions on product price, quality, reliability, user experience, etc.; and
  • benchmarking against competing products, services, and brands;
  • analyzing and comparing different consumer segments and share of voice.

The three main ways consumers express their feelings and opinions about a brand? They are product reviews, consumer surveys and social media posts.

The challenge for marketers is to collect, analyze and summarize these myriad opinions. This must be done in such a way that their results are understandable and actionable.

Using both topic-based sentiment analysis and machine learning and artificial intelligence, Sentimenti can help analyze customer sentiment and opinions.

It doesn’t matter if it’s qualitative or quantitative research. An important part of branding is understanding what’s behind the content of comments. By doing so, you can gain deeper insights into the minds of consumers or prospective customers, and that’s another step to improving your offerings.

Is positive emotion really always positive for a brand?

Is positive emotion really always positive for a brand?

Advertising, marketing, and communication activities based on influencing the positive emotions of the audience are used frequently today because they are effective. Why? Such emotions can be easily transferred directly to the brand, and thus build a positive image of the brand. Besides, influencing the emotions of the audience allows not only to better anchor the brand in their consciousness, but also to stimulate them to a certain action. But are positive emotions really always positive for a brand?

Positive emotions are generally those emotions that we experience with pleasure. To put it in scientific jargon, they are pleasant or desirable situational reactions, distinct from pleasant sensations and undifferentiated positive affect. Such emotions include love, joy, satisfaction, contentment, interest, amusement, happiness, peace or delight, among others.

How do they work? Experiencing positive emotions:

  • expands a person’s set of values and beliefs,
  • stimulates openness,
  • makes it easier to find solutions in a difficult situation,
  • makes it easier to get out of one’s comfort zone and pick up ideas or take actions that are not typical for a person.

This very mechanism is used by advertising, marketing and communication campaigns to strengthen the relationship of customers with the brand, encourage them to buy products and services, join newsletters, etc.

Positive emotions – two sides of the coin

However, what specialists do not want to remember is that every stick has two ends. Yes, yes: positive emotions also have their dark side. Take joy – an emotion with a clear positive connotation. As research shows [1], people who intensely pursue happiness have a greater ease of experiencing depressive states, feeling unhappy and being depressed. They also tend to be more selfish and feel lonely.

The study also indicated that anger may have an impact on performing certain actions more effectively, such as those of a confrontational nature (we wrote about the positive effects of negative emotions in a previous article). In contrast, emotions of a positive nature can sometimes have the opposite effect.

Positive and negative emotions are like yin and yang

They cannot exist separately, but complement each other. Sometimes people feel both positive and negative emotions simultaneously in certain situations (especially stressful ones). This happens, for example, when moving out of the family home or when graduating from college. In such moments, one may feel joy and sadness at the same time – emotions that, in the study, have opposite charges to each other.

People who have succeeded under ambiguous circumstances will have a similar emotional experience. Moreover, those who have suffered a failure, and this failure brought them a kind of deliverance from a difficult situation, or its consequences were not as terrible as it promised to be at first.

The above information shows that basing communication with customers solely on positive emotions is wrong. The study of the effects of such activities will be falsified because it is incomplete. To assess the emotionality of the recipients of marketing, advertising, PR and other activities, one needs not only a tool that cross-searches the emotions experienced by a person or group of people in a given situation. It is also necessary to take a more flexible approach to understanding the role of emotions themselves. It is best to assume that a given emotion can play several, sometimes conflicting roles.

Traps of positive emotions

The high effectiveness of marketing or communication activities based on positive emotions is very tempting for specialists. However, there is a considerable risk in this. Constant pumping of a positive emotional balloon causes frustration in the recipients of such communications after some time. This can lead to a loosening of the relationship with the brand, breaking that relationship, or even turning to the competition.

Imagine being in the company of an overly cheerful person for a long time, almost flooding with optimism or trying to make everyone around happy… by force. Tiring, isn’t it?

Receiving positively charged, but served in too large a dose of brand communication will work similarly. It will be perceived as intrusive, tiresome and inauthentic. Consumers will react skeptically to such actions.

Why? Because as emotional beings, we are only complete when positive emotions are joined by negative ones. When we accept their existence and integrate the whole with each other. This mechanism also translates into the reception of emotionally charged communication.

By focusing in communication exclusively on positive messages and not taking information about the negative emotions of its audience, the company commits a cognitive error (the so-called Pollyanna effect). Customers in the cognitive loop of positive emotions will contribute little analytical information about consumer behavior, and will not act. Instead, customers in negative emotions will provide this information. This is because their natural reflex will be to want to break out of the loop of negative emotions, so they will take appropriate steps to take so.

With incomplete data on negative reactions, the full potential of positive emotions cannot be realized

This reduces the value and effectiveness of the actions taken by the brand. A holistic analysis of emotions and sentiment will help in this case to identify potential changes in consumer behavior. And it is additionally important to remember that it is negative emotions that have a greater impact on changes in behavior and personality formation.

Negative emotions affect the way a company operates

The use of only positive emotions in marketing, advertising, and communications is often due to the social stigmatization of negative feelings, and therefore fear of them. This phenomenon can be observed especially where a social pattern of functioning is translated into the way an organization operates. For example, there is regular fearmongering and manipulation in public spaces, for example, by politicians, religious leaders, etc.

Under such circumstances, negative emotions are perceived as an enemy or at least as a weakness, and these, after all, should be protected against. Choosing only positive solutions is therefore here a socially expected form of protecting oneself, the organization or the brand. A way to manage the company and a method to build an image in the public space.

However, with the simultaneous denial of negative emotions, such a model of company operation or a way of brand communication becomes a powerful ballast in development and a barrier to achieving the goals. The vast majority of energy in this case is directed at building and maintaining an illusory positive image of reality. Human resources and financial outlays are directed to this goal.

If a crisis arises, such as an image crisis, it will be much more severe and long-lasting for the organization because its symptoms will be denied and repressed. It will be much more difficult for a given company to recover from such a crisis. It will need not only to take corrective steps against misguided actions, but also to reevaluate the entire way the organization functions.

Effects of building customer relationships on positive emotions only

Emotions are important components in creating the customer experience (CX) with a company. Placing communication solely on positive emotions and experiences deprives a company’s analytical cells of an significant range of information regarding specific consumer behavior. This leads to an incomplete depiction and therefore incomplete understanding of the needs of that company’s customers. As a result, these needs remain unmet, which eventually results in declines in sales, an overall reduction in brand appeal and a gradual exodus of dissatisfied customers toward competitors.

Consider customers’ experience of interacting with a brand in terms of the emotions they feel

These experiences result from the interactions between the various points of contact between consumers and the company at the pre-purchase, during purchase and post-purchase stages. The experiences from these interactions can sometimes be contradictory. A customer may experience positive emotions when reading enthusiastic product reviews or watching emotionally charged advertisements. However, he or she will also experience negative emotions when waiting for a long time for an ordered product, when demand increased by advertisements causes delays in delivery.

Such impressions in contact with the company can then:

  • reflect negatively on the image of the brand (it is inefficient, it does not cope with meeting the needs of customers),
  • project negative opinions about the product (when the frustration caused by waiting too long outweighs the level of satisfaction with the user, or when the product does not fully satisfy the user’s expectations, has defects, etc.).

In the case in question, betting only on creating positive communication misses the point. Again, the needs of the users of the product or service are overlooked, which leads to a progressive degradation of customer attachment to the brand or commodity.

Using only communication based on positive emotions and building customer attachment to them is, finally, fraught with yet another mistake. Without expected results from marketing, PR or advertising activities, there may be a tendency to assume that the absence of positive resonance (i.e., the absence of positive emotions in customers) means the presence of negative emotions. This is not necessarily true.

The actions taken by the company will then focus on covering negative emotions with positive ones. And this will be a mistake because building relationships on positive emotions is of a different nature than neutralizing negative emotions. In such a case, it is necessary to conduct an analysis of customer emotions in response to the actions taken, which will enable validation and corrective steps to be taken.

Conclusion: a single positive customer experience can be valuable to him and result in the transfer of positive emotions into loyalty decisions. It can therefore strengthen his relationship with the company. But this positive experience can also make the resulting secondary positive experiences less exciting and valuable to the same customer. This raises the possibility that these positive experiences will not positively influence loyalty decisions.

Positive emotions without negative ones? That’s not how it works

All activities aimed at promoting and advertising a product or service, increasing brand awareness among current and future customers, and creating lasting and engaging B2C relationships are dependent on emotional charge. Their effectiveness will be greater the emotional charge, however, using the full spectrum of emotions – positive, neutral, and negative.

Positive emotions cannot function isolated from negative ones – this is simply unnatural. Example: a blog article about a particular product, which only mentions its advantages. Such posts are not uncommon. Meanwhile, it is enough to have a glance at the reviews of this product on the web, and it is already clear that the mentioned text is unreliable, and describing only the advantages of the product – manipulative.

The same is true of more general communication directed to customers. Striking only a positive note falsifies the picture and does not provide full data on implemented measures. If there are problems, fixing such communication becomes much more difficult. Therefore, if we are already using emotions to increase the engagement of our users, let’s dose them carefully and with the whole palette in mind. We will then have the best effect.

Pfizer or Moderna? AstraZeneca or Johnson? Sentiment and emotion around vaccines

Pfizer or Moderna? AstraZeneca or Johnson? Sentiment and emotion around vaccines

The topic of vaccines, and specifically the differences between them, arouses great emotion. Analysis of Google search phrases shows that the public is still looking for information on the differences between preparations of different manufacturers. “Which vaccine to choose?”, “Moderna or Pfizer?” – are just some of them. This time we took to the workshop conversations of Internet users about five preparations. We explored the emotions and sentiment between them to see how they differ.

Pfizer or Moderna, or which vaccine to choose?

The material analyzed by our team represents a total of nearly 190,000 mentions of five vaccine manufacturers between January 1 and May 6, 2021. Content about specific manufacturers was filtered out of discussions about COVID-19 vaccines overall. Using the SentiTool tool, it only took a few minutes to get the full results.

Preparations were most often discussed by Internet users under articles on news sites. The second place in terms of discussion sites was Twitter, followed by Facebook. On news portals, Internet users discussed AstraZeneca most often. This is also where a lot of opinions about the Sputnik-V vaccine from Russia were reported online. On Twitter and Facebook, but also in other places, it was Pfizer. These two vaccines are among the most commented on.

pfizer moderna johnson astra zeneca jaka szczepionek wybrac opinie skutecznoscThe graph of discussion dynamics (monthly data) shows that the AstraZeneca controversy has had its effect. This vaccine was discussed very intensively – most often in March this year. Johnson&Johnson was the least discussed. It is worth noting that in a significant number of opinions there were more names of preparations.

COVID-19 vaccines – a comparison. Sentiment and emotion

When looking at the differences in sentiment of opinion of a given vaccine versus the others, two extremes are clearly visible. Johnson’s was the most positively commented on during the study period. In contrast, the most negative statements were reported around Sputnik-V. The controversy around vaccine reactions after AstraZeneca’s vaccine ultimately did not impair the online image. More unfavorable content was seen from Pfizer, which may come as a surprise.

DIFFERENCES IN OPINION SENTIMENT BETWEEN VACCINES DURING THE STUDY PERIOD

Differences in the intensity of negative and positive sentiment inopinions on COVID-19 formulations. The graph shows how sentiment differed for each vaccine from the other four

This preparation is one of the most popular among Poles and it is with this preparation that citizens most often want to be vaccinated. This may have been influenced by the co-occurrence of other preparation names together with Pfizer. This is because we are primarily examining the general climate of online discussion around the products of pharmaceutical companies. Thus the graph below shows how opinions on one preparation could have influenced the discussion climate around another preparation on the basis of co-occurrence. Moderna and AstraZeneca co-occurred most frequently with Pfizer.

Co-occurrence analysis of formulations in online reviews

The chart below breaks the positive and negative sentiment into eight base emotions for an even closer look at vaccine opinions. Sputnik-V differed from the other formulations with a high dose of disgust (about 13% more than the others) from internet users, while Pfizer differed with fear and sadness. AstraZeneca reported a slight increase in confidence and expectation. With the same emotions, single-dose Johnson differed from the rest most on the plus side (with strong decreases in negative emotions relative to competitors). As for Moderna, there is little difference from the other four formulations against COVID-19.

The above analysis is illustrative and informative. It does not reflect the full results of the study.

How to properly analyze emotions?

How to properly analyze emotions?

Each data analysis is aimed at understanding what information it contains. Has something changed, or is there a difference between A and B? Do changes in A correlate with those in C or D? Only these steps allow us to draw conclusions about the results.

First stage: measurement. What is the text?

The above statement also applies to the analysis of emotions or sentiment. Its first stage is the MEASUREMENT, checking how many and what kind of emotions we find in a given text or set of them. The result of a simple emotion measurement shows the intensity of each of Plutchik’s 8 basic emotions supplemented by positive and negative sentiment and arousal (overall emotional temperature of the text). Sometimes we can afford to interpret it already at this stage. We did it in one of our first entries, where we analyzed short ads (what is important, we have already managed to improve the way the results are presented). By analyzing the ads we wanted to show something characteristic for the whole type of texts: the most important emotions are joy and trust, only at the very beginning of the story about the product the creators allow themselves to remember the negative ones – to show the hardships of life before the era of the best shampoo or grease in the world.

The correct results of the Emotional Measurement are those that are consistent with people’s feelings, after all, each of us is an expert on feelings. Our tool owes its correctness to the participants of research on emotions in Polish, which we conducted according to the best scientific standards.

Measurement is only the first step towards understanding the message and the emotions it contains. When we deal with many similar texts or collections of texts, we have to do something else. We want to find out which shop has the best opinions? Which version of our marketing content expresses the most enthusiasm or best shows interest in the subject? Which of the texts in the “Beauty” section will delight, move or warn the reader? We are talking about COMPARISON.

The second stage: comparison. Does this text differ from the average?

Comparison is perhaps the most important stage in the analysis of emotions – thanks to it we not only find out what the text is like, but also how it compares with others. We can compare directly – as we did when writing about lipsticks and lipsticks. Then we were interested in which of the topics has an advantage in terms of positive emotions and whether this difference is statistically significant. However, comparing several or a dozen or so different cosmetic brands cannot be done in this way, it would not be the correct approach. That is why in the text about beauty companies we used a comparison to the average – we needed some kind of background measurement, so-called baseline. This approach will be useful, for example, when comparing shops and brands. We then answer the question which brand has better or worse results than most of the industry.

The most general type of baseline would be the sum of emotions that characterize not only the domain, portal or texts of a given author, but simply language. In linguistics, the so-called Polyanna effect is known, which is that there are more positive than negative expressions in every language. Not only in the dictionary, but also in what we say – this effect expresses quite a general tendency of our minds to spend time and energy rather on pleasant things. In our research we very often see this tendency – joys and trust are emotions that appear in the greatest intensity not only in advertising. The fact that language has its emotional mean all the more reason to draw conclusions only based on comparison and not on the measurement itself.

Third stage: trends. What do emotions do?

The analysis of emotions is also about tracking changes in time, i.e. monitoring emotions. We can check whether sadness or revulsion show a growing trend, i.e. there are more and more of them in statements on a given topic or in the opinions of customers. If we notice a trend, which is statistically significant, we can predict what will happen in the future and if by chance it does not mean an impending crisis (depending on the slope of the trend line).

At this stage it is also possible to go beyond the data from the Sentimenti tools. We started with something simple, accessible and yet untouched by others – we compared the emotional temperatures of mentions of listed companies with the prices of their shares, published publicly. Sentistock is great, it allows you to determine what the investor mood really is and how it translates into stock market fluctuations.

This part of the analysis of emotions depends entirely on who and for what purpose wanted to examine the overtones of the text, the notes, the conversation. We have also managed to show which emotions correlate positively with reactions on Facebook and Twitter – that is, how to write, so that the observers would like to like or comment on the post. However, we might as well ask how emotions correlate with remembering information from the text. Studies on the psychology of emotions, including those conducted by our colleagues from LOBI, indicate that the overtones of the text have an impact on what and how well we remember. Correlation between customer feedback and online store sales? Our tools are designed for this type of research.

Why so many stages?

Emotions were not created for themselves. This is our advisory mechanism: they tell us what action to take. Tversky and Kahneman did not receive the Nobel Prize for their research, but for showing that the consumer, including the stock market, is not rational. This statement tells us two things:

  1. emotions shape the market,
  2. we need good tools and methods to study this impact.

Trying to understand the emotions “on the eye” we won’t know more than the average customer wanting to buy a new computer, reading all the available reviews and then deciding on the brand for which he or she has (and had) the warmest feelings. Maintaining scientific standards, checking whether differences and trends are statistically significant and even better correlate with other, harder indicators is the best way to find out. After all, we live in the era of big data and data analysis.

What emotions make the social entry popular?

What emotions make the social entry popular?

Not only our analyses show from time to time that the more emotions we show, the more we receive from our interlocutors. What’s more, I think everyone knows the basic principle of advertising and marketing: emotions pay off because they generate reactions. On the wave of these beliefs, we decided to check whether actually popular Facebook and Twitter posts are more emotional than those that evoke less liking or commentary.

We used Facebook Insights and Tweet Activity analytics statistics. We focused only on those that tell us about the actual interest in a given post: the number of reactions (likes and comments). We ignored the number of page views and similar “big numbers”. – Emotions do not necessarily affect the algorithms of social media, but they should be the reaction of those who observe a particular channel. Where did we get the data from? Not from the Sentimenti website, we have too few fans so far. But we happen to be familiar with a certain pop-culture portal – the statistics are for posts from Nie Tylko Gry.

Emotions on Facebook

We have collected data from January 2019. The portal has more than 1700 viewers on Facebook, at that time it published about 3 posts a day, including multimedia (e.g. containing film posters), links to its pages, several links to films (mainly trailers). We did not analyze these types of posts separately, we would need much more material to do so. We focused only on the text. The posts evoked 20 likes or 23 reactions on average (3 comments on average). We filtered out the 20 most popular posts and the 20 with the least reactions.

In both types of posts we found less than 20% of emotional words, 15% in the most popular ones and 18% in the second group. This difference is not statistically significant, so the “amount” of emotions alone is not responsible for the popularity of the post.

If we look at the distribution of individual emotions, we can see that these two groups of posts differ most in the presence of sadness and trust. If we analyze 8 emotions together, the difference is not statistically significant, but already the set of trust, sadness and anger gives a result that shows the difference between the groups (Chi=13,945; p=0,002). Interestingly, in the most popular posts there are more negative emotions and less trust. Does writing about pop culture feed on criticism? Or maybe doubts about the quality of the promoted works?

Emotion on Twitter

We also analyzed the twitts of the portal, this time in December 2018 and January 2019. The portal has less than 200 observers on Twitter, published, as on Fecebook, about 3 posts a day: photos and links (mainly to its website). On average, they generated 4 reactions to the post (including: likes, comments and sharing). We filtered 60 of the most and least popular posts.

The percentage of words carrying emotion was slightly smaller for twitts, about 13%, and did not differ between the two groups of entries. Interestingly, longer twitts seem to be more popular than shorter ones. However, none of these differences are statistically significant.

Joy and trust seem to differentiate twitty most strongly. In fact, if we consider these two emotions and fear, we get statistically significant differences between the groups (Chi=8,569; p=0,014), while for 8 emotions the result is not significant and we can only talk about the tendency. In popular twitts trust is more often expressed, and less often – fear and joy.

Emotions in social media

As you can see, Twitter and Facebook posts from the Not Only Games portal we are analyzing show the opposite relationship between emotions and popularity, although on both platforms trust seems to be an important emotion for the audience. It is more important than the ratio of words that carry emotion to neutral. In the case of social media, therefore, the portal cannot simply talk about the influence of emotions on the popularity of an entry, no differences remain relevant if we compare all popular and less popular posts. We have to treat both portals as separate collections of texts to say something about evoking readers’ reactions.

What does this difference mean? Perhaps the author-reader relationship on Twitter and Facebook is different, but the administrator may also be key here – and for “Nie Tylko Gry”, someone completely different deals with each of these channels of communication with the recipients.

In this analysis we have shown that emotions, especially trust, influence the popularity of a post on a social network. At the same time, there is no uniform, valid for all media – on Twitter, positive emotions “won”, and on Facebook, negative emotions. It is also not enough to simply show any feelings – the proportion of emotional to neutral words was not important.

Getting to know your audience is crucial for promotion in social media. In order to encourage interaction with our post, we should find out what our readers, potential clients and supporters actually react to. Do they prefer a photo or text? Probably the first. But do they prefer joy, anger or trust? That’s what we won’t find out from the Facebook algorithms yet. As we have shown, these preferences depend on the communication channel and without reliable data analysis we will not be able to evaluate them.