How AI-Powered Text Analytics Help Retailers Understand Their Customers

The retail industry is drowning in customer feedback. From reviews to digital surveys to in-store polls to social media posts, customer feedback is everywhere.

Smart businesses know this feedback is valuable. They try to glean insights about their business based on customer feedback. The most valuable feedback is qualitative feedback, left in open text boxes and essay boxes in surveys.

However, qualitative feedback has been too difficult to categorize at scale — there are simply too many responses to read. Qualitative feedback simply isn’t actionable for many retailers due to the complexity and scale of the feedback received. Thus, collecting feedback in open text has made the data unworkable and unusable for most businesses.

Retailers have three options for collecting qualitative data:

1. Don’t collect qualitative data. Some retailers have forced their customers to deliver their feedback using radio buttons or checkbox questions. However, these miss the richness of customer feedback. Radio buttons can’t encompass the entirety of the customer experience.
2. Collect qualitative feedback, but don’t do much with it. Retailers may opt to collect qualitative feedback in open text boxes, but due to the size and complexity, many retailers find this analysis unmanageable. They may pay an employee to manually tag feedback with sentiment data (is it positive or negative?). This can give a general indication if you're on the wrong or right path, but it doesn't deliver insights teams can use, plus is ripe with bias.
3. Use open text analysis (OTA) to segment customer feedback. OTA is an emerging technology that uses artificial intelligence-powered large language models (LLMs) to analyze and measure sentiment and themes found in open text responses. OTA quickly summarizes large volumes of feedback, identifies and quantifies the themes/sentiments, and provides visibility into the combined analysis of cross-channel feedback sources.

The most skilled retail professionals will opt for OTA to analyze and measure customer sentiment and themes. This is the best way to receive actionable qualitative customer feedback.

There are four key capabilities that retailers should look for in an OTA platform:

1. Text analysis: Instantly summarize unstructured, open text feedback at scale using phrasal analysis to identify and categorize your themes/sentiments.
2. Analytics: Understand the voice of your customers by tracking and visualizing key feedback metrics and measuring how — and why — customer sentiment changes over time.
3. Dashboard: Eliminate siloed spreadsheets and static reports. A reputable OTA platform will empower you to customize dashboards that instantly align every team on your target outcomes and show stakeholders metrics to identify where they need to act.
4. Workflows: Use workflows to trigger instant alerts when important events happen (e.g., your NPS score crashes or you see a spike in login complaints).

OTA can be incredibly helpful to instantly identify issues your customers are experiencing. Let’s examine how that could play out in the enterprise.

CX/Marketing

CX and marketing need to connect their numbers to overall business metrics. How is their work impacting the business’ success? They also need to maintain or improve NPS or CSAT scores as their business scales.

OTA provides CX and marketing professionals with an all-in-one solution to understand the link between their work (including NPS/CSAT) and company revenue. Plus, OTA platforms can automatically alert CX and marketing professionals when trends in customer sentiment start to occur.

Product

Product teams are tasked with understanding the root cause of customer churn and increasing average revenue per unit (ARPU).

OTA helps identify gaps in product quality, helping product teams understand why people start (or stop) using their product. This helps avoid misguided investments and prioritization decisions, saving retail product teams thousands of hours.

Support

Directors of customer support want to drive loyalty and retention. Therefore, they need to know the underlying causes of poor customer support experiences.

With an OTA platform, support professionals can pinpoint the issues that impact revenue and growth, taking steps to intervene. They can also locate the root causes of customer complaints in an effort to reduce churn. Support professionals leverage OTA to resolve issues faster.

AI-powered OTA platforms enable retailers to interpret customer comments, enriching unstructured data such as NPS comments and other open text feedback. Most importantly, OTA platforms translate qualitative data into quantitative data that teams can use to take action — to save a customer or to generate a new one.

OTA crunches feedback into numbers. Wise retailers will be using OTA now or in the near future to identify trends, understand sentiment, gain visibility into complaints, and make better decisions with real data.

Peter Zaidel is the director of product management for Alchemer, an enterprise feedback platform.