In what ways do apps get better? For many of today’s top applications, improvement comes from heeding customer feedback.
- Instagram In response to user demands for in-app editing features, they introduced options like filters as well as controls for adjusting brightness and contrast.
- Offline maps, by Google Maps responded to users who wished to utilize the tool during periods of unavailability online.
However, paying attention to user feedback is not a simple endeavor. Just the Apple App Store by itself provides 3.8 million apps up to 1.8 million reviews each.
Recent findings from Texas McCombs provide insights into whose advice to heed and whose to dismiss. By employing artificial intelligence, this study examines user feedback and its impact on app modifications. The outcomes indicate scenarios where developers ought to consider user input for incorporating new functionalities versus instances where they might be better off pursuing their original concepts. Should You Follow Your Heart or Heed User Feedback? The Dilemma in Mobile App Development " is forthcoming in Information Systems Research .
Companies can gather insights into desired product features directly from customers, as these individuals often understand their requirements best," explains Anitesh Barua, chairman and professor at the IROM department. "Nonetheless, research has yet to explore how effective suggestions made by users really are when it comes to new features.
These elements typically belong to two groups, as he explains: they either mimic functionalities found in rival applications or introduce innovations not offered by competitors.
Ashish Agarwal, an associate professor from the Department of Information, Risk, and Operations Management (IROM), along with Aditya Karanam, who was formerly a doctoral student at the National University of Singapore, collaborated with Barua to examine both innovations and imitations—those sparked by developers as well as ideas suggested by end-users.
From a collection of 853 highly rated Apple applications spanning from 2012 to 2016, they employed artificial intelligence techniques to analyze over seven million review comments for insights. To accomplish this, they utilized an adjusted version of the BERT language model, which was instrumental in extracting relevant information based on word sequences found both in app update descriptions and within user feedback sections.
During test runs, their model surpassed the well-known GPT-4 in recognizing both user-proposed and developer-introduced functionalities, particularly when data was scarce.
Next, the researchers examined each application along with its updates to determine which feature concepts originated from users versus those developed internally. They also looked into whether these features were replicas of functionalities found elsewhere. Ultimately, their analysis aimed at quantifying how significant an impact individual new features had on user engagement, gauged through ranking metrics and overall demand trends for applications.
They discovered that the worth of user recommendations varied based on the type of feature.
- Innovations—features not copied from other applications—are driven by developers tend to increase demand. However, those initiated by customers often lead to a decrease in demand.
- On the contrary, for imitations, the outcomes were opposite. Features proposed by users boosted demand, whereas those introduced by developers had no impact.
What accounts for this discrepancy? Researchers propose that customers might excel at articulating attributes from products they have encountered before rather than those they haven't.
App developers, Barua says, can use the results to help determine when to heed their customers—and when not to.
He suggests that for incorporating novel features, developers ought to trust their intuition unless they can thoroughly understand user feedback, which tends to be unclear and could potentially lead to negative outcomes rather than beneficial ones.
When it comes to mimicking other applications, though, customers tend to be correct more frequently, he notes. "In these situations, they ought to pay attention to user feedback rather than solely following their instincts," he says.
More information: Subrahmanyam Aditya Karanam et al., Should You Follow Your Intuition or Heed User Feedback? The Case of Mobile App Development, Information Systems Research (2025). DOI: 10.1287/isre.2023.0060
Furnished by the University of Texas at Austin
This tale was initially released on Tech Xplore . Subscribe to our newsletter For the most recent updates on science and technology news.