How AI can turn customer reviews into innovation funnels

Customers know best. And now that so much commerce has moved online, they’re telling us more than ever. But how best to hear them and engage?

About half of internet users post reviews at least once a month¹. Many of these are comments about electronics they’ve just purchased. This unmediated offering of market intelligence can be crucial not just for customer service and marketing, but also for product development and innovation.

Yet many businesses aren’t taking the opportunity to respond and learn more from consumers. A separate survey showed 63% of consumers say they’ve never heard back from a business about their negative reviews². This potentially means millions of businesses are missing out on the opportunity to funnel key product information from users to innovators who are designing new products. 

We also know that almost all consumers refer to online reviews -- 9 out of 10 according to one survey -- before making a purchase. This highlights just how important reviews are to both buyers and sellers. Research also suggests that search engines are using review sentiment as a ranking factor for products and brands. Bottom line: Every business needs to understand what consumers are trying to say in online reviews and to analyze and digest this data continuously.

Businesses that respond to customer reviews make more sales. On average, businesses that responded to just one customer review earned 4% more revenue. When an enterprise replies to at least 25% of their online customer reviews, on average, they earn 35% more revenue.

Now for the hard part. Assigning workers or even executives to directly monitor customer reviews can be like asking them to wrap their lips around a firehose. And even if the volume is small enough to keep up with, the time necessary to objectively, consistently analyze each response is prohibitive.

Luckily, machine learning and artificial intelligence models can analyze customer reviews quickly, objectively, consistently and continuously. The longer these models run, the more accurate and insightful they become. The results are displayed in a dashboard comparing the performance of products across a range of metrics drawn from consumer sentiment about the products and services. Most important, perhaps, the results can tell managers when it’s necessary to engage directly with consumers posting reviews, and when to simply take in the information.

The ability to compare consumer sentiment about products with that for rivals’ offering and across time, especially when it relates to adding new features or design changes, can be a very powerful tool for product development. Think of it as an innovation information funnel delivering broad, accurate insights to guide marketing, customer service and product development. 

The full power of this approach becomes clear when dealing with reviews across global markets in different languages with varying cultural settings. An algorithm designed for this can be consistent, objective and accurate regardless of language or market culture.

IgnitusAI adopts a data-driven approach to analyzing consumer reviews. Powerful scripts “scrape” the entire internet for relevant product reviews, then annotate the data for training algorithms.

The sooner you begin this process the more powerful your costumer review analysis platform can become for understanding what consumers really want and what they’re willing to pay for. 

Let’s get started. For more information, connect with us here.

‍

‍

_{*1 GlobalWebIndex, 2019; *2 Brightlocal 2020 research; *3 Trustpilot 2020 research}