bentleystrategy.com

Not All Customer Feedback Is Created Equal

Your VoC data is almost certainly skewed, because the customers who give feedback aren't the customers who matter most. A framework for getting the customer insight that actually drives growth.

Your VoC (Voice of Customer) report just flagged a CX problem. Customers are frustrated. The feedback is clear. It's time to take action.

That's living the data-driven and customer-centric culture, right?

Not so fast.

Before you act, do you know which customers are frustrated? Most VoC data aggregates all customer responses together. It's not segmented to tell you how your worst, average, or best customers are each responding. And it's almost certainly not weighted to account for the fact that your best customers are less likely to respond to surveys in the first place.

So when you find that CX issue customers are frustrated about, have you confirmed whether it's consistent across all customer types, or concentrated in a particular segment? Does it matter more to your most valuable customers or your least valuable ones?

If you can't answer that question, you're not alone. But the decisions your team is making without that answer are costing you more than you think.

Your VoC program is treating all customers equally. It shouldn't.

Most companies treat customer feedback like a democracy. Every vote counts equally. A survey response from a customer who spent $50 with you last year carries the same weight as one from a customer who spent $5,000.

This creates a structural problem that compounds quietly in the background of every roadmap meeting, every CX investment decision, every customer service policy.

Your VoC data is probably shaped more by your average and low-value customers than by the customers who matter most to your business.

Your best customers are often your quietest ones. They don't complain. They skip surveys. They rarely write reviews, because they're busy, satisfied enough not to feel compelled to respond, and their threshold for raising a concern is higher than the average customer's. Research on survey response behavior shows that people at the extremes of satisfaction (very unhappy or very happy) are most likely to respond. Your highest-value, loyal customers are systematically underrepresented from the start.

Most companies aren't segmenting their customer feedback data by customer value, or any customer dimension. Every response goes into the same bucket. That's where well-intentioned, data-driven teams can get it wrong even when the surface data backs them up.

Your loudest customers may be leading you astray

Picture this: you're in a leadership meeting where the latest VoC report is presented. The report shows a clear CX issue on your website. Customers are frustrated and the feedback is clear. You ask your Digital team to reprioritize other work and fix the issue.

Great customer-centric, data-driven leadership, right? You used data to identify the most pressing issue, empathized with the customer experience, and prioritized investment to fix it. That's exactly what you're supposed to do.

But did anyone in that room ask which customers were frustrated?

If you segmented that same feedback by customer value and discovered that your top customers, the ones driving 50% of your revenue, don't even rank this as a top issue and are actually more frustrated by something else entirely, would you have made the same call?

The question isn't whether you have customer data. It's whether you know whose data you are prioritizing and acting on.

This is the kind of mistake that results in big CX investments without commensurate revenue or profit improvements. The CX ends up being optimized for low-revenue or low-profit customers, those that spend little and cost a lot to serve. Meanwhile, high-profit, high-growth, low cost to serve customers don't see any meaningful improvement in the parts of the experience that matter most to them.

This impacts more than just CX decisions. Consider a major decision on whether to offer bulk ordering capabilities or volume discounts. The majority of customers, who buy infrequently and in small amounts, would probably not rank this highly on surveys. But your high-frequency, high-volume customers would. One way to avoid this trap: weight your survey responses by last-12-month spend. If a customer spent ten times what others spent, shouldn't their vote count more?

The Five-One Playbook (five steps, one week)

The good news: this doesn't require a large-scale analysis. In most organizations, the data to do this already exists. It just hasn't been filtered or connected. Here's a framework to help you get results quickly.

1. Get the instinctive read first. Ask your Consumer Research or Customer Insights team for their top 3 opportunities and challenges based on everything they know. Customer feedback data is often a vast ocean with a long history. Trust your subject matter experts to boil it down to the top few things to focus on. This is your baseline, and is likely the underpinning of current CX initiatives.
2. Segment the data by customer value bands. Ask the team if they segment survey responses, NPS scores, or behavioral feedback by spend tier, tenure, and/or purchase frequency. This might mean matching survey data to your CRM or transaction records. If it takes more than a few days to produce, note that. It's a signal about your survey methodology and data infrastructure worth investigating. Improving your customer data supply chain and chokepoints pays dividends far beyond this exercise.
3. Ask the same question again, but only for your best customers. What are the top 3 opportunities and challenges, now filtered to your highest-value segment. Where does the priority list change? Where are your best customers saying something meaningfully different from the aggregate? That gap is your insight.
4. Run your current investments through this lens. Look at your marketing messages, CX improvements, Customer Service policies, promo strategies, and product roadmap priorities. How many were shaped by undifferentiated feedback? If your best customers' top frustrations aren't on your priority list, reprioritize at the earliest opportunity.
5. Find one thing to do differently and act on it this week. You don't need to rebuild your entire VoC program. Start with one insight, one decision, one change, and test whether it moves the needle for your best customers. Momentum matters more than completeness.

Insights you'll gain

When you start segmenting VoC data by customer value, a few patterns tend to surface. These will be unique to your brand, customers, and current situation, but here are three worth watching for:

1. Your best customers are getting average-customer treatment when something goes wrong. A customer spending 10× the average who receives the same $10 apology credit as everyone else notices the disconnect. It signals that you don't know who they are and what they mean to your business. This quietly erodes the relationship over time. Work with Customer Service and Finance to test a differentiated recovery offer for your highest-value customers.
2. Your best customers are your earliest warning system on product issues. Because they use your products more intensively and frequently, they surface quality or design issues before anyone else. If your top customers are significantly more likely to flag a problem with a new product, take it seriously. They may be identifying a flaw the average customer hasn't encountered yet.
3. Your best customers are telling you what they want you to sell to them. If your top customers are consistently frustrated when they can't complete a related purchase in the same transaction, that's a commercial signal worth investigating. Carrying a lower-volume product might be worth it: high-spending customers value their time and convenience, and if another brand meets that need, you risk losing their entire basket, not just that one purchase.

Build a system that amplifies your best customer feedback

Your VoC listening station isn't broken, it just might not be dialed into the right channel. If it is giving you a democratized, blended picture of your customer base, it may be steering you toward decisions optimized for the average, rather than your most important customer.

The shift isn't about ignoring everyone else. It's about making sure the voice of your most valuable customers isn't getting lost in the aggregate. Build a system in which best customer feedback is amplified and weighted above the rest. This shouldn't be the only way you look at your data, there are real risks to over-indexing on any single segment, but having a view that weights best customer input more heavily will help you see their needs more clearly.

When you know what your best customers actually want, and where they're most frustrated, you make better decisions about where to invest, what to fix, and what to build next.

Most companies compete for the average customer. That's a fight for incremental, low-margin growth. Understanding and capturing more of your best customers is the path to high-margin growth.