Why Influencers Should Pay Attention to Word Choice

Why Some Social Media Influencers Have More Impact

Angie Basiouny: Marketing professor Jonah Berger is here to teach us about how to be a more impactful influencer. He has been doing some co-authored research that looks at how sensory language can help social media marketers drive engagement, connect with their customers, and hopefully drive sales. The paper, “How Sensory Language Shapes Influencer’s Impact,” and appears in the Journal of Consumer Research.

Jonah, welcome back. I was reading recently that social media spending on influencers has passed the $20 billion mark. That is a huge spend for brands. Tell me about this research and why you wanted to look into it.

Jonah Berger: Yes, so if we go back about a decade ago, I did some conversations with Knowledge at Wharton about that time around the power of word of mouth. My first book, Contagious, was all about word of mouth, how do we get it? How do we get people to talk and share? At the time, people were still used to sort of a traditional advertising mindset. We’d go on television, a spokesperson or an actor or a regular-looking person talks about a product or service, says how great it is. People were starting to realize in the early 2010s, “Wait. Word of mouth is much more impactful than traditional advertising.” We trust it much more because it feels like it comes from a peer rather than a company, and it can be much more targeted.

Given that revolution, and attention of word of mouth, companies and organizations have spent the last decade-and-a-half starting to think about, “Well, word of mouth is really valuable. How do we get it?” Some of that is encouraging your existing customers to talk and share, which is great. But companies used to that traditional advertising model said, “Wait a second. I’m used to paying, buying attention for my stuff. Is there a way to do that?” And that really started the influencer boom. It’s a new type of paid media. Rather than me designing an ad as a company and going on television or putting it in a magazine, now I pay an influencer, an online individual who has some sort of following, to talk about my product, my service, and share my message.

Now this $20 billion industry has grown up, where people are talking about products and services and using them to influence others. But importantly, there’s a catch, right? Influencers are interesting. They often have large followings. Some people pay attention to them. Sometimes they can raise awareness to products and services. But they have a trust problem. Not everybody trusts influencers because they’re not the same as their friends. If your friend says, “Oh my God, I went to this restaurant. It was great,” or a colleague in another organization says, “I tried this software, and it worked really well for us,” you  listen to them because you know they didn’t get paid. But when someone online that you don’t really know so well talks about something or shares a piece of content, you don’t know whether they did it because they actually like it or not, or just because someone paid them. So, there’s this challenge of, “Hey, lots of money is spent on influencers. Is all that money well spent?”

That’s really where this project started. Lots of companies and organizations are paying lots of individuals to post lots of content. Some of it is impactful. Some of it isn’t. Why? What drives impact? What types of posts from these so-called influencers have more impact? And how, by understanding that why, can we both increase the effectiveness of our spend, and make it more impactful?

Basiouny: Let’s talk about that sensory language that’s at the heart of your paper. What do you mean by sensory language? Give us some examples.

Berger: Sure, let’s talk about something like a food product. Imagine I said, “Oh my God, I tried this food, and it was really good.” “Good” is not a sensory word. It doesn’t relate to the senses in any way.

Alternatively, I could say, “Hey, I tried that food, and it was really tasty.” “Tasty” relates to the senses. It speaks to sensory experience — in this case, the experience of taste. I tasted the food, and I liked the way it tasted. It was tasty.

Similarly, I could talk about, let’s say, a dry skin cream. I could say, “Hey, you should put it on your hands.” And that’s not really a sensory word. I can talk about “rubbing” it on your hands. That’s a word that relates in some way, shape, or form to the senses. We can think about our sense of touch, our sense of smell, our sense of taste, our sense of hearing, even our sense of sight. Some words relate to those senses more than others, unlike a positive word like “good” or even “great.” It could be negative words, also. These sensory words really touch on the senses in some way, shape, or form.

Rather than “putting” peanut butter on bread, I could “spread” it on that bread. Rather than “adding” flakes on top of something, I could “crumble” them on top. Steak could be really “good,” or it could be really “juicy.” All of those latter words speak more, in this case, to a sense of taste or a sense of smell or a sense of touch, and that’s what makes them sensory language.

Why Word Choice and Specific Types of Language Matter in Marketing

Basiouny: It seems like sensory language would be a no-brainer. They’ve been doing this for a long time in television commercials and radio ads. But your paper points out that the effect of sensory language on consumer behavior, especially online consumer behavior, has been overlooked in the literature. What does your paper contribute to this?

Berger: Yes, it’s a really good observation that particularly food marketers have, for a long time, recognized that when you show a hamburger, you want to show it looking plump and juicy. But I think even traditional advertisers have not always thought about language. There hasn’t been amazing research that has been done. There has been a little bit of work, but not a lot of work on the exact words we use to describe things — in part because it was tough to study. In the last five to 10 years, there has been a revolution in our ability to extract insight from content, and that has really opened up some exciting work on language.

In this case, we tried to begin to say, “What is the language that influencers use?” Different people use different types of language. Most influencers are not necessarily experts on language. Some of them just post content, and they try to figure out what works and what doesn’t, but they may not always have insight into why. We got a large dataset of hundreds of different influencers posting thousands of pieces of content across multiple different platforms, and we began to look at actually the content they were posting.

Controlling for everything else, controlling for what topic they’re talking about, what brand they’re working with, how many followers they have, their gender, their age, what time of day they posted it, the other words they’re using. Might the specific type of language — rather than saying a food is “good,” saying it’s “tasty,” rather than saying it’s “great” or “wonderful,” saying it’s “juicy” or “fragrant” — these words that suggest direct experience, might they actually have a bigger impact? In fact, we found that they did. In a variety of domains, we found that using sensory language, rather than this other type of language, had an impact, a pretty large one. In TikTok, for example, just one additional sensory word in that video was associated with 11,000 additional likes and comments.

We not only looked at the impact, we looked both in the field as well as experimentally. We manipulated those words and asked people to imagine that an influencer posts one thing versus something else. We showed a causal impact of that language on behavior, but we also dug into why.

As a scholar of consumer behavior, I always find the “why” really interesting and important. My co-author, the first author of this paper, Luca (Giovanni Luca Cascio Rizzo, marketing professor at USC Marshall School of Business), really did the hard work on this. What he pointed out is, “Look, influencers have a big challenge. They’re talking about all these products, all these services, but people don’t know whether to trust them or not.” Even if you say something is “good” or it’s “great” or it’s “wonderful and I love it,” lots of people say those things. How do I know that I can trust you or not?

But sensory language does something interesting. It suggests that you’ve actually had direct experience with that thing. If you’re saying, “Hey, it’s really juicy or really tasty or really fragrant or really smooth.” “Good” is a positive adjective, but “smooth” suggests you’ve actually felt it. “I love these cookies.” That’s really positive, but, “They have a big crunch when you bite into them,” suggests that you must have actually bitten into them. Anybody can say that cookies are “great,” whether they’ve tried them or not, but to say they have “crunch,” you must have actually experienced it. And you can imagine the same thing for other types of experiences, whether it’s a consumer product or whether it’s more of a B2B software, back end sort of thing.

Using sensory language suggests direct experience. It suggests someone actually tried that product or service they’re talking about. And because of that, it makes the speaker seem more authentic. It makes them seem like they’ve actually used that shampoo or relied on that software package. “I liked it enough to use it myself, and so I’m not just recommending it because someone paid me, I’m recommending it because I actually liked it, because I tried it.” And wow, if you actually liked it and you tried it, I’m much more likely to trust you.

How the Right Word Choice Can Help Influencers Build Trust

Basiouny: Using sensory language can move the influencer from someone who is perceived as just getting paid to endorse the product, to someone who has experienced the product, likes the product, is endorsing it because they believe in it.

Berger: Exactly, and that’s a big challenge. Do I trust this person? If you just met someone, you try to figure out, “Do I trust them or not?” It’s particularly difficult because we don’t have the most trust for them to begin with. But we use cues, like the language they use or the photos they post or other things, to make inferences about them. In this case, sensory language suggests direct experience. “Hey, I’m not just saying this. I actually experienced it, and if I’ve actually experienced it, I’m more authentic.” Now, I’m going to trust that person, and I’m going to be more likely to buy the product we’re talking about.

Basiouny: What is your takeaway for marketers or social media influencers who want to get better at what they’re doing?

Berger: I think this has both direct takeaways for individuals who are “influencers” or who want to have influence. Many of us may not see ourselves as influencers, per se, but we may want to have influence — more on the folks that listen to us, whether online or off. But also the brands of the organizations that work with these individuals. I think what’s particularly neat here is this is not a big ask. We’re not asking you to have 10,000 more followers or drastically overhaul everything you’re doing. Subtle shifts, really small shifts in the language you’re using can have a big impact. As I mentioned, one additional sensory word is associated with over 10,000 additional likes and comments on TikTok.

By understanding what sensory language is and understanding how to work it into what we’ve done, what we’ve posted, what we’re sharing, we can have a big impact on our content’s eventual influence.

Basiouny: And it doesn’t cost anything extra.

Berger: It doesn’t cost anything extra, no.