Traditionally, marketers have conducted brand research using surveys, focus group discussions and other primary research methods to understand consumer needs and generate insights on consumer behavior. These insights can be both qualitative and quantitative and can help a business make better decisions. But there are obvious limitations with this approach, not the least of which is the amount of time one of these studies takes (around three to four months). But in today’s digital age, there’s an entire mass of readily available marketing research data generated every single day: the consumer review.
The total number of reviews and opinions on TripAdvisor worldwide, for instance, has increased gradually since 2014 and an estimated aggregated number of reviews peaked at roughly one billion in 2021, rising by approximately 13% over the previous year. India alone saw over 551K smartphone users post reviews online in the first half of 2020. Going through consumer reviews for a product has become an integral part of the buying journey in the digital world.
For brands, this means an impact on conversion rate and sales. On average, the conversion rate of a product can increase by 270% with a steady stream of new reviews. High-priced items can see their conversion rate increase by as much as 380% as they accumulate reviews, and low-priced items can see their conversion rate increase by as much as 190%. The existence of reviews clearly provides a valuable signal to customers, increasing their propensity to purchase the product.
There is a great deal to be gained from thinking of reviews as data. Companies that ignore this are not only doing themselves a disservice but also losing what is going to be an increasingly important line of vision into their customer base.
There are a lot of granular insights that can be teased out from this review data. In most cases, consumers leave very detailed feedback, such as, “The store experience was great, but the design variety could have been better,” or “The fit and feel was excellent, but the size was smaller than usual.” What’s needed then is an artificial intelligence (AI) tool that can extrapolate from these reviews things like brand perceptions, product attributes and consumer preferences.
A solution that effectively mines reviews must be able to identify the factors that matter most to consumers and understand the sentiments associated with them. We’ve found that it’s critical in mining these reviews to link sentiment factors and quantify the impact those factors have on ratings. In addition to this, an AI tool needs to do four other major factors in order to create actionable insights for a brand:
- Identify the consumer need. This is the starting point for finding a brand’s perception and formulating a growth strategy around that.
- Quantify the impact of that need. There needs to be real data that can be used to show that the insight you glean from reviews can equate to a real change for the company.
- Inform competitive strategy. The tool should be able to take insights from the reviews and compare them to insights found on competitor sites.
- Understand the buying context. This is the sentiment aspect, and it’s critical to using reviews for market research purposes.
We’ve found in using our own tool that these factors have been successful with clients, as demonstrated in the two real-world examples below.
A lingerie brand came to us to get information on how their brand was performing in the Singapore market. Its goal was to position the company as a leader in one specific segment of the category. We started by scraping data from multiple marketplaces, collecting around 150K reviews—which allowed us to create an analysis that had some very insightful observations.
There were very specific category-level challenges related to some very important product attributes, like size and fit, which were discovered during the project. The product team worked with suppliers to alter the product profile based on the feedback received. This helped them to move ahead of the competition since these problems were brand agnostic.
We also derived insights related to challenges in product delivery. The brand team worked with marketplaces to understand the challenges in detail. Eventually they were able to streamline the delivery process and ensure the right products are delivered to customers.
The AI product is now being leveraged across various markets in the Asia-Pacific region to better understand consumer needs and product preferences.
In another case, we worked with a retailer that had rich net promoter score (NPS) data collected at their store level. Along with the NPS score, the free text option allowed consumers to write detailed reviews about products, store-level service, etc. ZS’s AI solution enabled the retailer to look at store-level nuances and regional consumer preferences to inform the retailer’s operations model and pricing and product strategy.
Our customer experience team got a view into consumer sentiment on the purchase process, overall store experience and post-purchase store experience, which had many areas of opportunity. The retailer revamped the billing and purchase process, making it streamlined and faster.
The insights generated were also able to point out specific challenges on staff behavior and the shopping experience with key stores in one region. The retailer focused on in-store staff training and rotation while also tagging better performing stores with the underperformers to alleviate the people challenges.
Digital has enabled consumers to be more vocal than ever before, and it’s imperative that brands utilize this rich information as part of their strategy. Relying solely on traditional market research methods is a serious disadvantage—one that may cost brands dearly in the future. In today’s digital world, that means utilizing advanced AI tools that not only enable insight generation at speed and scale but also read these insights within the wider context of like-minded consumers.