Using Feedback to Fuel a Better Customer Experience

Without customer feedback, you’re basically operating in the dark.

If you don’t know what you’re doing right, you can’t do more of it. If you don’t know what you’re doing wrong, you can’t do less of it.

In short, feedback is the lifeblood that allows us to course correct points of failure and double down on what we’re doing well.

It can help almost any team throughout the organization, and nowadays, it’s so easy to collect, that there’s no excuse not to do so.

This article will outline the importance of feedback with regard to multiple teams and then cover some tangible ideas for using feedback to improve the customer experience.

Customer Feedback is the Breakfast of Champions

“Feedback is the breakfast of champions.” – Brian Halligan, HubSpot, CEO

Customer feedback is important, primarily because we’re all too close to our own company and product to see things objectively. In addition, in the big drive to become more customer-centric, it’s rare that companies actually, well, listen to their customers.

Real-time feedback is the remedy, the course corrector to help inject some customer-centricity into your organization.

If you’re familiar with marketing and conversion rate optimization, you know there’s a whole class of quantitative behavioral data that can help you understand problems with your product, sales, support, or marketing operations. Tools like Google Analytics can really help you understand where your funnel is broken and what potential issues you may have.

However, while this information gives you great insight on what’s going wrong (e.g. users are dropping off on the checkout page, or no one is using your product’s newest feature), it’s not very good at giving you insight into why that is or what the problem could be.

In fact, here’s a video for you to choose your customer feedback tools smartly:

That’s where feedback comes in. When you combine the quantitative with the qualitative, you can come up with truly evidence-backed A/B test hypotheses to continuously improve the customer experience.

You can use this insight everywhere, in fact. Marketers and optimizers can churn out better experiments, but you can also use it to inform product and feature design, customer service operations, sales messaging, and more.

Offline teams can apply the same approach. At live events, in-the-moment polls and post‑session surveys surface issues like audio dropouts or slow changeovers so organizers can act fast. Translating those insights into staffing decisions to manage rigging, load‑ins/outs, and backstage cues helps deliver smoother production and a measurably better attendee experience.

As Aarron Walter, VP of Design Education at InVision, put it:

“Designers aren’t the only ones who stand to gain from the data collected in a survey—anyone who makes things for or communicates with customers will find themselves empowered to do better work when they know more about the people they serve.”

Outside of insight for iterative customer experience changes, some forms of customer feedback, such as Customer Effort Score and Net Promoter Score, have been shown to correlate with customer loyalty.

Metrics like NPS and CES can correlate with customer loyalty in many contexts, and satisfied customers do tend to have higher lifetime value and refer others more often. That said, the relationship between survey scores and business outcomes is directional, not deterministic. Treat these metrics as useful signals that warrant investigation, not as standalone predictors of revenue or retention.

As Rob Markey, Fred Reichheld, and Andreas Dullweber put it in an HBR article:

“Increasing positive customer feedback and meeting conventional financial objectives are becoming one and the same goal.”

Three Ways to Use Feedback to Create a Better Customer Experience

These aren’t the only places you can utilize the customer feedback you collect; if you get creative you can use real-time surveys or other forms of feedback collection to capture data at almost any point in the customer journey.

But the following three ideas will give you a valuable place to begin collecting and using feedback.

1. Customer Service Touch Points

One of the most basic ways to use customer feedback is to improve customer service touch points.

This could mean active support encounters like live chat or support tickets, or it could mean self-service support using a software like ProProfs Knowledge Base.

The easiest way to get started — if you’re not doing this already — is to ask for feedback after a support encounter.

Most customer support tools do this automatically and you just have to analyze and act on the data. If, for some reason, you’re not doing this already, simply implement a post-support survey, like the following:

Stepping back, though, the actual support experience itself can serve as customer feedback, too.

For instance, if you codify customer service tickets and find tons and tons of people are asking the same question or struggling with the same thing, you can a) fix the problem (if that’s possible) or b) create self-service content in your knowledge base to help scale your support efforts in the meantime.

Which leads to my next point: you should be collecting customer feedback in your knowledge base articles as well. Here, it’s as simple as asking a binary question: was this article helpful?

With this simple question, you can gather some straightforward data on what articles aren’t helpful and if you’re improving the helpfulness of your knowledge base over time.

Asking a binary feedback question on help content tends to be more effective than using a proxy metric like time on page.

While you may have an opinion on the matter, there’s a lot of ambiguity over the effectiveness of engagement metrics. In other words, do people stay on site longer because they’re more engaged with the great content or because they found their answer fast and are super satisfied?

In an article for HubSpot, Adam Avramescu talked about how they measure success at Optimizely’s Optiverse. Essentially, they use Discoverability and Value as the key variables.

They measured Discoverability in terms of pageviews, but also looked at typical referral paths to see where people are coming from.

They measured Value through a one-question feedback survey asking a simple question: did you find this article helpful?

They then ran a basic calculation where the number of “Yes” responses out of the total is that article’s helpfulness index.

That way, Optimizely can determine which articles in relation to the rest of them is helpful, unhelpful, or otherwise. With this data, they can also determine helpfulness over time and see if things are generally improving, both at the aggregate and single article level.

2. Product Suggestions and Improvements

If you sell a software product, it’s easier than ever to ask for feedback directly inside the app. Heck, even if you don’t sell software, you can usually get “product” feedback and suggestions if you ask. Here’s a form in an apartment complex, for instance:

While it can be daunting to determine which question to ask in the product experience (you want to avoid disrupting the UX too much but also get valuable insights), there are a few popular questions to ask depending on your goals:

• “What is {product feature} helping you accomplish?”

• “What issues, if any, are you having with {product feature}?”

• “What features do you think we’re missing today for {product feature}?””

Using questions like this you can accomplish goals as different as learning your customers’ Jobs to Be Done, hearing complaints and frustrations, and crowdsourcing desire for new features.

Furthering that last point, many companies are now collecting feedback on feature requests using a straightforward “feature request board. For example, here’s the page we use at HubSpot to source feature requests:

Note, however, that this tends to be biased towards your most passionate, highest-intent customers. That’s not a bad thing in itself, as you can clearly still get good ideas from these people, but it’s good to know that they may not reflect your customer base at large.

In summary, using real-time surveys, feature request boards, and qualitative data from live chats and support tickets, you can gather a variety of product-level data, including:

• Major product bugs: Things that are detrimental to the customer experience and should be prioritized.

• Minor product bugs: Minor annoyances that may not destroy the value of your product, but can build up frustration in users.

• Feature requests: Forward-facing new ideas to consider in balance to bugs and frustration issues.

Listening to the customer opens up powerful product design insights. As Jeremy Bailey wrote in TechCrunch, “I’ve found that the most valuable design tool in the 21st century is rather old-fashioned: customer feedback.”

Side note: Sprint by Jake Knapp outlines a framework for solving big problems that is largely based on modeling early iterations of a product or experience and getting early feedback. The plot doesn’t quite fit this article, but the book is worth a read if you’re into innovation.

3. A/B Test Ideation

On-site surveys are a crucial tool used in conversion research.

I’ve used Qualaroo and other tools extensively both with clients and in my work at ConversionXL. Almost every project proves that — if you’re asking the right questions — real-time surveys are one of the most insightful data points for test ideation and spotting issues and opportunities.

In the past, I’ve used on-site surveys mainly as an extension of my research goals. For instance, if I’ve found via analytics analysis that a certain page is converting poorly or has a much higher bounce rate than other similar pages, I may add a survey that asks one of the following questions:

• “Were you able to find the information you were looking for?”

• “Did this page meet your expectations?”

Or if, for example, I’m doing more general persona research and want to understand visitors at a deeper level than their digital behavior shows, I may simply ask one of the following questions:

• “What’s the purpose of your visit today?”

• “Why are you here today?”

Finally, when I want to get broader Voice of Customer insights into why users prefer our brand and end up purchasing, what hesitations and doubts they may have had, or what their comparison shopping experience may have been like, I could ask one of the following:

• “What persuaded you to purchase from us?”

• “Is there anything holding you back from completing a purchase? Y/N” (and then ask for explanation)

• “Which other options did you consider before choosing our product/service?”

So there’s really an unlimited variety of questions you could ask, depending on your goal (the above are just a few common ones I’ve used). In addition, when you pop the question depends a lot on your goal as well.

For instance, the above example question, “What persuaded you to purchase from us?” would be best suited for a Thank You Page. But something like, “Were you able to find the information you were looking for?” would be better off on a product page or somewhere the visitor is seeking information.

Here’s a tangible example from Ryan Farley, founder at LawnStarter.

They had created a pillar page of sorts for SEO purposes – the definitive guide to lawn care service. It worked to that end and garnered a bunch of traffic. But compared to other blog articles, it had a high bounce rate and strange user behavior patterns.

So Ryan ran an open-ended survey on the page, asking “What did you come here for?” As expected, many people were looking for info on lawn care, but many were also seeking to purchase lawn care services.

They took the open-ended question data, coded them into four variables, and then put up another poll with four choices. Here are the options and the corresponding response numbers:

I want to…

• Learn about lawn care (33%)

• Answer a specific question I have about my lawn (14%)

• Read reviews about local lawn care professionals (7%)

• Hire a lawn care professional (47%)

Most SEO pages they had done had a pretty singular intent, but this one was split between 1) informational and 2) transactional intent. So, they split the page.  

In the new version, the top part of the page explains how LawnStarter works and how to book lawn care, and the bottom has their lawn care guide, broken up into chunks.

The results: an increase in conversion rate, a decrease in bounce rate, and even an increase in traffic (presumably because Google values better engagement metrics)

Survey-driven hypotheses are a strong starting point. They are not a substitute for properly run experiments.

A few principles worth applying before you move a survey insight into a test:

Define Your Success Metric Before You Design the Test: If your survey found that users were confused by your pricing page, your A/B test needs a primary metric, such as clicks on the pricing plan CTA or signups from that page, agreed upon before the test runs. Post-hoc metric selection is one of the most common sources of false positives in CRO.

Run Tests Long Enough to Reach Statistical Significance: A test that runs for three days and shows a 30 percent lift is almost always noise. As a baseline, aim for a minimum of two full business cycles (typically two weeks) and a sample size large enough to detect a realistic effect. Most A/B testing tools will calculate the required sample size if you input your baseline conversion rate and minimum detectable effect.

Treat Survey Insights as Directional, Not Prescriptive: A survey that says users want feature X does not mean building feature X will improve conversion. It means that feature X is worth hypothesizing about and testing. The test result, not the survey, is what validates the change.

Still not sure what to ask after all that? You could just follow Avinash Kaushik’s “three greatest survey questions ever.” He says that when he was asked which analytics tool to recommend to a VP on a short time frame, he responded that she shouldn’t install an analytics tool. Instead, she should install a customer feedback survey tool and ask these three questions:

• What is the purpose of your visit to our website today?

  Were you able to complete your task today?

• If you were not able to complete your task today, why not?

There you have it: Avinash Kaushik, an analytics Evangelist at Google, wants you to start listening to your customers. What are you waiting for?

How Do You Turn Feedback Into Decisions, Not Just Data?

Collecting feedback is the easy part. Most teams have surveys running. What separates teams that improve the customer experience from teams that just measure it is what they do after the responses come in.

Validate themes before acting on them. A single comment about a confusing onboarding step is anecdotal. Twenty comments about the same step across different user segments is a pattern. Before escalating any piece of feedback, group responses by theme, filter by volume and recurrence, and confirm the pattern holds across multiple cohorts or channels.

Separate signal from noise with a simple triage framework. Not all feedback deserves the same response speed. A useful starting point:

Assign ownership at the point of collection, not after. Feedback without a named owner tends to stall. Before you launch any survey, decide which team is responsible for reviewing responses, who escalates critical findings, and what the expected response window is. Build this into the survey workflow, not as an afterthought.

Validate qualitative findings with behavioral data. If survey responses suggest that users find a particular feature confusing, check whether session recordings or drop-off data from that feature corroborate the claim. Qualitative feedback tells you what people say. Behavioral data tells you what they do. When both point in the same direction, you have a strong basis for action.

Close the loop with users and internally. When feedback leads to a change, communicate it. A brief changelog note, an in-app message, or a follow-up email to the respondent cohort builds the trust that keeps response rates healthy over time. Internally, share what you heard and what changed in a regular cross-functional review so that product, support, and marketing are working from the same feedback picture.

Conclusion

Customer feedback is only as valuable as what you do with it. Collecting it is a starting point, not an outcome.

This article covered three practical entry points for feedback collection: customer service touchpoints, product improvements, and A/B test ideation.

But the work that creates a measurably better customer experience happens after collection: validating themes, assigning ownership, cross-referencing with behavioral data, and closing the loop with the people who gave you the feedback in the first place.

No single department owns this. Product, support, marketing, and leadership all benefit from feedback when it flows to the right place. The teams that get the most out of it are the ones that treat it as an operational input, not a reporting exercise.