How to Use Customer Feedback Analysis to Make the Most of Your Data

July 3, 2020
Reading time:

The reasons to collect customer feedback are obvious; in order to create the best possible customer experience, you need to find out what people think about your current offerings. However, it becomes less obvious what to actually do with the data once you’ve gathered it, especially when it comes in the form of open-ended, qualitative responses. Below we cover the steps companies should take in order to draw insightful, actionable information using customer feedback analysis.

What is Customer Feedback Analysis?


Customer feedback analysis is the process of extracting insights from customer responses to things like surveys or online reviews. Having a lot of data to work with is a good start, but you’ll need to categorize, tag, and manipulate the data in order to uncover the hidden nuggets of information that will allow you to make beneficial, data-backed decisions for your business. Customer feedback analysis matters because without it, companies run the risk of becoming “data rich but information poor,” a modern business adage applicable across a wide variety of industries.

Gathering data is costly, both in terms of revenue and employee bandwidth, and data by itself means nothing. A single negative review isn’t enough evidence to confidently make a major alteration to your business. It’s only when you’re able to analyze the data and uncover overarching trends that you can make changes that truly affect your bottom line. Below we go through the basic customer feedback analysis process along with a few extra tips and tricks to make sure you’re getting the most out of your efforts.

How to Conduct Customer Feedback Analysis


Customer feedback analysis works much like any other type of data analysis. You gather the data from reliable sources, you transform the data from qualitative to quantitative, and then you analyze and manipulate the data to reveal trends.

Step 1: Gather Customer Feedback

For most companies, there are three main sources of customer feedback data:

- Online Reviews - There are a plethora of online review sites tailored to different industries. For example, Yelp is a popular platform for reviewing restaurants and attractions, while CNET is a trusted resource for electronics. You can scrape customer data from these sites manually, or, more likely, you’ll want to use a data scraping tool to save time.- Social Media Mentions - Another valuable source of customer feedback is social media. Again, this will be industry-dependent, as B2B customers interact more on platforms like LinkedIn while lifestyle and product-based companies are more active on Instagram. You can scrape customer data from comments, posts, and direct messages to your own accounts.- Feedback Surveys - For more targeted responses, many companies use customer feedback surveys sent via email to customers who recently interacted with the business in some way (such as a recent purchase or customer service call).

Take a look at our guide on avoiding response bias to learn how to structure effective customer feedback surveys.

Once you’ve gathered the reviews and comments from these sources, you can simply add them to a spreadsheet like Google Sheets or Excel, or any other data analysis tool you prefer.

Step 2: Categorize the Feedback

Now that you have your data, you have to find a way to transform the qualitative, open-ended responses into quantitative points. For example, you can’t numerically compare the reviews “The price is too high, but the quality makes it almost worth it” and “The price seems fair to me for what you get.” But if you change these reviews into “Price: ⅖” and “Price: 5/5,” then you’re able to better compare your data. Here’s how:- Option 1: Manual Coding - Transforming qualitative data into quantitative data involves something called “coding.” This is not coding in the computer programming sense, but the process of assigning codes, otherwise known as tags, to each piece of data so it can be categorized properly. With manual coding, you will need to go through each customer review and assign it a category (price, quality, customer service, etc.) and a sentiment based on a scale you determine (very negative, ⅗ etc.). For more details on this process, take a look at our in-depth guide on coding qualitative data.- Option 2: Rule-Based AI - The rule-based approach involved human-crafted and curated rule sets. Rules-based approaches look for linguistic terms such as “love,” “hate,” “like,” and “dislike.” The presence of positive and negative words defines whether a sentence is positive or negative. However, multiple word meanings make it hard to create rules. The most common reason why rules fail stems from something called “polysemy,” when the same word can have different meanings. Think of the word “hot” in context of both temperature and spicy.- Option 3: AI + Sentiment Analysis - If manual coding sounds like too much of a pain, many businesses are instead investing in AI-backed software that can actually do this process for you. Using sentiment and topic analysis powered by deep learning algorithms, these programs can go through your customer feedback data and tag and categorize each piece of feedback, automatically changing your data from qualitative to quantitative. What used to take days can now be done in minutes. Learn more about AI and text analytics here.

Step 3: Analyze the Data

Now that you’ve turned your customer responses into a set of quantitative metrics, you can start digging into your data to uncover trends. You can use spreadsheets to manipulate the data and look for things like:

  • Unexpectedly high or low quality ratings on certain products
  • Different demographic trends, such as product preferences based on age or location
  • High amounts of customer service complaints in a specific location or via a specific channel

The point of this process is to get a full view of the customer journey by understanding what is driving positive sentiment with your CX and what’s letting it down. Sometimes this can reveal hard-to-see places on the customer journey like delivery-to-door issues or a broken product.

For example, perhaps you notice that 45% of customers aged 18 - 35 rated your pricing as just 2/10 on average. From this, you might conclude that your product is priced too high to appeal to the younger portion of your customers. However, before you go about actually making changes to your business based on these trends, it’s important to ensure that the segment of responses you’re looking at is statistically significant.

If something is statistically significant, it means that you can be reasonably confident that these trends will map to the rest of your customer population and not just the sample you have data from. Take a look at our guide to survey data analysis to learn more about making sure your data is reliable.

Once you’re able to identify trends and ensure the data is statistically significant, you can start to roll out small, data-backed changes to your business with confidence. For an example of how customer feedback analysis can be used in practice, check out our report on how COVID-19 affected customers in the travel industry.

Do I Really Need Customer Feedback Analysis?


In short, yes! Every business should be conducting customer feedback analysis. You might think that just because your company is doing well and staying profitable that you don’t need to see what your customers are saying. But what if you could be doing even better, and therefore improving your overall revenue? Though customer feedback analysis is undoubtedly useful for shedding light on existing issues, the real value of this process is to uncover problems or successes that you didn’t know about or only suspected.

For example, say your Net Promoter Score increased by five points in one month. Do you know why that happened or what you can do to ensure it continues to improve? You can understand the “why” behind the score through customer feedback analysis and uncovering the “unknown unknowns.” Otherwise, these valuable but hidden insights would have stayed out of sight, out of mind.

Extra Customer Feedback Analysis Tips


The steps above outline the basic customer feedback analysis process, but any sort of data manipulation can be tricky. Here are a few extra tips for ensuring that your time and effort goes into initiatives that will affect your bottom line, instead of getting stuck in the weeds.

1. Share Insights Across Departments

Understanding the voice of the customer is essential in every department of your business, not just for C-level executives. Share your data, theories, analysis, and insights via dashboards so that everyone has access to the information they need.

2. Account for Demographic Differences

When you’re staring at numbers on a spreadsheet all day, it can be easy to get lost in the data. For that reason, it’s important to keep your customers in context and consider the obvious answers in terms of demographic differences. For example, say you’re looking at the percentage of your college-aged customer segment that either did or did not purchase swimsuits this summer.

To get a more accurate representation of this percentage, you would need to filter out the locations where people aren’t likely to be swimming during this time period because it’s too cold. You can account for similar demographic differences in terms of lifestyle, seasonality, income, etc. to make sure you’re getting data that takes these variables into account.

In addition, segmentation adds a lot of essential context to data. As you can see in the below chart, the churn rate of Premium customers due to negative customer service experiences is arguably more of a problem than the negative customer service experiences of Freemium customers.

segmentation of users

3. Look at Root Problems to Understand Small Complaints

If you’re receiving a lot of niche complaints about a specific product, it’s worth taking a deep dive into that product to see if it needs to be overhauled. Though the goal of customer feedback analysis is typically to identify overarching trends and themes, sometimes a variety of small problems can point to a deeper issue.

For example, if multiple customers leave reviews about a product all complaining about different things, such as “I couldn’t get the export function to work” or “My computer crashes every time I open this,” you may investigate the product and learn that the onboarding manual file is broken. Customers weren’t able to learn how to install the software properly in the first place, leading to the variety of complaints. Remember, you don’t know what you don’t know.

4. Prioritize Your Most Valuable Customers

Though the ultimate objective of customer feedback analysis is to make changes to your business that improve the customer experience, change is hard and for businesses, often expensive. Especially when making large overhauls, it’s important to prioritize changes that are going to positively affect your most important customers first.

This could be a specific customer segment if you’re a large company, but in general, existing customers are more important than new customers. This is because earning a new customer can cost up to 25x more than keeping an existing customer, so reducing churn by focusing on these specific customer needs is much more important than trying to appeal to new buyers.

Take a look at our article on the top five things we’ve learned about successful customer feedback analysis for more important tips.

Chattermill helps companies gain actionable insights from their customer feedback analysis through AI-backed sentiment tracking and text analytics software. Contact us for a free product demo to see if our solution could be the right fit for you.

See Unified Customer Intelligence in action

See the customer reality with Unified Customer Intelligence.