Typing Less, Saying More? – The Effects of Using Generative AI in Online Consumer Review Writing
Maximilian Habla
This study investigates how using Generative AI (GenAI) impacts the quality and informativeness of online consumer reviews. Through a scenario-based online experiment, the research compares reviews written with and without GenAI assistance, analyzing factors like the writer's cognitive load and the resulting review's detail, complexity, and sentiment.
Problem
Writing detailed, informative online reviews is a mentally demanding task for consumers, which often results in less helpful content for others making purchasing decisions. While platforms use templates to help, these still require significant effort from the reviewer. This study addresses the gap in understanding whether new GenAI tools can make it easier for people to write better, more useful reviews.
Outcome
- Using GenAI significantly reduces the perceived cognitive load (mental effort) for people writing reviews. - Reviews written with the help of GenAI are more informative, covering a greater number and a wider diversity of product aspects and topics. - GenAI-assisted reviews tend to exhibit higher linguistic complexity and express a more positive sentiment, even when the star rating given by the user is the same. - Contrary to the initial hypothesis, the reduction in cognitive load did not directly account for the increase in review informativeness, suggesting other mechanisms are at play.
Host: Welcome to A.I.S. Insights, powered by Living Knowledge. I'm your host, Anna Ivy Summers. Host: Today, we're diving into a fascinating new study called "Typing Less, Saying More? – The Effects of Using Generative AI in Online Consumer Review Writing." Host: With me is our expert analyst, Alex Ian Sutherland. Alex, welcome. Expert: Great to be here, Anna. Host: So, in a nutshell, what is this study about? Expert: It investigates what happens when people use Generative AI tools, like ChatGPT, to help them write online consumer reviews. The core question is whether this AI assistance impacts the quality and informativeness of the final review. Host: Let's start with the big problem. Why do we need AI to help us write reviews in the first place? Expert: Well, we've all been there. A website asks you to leave a review, and you want to be helpful, but writing a detailed, useful comment is actually hard work. Expert: It takes real mental effort, what researchers call 'cognitive load,' to recall your experience, select the important details, and structure your thoughts coherently. Host: And because it's difficult, people often just write something very brief, like "It was great," which doesn't really help anyone. Expert: Exactly. That lack of detail is a major problem for consumers who rely on reviews to make purchasing decisions. This study wanted to see if GenAI could be the solution to make it easier for people to write better, more useful reviews. Host: So how did the researchers test this? What was their approach? Expert: They conducted a scenario-based online experiment. They asked participants to write a review about their most recent visit to a Mexican restaurant. Expert: People were randomly split into two groups. The first group, the control, used a traditional review template with a star rating and a blank text box, similar to what you’d find on Yelp today. Expert: The second group, the treatment group, had a template with GenAI embedded. They could simply enter a few bullet points about their experience, click a "Generate Review" button, and the AI would draft a full, well-structured review for them. Host: And by comparing the two groups, they could measure the impact of the AI. What were the key findings? Did it work? Expert: It made a significant difference. First, the people who used the AI assistant reported that writing the review required much less mental effort. Host: That makes sense. But were the AI-assisted reviews actually better? Expert: They were. The study found that reviews written with GenAI were significantly more informative. They covered a greater number of specific details and a wider diversity of topics, like food, service, and ambiance, all in one review. Host: That's a clear win for informativeness. Were there any other interesting outcomes? Expert: Yes, a couple of surprising ones. The AI-generated reviews tended to use more complex language. And perhaps more importantly, they expressed a more positive sentiment, even when the star rating given by the user was exactly the same as someone in the control group. Host: So, for the same four-star experience, the AI-written text sounded happier about it? Expert: Precisely. The AI seems to have an inherent positivity bias. One last thing that puzzled the researchers was that the reduction in mental effort didn't directly explain the increase in detail. The relationship is more complex than they first thought. Host: This is the most important question for our audience, Alex. Why does this matter for business? What are the practical takeaways? Expert: This is a classic double-edged sword for any business with a digital platform. The upside is huge. Integrating GenAI into the review process could unlock a wave of richer, more detailed user-generated content. Host: And more detailed reviews help other customers make better-informed decisions, which builds trust and drives sales. Expert: Absolutely. But there are two critical risks to manage. First, that "linguistic complexity" I mentioned. The AI writes at a higher reading level, which could make the detailed reviews harder for the average person to understand, defeating the purpose. Host: So you get more information, but it's less accessible. What's the other risk? Expert: That positivity bias. If reviews generated by AI consistently sound more positive than the user's actual experience, it could mislead future customers. Negative aspects might be downplayed, creating a skewed perception of a product or service. Host: So what should a business leader do with this information? Expert: The takeaway is to embrace the technology but manage its side effects proactively. Platforms should consider adding features that simplify the AI's language or provide easy-to-read summaries. They also need to be aware of, and perhaps even flag, potential sentiment shifts to maintain transparency and consumer trust. Host: So, to summarize: using GenAI for review writing makes the task easier and the output more detailed. Host: However, businesses must be cautious, as it can also make reviews harder to read and artificially positive. The key is to implement it strategically to harness the benefits while mitigating the risks. Host: Alex Ian Sutherland, thank you for these fantastic insights. Expert: It was my pleasure, Anna. Host: And thank you for tuning in to A.I.S. Insights, powered by Living Knowledge. Join us next time.
