.jpg)
We spoke with Marley Dizney Swanson, Senior UX Researcher at MPB to learn how they use feedback and customer insights as the foundation to drive tangible business results.
Key takeaways:
- Marley's fundamental rule: always set a benchmark before starting any new work - understand exactly what metric you're targeting and your current baseline.
- Using this philosophy, at MPB Marley combines qualitative customer feedback with specific business metrics (NPS, conversion, revenue, GMV), ensuring teams know with confidence whether their work genuinely impacts both CX and business results.
- By quantifying qualitative feedback, Marley is able to build compelling business cases executives can't ignore - showing the true impact of customer issues.
- This approach led to a new requirement: every product decision at MPB now requires evidence linking to specific business metrics - whether it's return rates, GMV, or conversion - ensuring resources go toward problems that actually affect the bottom line.
A conversation with Marley Dizney Swanson, Senior UX Researcher at MPB
Q: Can you tell us what being a mixed methods researcher means, and how it plays out in your work at MPB?
Marley: Being a mixed-methods researcher essentially means I'm a storyteller with multiple ways to tell that story. At MPB, we leverage extensive quantitative data to understand how customers behave on our platform, tracking key metrics like revenue and GMV.
I then enrich this data with qualitative insights from interviews, usability testing, and NPS survey results that come through Chattermill. By triangulating these different data points, we can confidently determine whether our work is truly making an impact on our customers' experience.
Q: How big is the team, and how are you working with the wider product organization?
Marley: Our product design team is about six people, and I'm the sole researcher supporting an organization of over 500. To address this scale, we've invested heavily in empowering team members to conduct their own research.
We have six scrum teams with product managers, engineers, designers, copywriters, and visual designers. Through monthly seminars, I teach them essential research skills - conducting interviews, designing surveys, and other techniques - so they can handle their own research while I focus on synthesizing insights across projects and tackling strategic initiatives.
Q: As the only researcher, how do you handle all the questions coming from the business and product teams?
Marley: We do have a backlog of things that we've just not been able to answer yet. But one way we prioritize the research requests coming in is by using hypothesis statements.
We'll say, "Unlocking this information would allow me to do X, and we predict Y outcome." This framework allows us to assess whether the issue is urgent, whether it impacts a really big metric, or if it's something that will unlock future work. That way, we're able to put our efforts where they would be best served in that moment.
The challenge: blind spots in customer feedback
Q: What are the biggest challenges you're seeing in product and UX at the moment?
Marley: I think the biggest problem we have right now is that our data lives in a lot of different places. We probably use 10 tools to collect customer feedback, which isn't ideal because it means a lot of our time is spent going from source to source trying to triangulate that data.
So we're working with Chattermill to consolidate everything in one place. It means that our NPS survey results, customer service tickets, Trustpilot reviews - all of that will be in one place. We'll have a more automated process of analyzing that feedback, which saves us time and ultimately means we can get to the actual work a lot faster.
Q: What made you consider expanding your use of Chattermill to not only store NPS results, but also gather feedback from across the user journey?
Marley: The instigation for this project was discovering that our NPS survey results, which are what we have in Chattermill right now, only represent 6% of our customer base. And I'm being very intentional about that language - it represents 6% of our customers, but not our wider user base.
We only send out NPS surveys to people who have successfully transacted with us, so we're losing a huge amount of data. We're not capturing feedback from our wider user and potential customer base.
So we wanted to work with Chattermill to get everything in one place to understand how we could convert those non-customers into customers. From a UX perspective, we wanted to understand if there were particular funnels they were dropping out of, or usability issues we could address to help them achieve their goal of buying a camera.
It's a tricky measurement because when we looked at competitor numbers, it seems like e-commerce can typically expect about 10 to 20% of their customers to give feedback.
Q: What role does customer feedback play in your day-to-day decision-making and UX strategy?
Marley: It's huge - it shapes everything. We use customer insights regularly through voice of the customer forums and reports to keep them at the forefront of decision-makers' minds.
When our executive team conducts its annual planning, they have customer feedback in the back of their minds, whether it's specific issues customers have raised or metrics that are trending negatively. This enables them to create roadmaps that directly address customer needs.
It's an ongoing process - we're never going to get it exactly right - but the fewer longstanding problems you have, the happier your customers will be.
Q: From what we hear from other Chattermill users, it's not easy to ensure the voice of the customer stays top of mind for leadership and shapes company strategy.
Marley: It's really difficult. We've recently started an initiative we're calling "cite your sources." We're challenging everyone across the business to explain the evidence behind their decisions and proposals.
I think this is really important for building trust across an organization - knowing that due diligence has been done and understanding where decisions come from. Customer feedback plays a huge role in that.
When I propose things for the roadmap or to prioritize work, I'll bring a body of evidence showing that customers have been saying this in interviews, we've seen this behavior on the platform, and for these reasons, we believe we should take action.
We're now sharing this with other teams, challenging them to do the same, so that everyone across the company is confident that we're doing what's best for both the customer and the business.
Using Chattermill day-to-day
Q: Tell me about your experience with Chattermill. How did you start using it, and what were your first impressions?
Marley: Chattermill was being used by my organization before I joined about three years ago. I remember going in and being overwhelmed by the amount of information - I wanted to read every single piece of feedback. It was the best part of my onboarding experience - seeing what customers were saying about the organization helped me frame my first 90 days.
I spent a good two full days of my onboarding just on Chattermill - going through feedback, learning how to create reports, making my own dashboards of things that stood out to me, and flagging those up to my manager and team saying, "Hey, are we aware that this is happening?"
Q: How does Chattermill fit into your day-to-day workflow today?
Marley: My day-to-day workflow focuses on the top drivers of negative feedback - that's probably my number one report.
I consider it part of my job to raise any key customer concerns to our team to ensure our core journeys are being taken care of. I'm also using Chattermill to train our product managers on how to read qualitative feedback and use it to support their own decisions.
A lot of it involves teaching people how to use reports, how to identify trends in qualitative feedback, and figuring out how to share Chattermill data, reports, and dashboards to tell a story. Ultimately, we're trying to convince people to believe in what we believe in, and Chattermill has been a highly effective tool for doing so.
Q: When working on specific projects or product areas, are you looking to answer specific questions or solve particular business problems with customer feedback?
Marley: Yes, and there are two ways we use Chattermill. One is to answer a specific question, and one is for more generative research.
When we're answering a question, it'll be something like, "We released a piece of work four weeks ago. Have our customers noticed? Has it made a difference?"
We'll use a lot of quantitative data to support that, but then it's really important to get that qualitative piece to understand the why. Perhaps a metric has shifted, but can we be certain it's actually due to our actions? The qualitative piece helps answer that question for us.
Chattermill has also been so valuable in showing us where to focus our efforts. For example, we have these voice of the customer forums once a quarter, and we recently learned through Chattermill that our top driver of negative feedback is customers receiving defective items.
We gathered a significant amount of information from Chattermill, brought together 40 people and decision-makers to understand how we could address this problem, and reviewed all the feedback collectively.
It's truly remarkable to observe how different individuals interpret the data in distinct ways, leading to varied insights and solutions. We've now got a list of about 30 different action items we want to take based on that data from Chattermill.
Q: Taking action is often the most difficult step. How does that work at MPB?
Marley: The best way I've learned is to just be the squeaky wheel.
I'll give you an example: when I first came, the top driver of negative feedback was that people trying to sell cameras to MPB didn't know what accessories to include. Every week I would come to my team and say, "This is still the top driver of negative feedback." Eventually, it starts to wear on them.
So, it's a little bit about being persistent and a little bit about getting your key decision-maker to care about it.
One of the best ways you can do this is to sit down with them and ask, "What metrics are important to you?" In that example, the metrics were sales conversion, returns, and missing accessories. Eventually, you slowly persuade people around the business until you have a team and can push that work forward together.
It's really hard to get work done alone. So I would say the biggest part of getting that action is building relationships with the right people.
Connecting feedback to business outcomes
Q: Speaking of metrics, which ones are you looking to improve by acting on customer feedback?
Marley: We've got quite a few. The top drivers of negative feedback are the ones I check the most. I'm particularly passionate about return rates right now because they have numerous implications – for customer satisfaction, for our GMV, and for the amount of work that coworkers in our warehouse have to do.
Then, we're doing a lot of work with conversion rate optimization, focusing on buy conversions, adding to the basket, and conducting A/B testing. Those are probably my top three points of focus at the moment.
Q: How do you share insights with the wider business and ensure people know what customers have top of mind?
Marley: There are a couple of different ways I do this. One is that I literally put them up on the wall in the office. My desk has become what I'm calling the "customer corner" - above where I sit, we have personas and different driver trees for what customers are thinking, doing, and saying at any point during their journey.
The second way is through our customer insights channel, where I'll share all my voice of the customer reports and any interviews I've conducted that I think people will find interesting.
And maybe the more informal one is that I'll just attend different teams' meetings. I've got relationships with different teams, and maybe once a quarter, I'll check in and say, "Hey, we haven't heard from you in a while. What are the top things facing your team right now that we might be able to help with?"
That gives me an invitation to come in and say, "Well, here's what I'm hearing from customers. Why don't we help each other help our customers?" Again, it's that relationship building that seems to make a big difference for me.
Q: When building reports and dashboards, how do you approach understanding the customer experience?
Marley: I start with an overarching metric like NPS, then break it down to find pain points. For example: NPS by market shows UK customers are happier than France. In France, logistics drives the most negative mentions. Looking at the last 30 days of feedback reveals specific issues - unhappy with courier, delayed deliveries, packages left unattended. Those three concrete actions go straight to my operations team.
The more granular the report, the easier it is to take action. Then I connect it back to business impact - this is dragging NPS down by five to ten points, which translates to X revenue loss. Speaking both the customer and business language gets decision-makers to care.
Q: What would you say is the biggest benefit of working with the Chattermill platform?
Marley: Chattermill automates so much of the work I had to do previously - creating graphs to show how specific work has moved metrics by specific amounts. All of that is done for you, and the UI is significantly better than anything I've ever created on my Tableau dashboards.
It makes it really easy to break down customer data and quantify qualitative data, which is a challenging skill. It allows you to create really clean metrics that show concrete point increases, instead of providing more generic statements about people being upset about something. That number seems to resonate with people a little bit more than just the pure qualitative side.
Examples of feedback driving real changes
Q: Let's dive into some examples you found in Chattermill that really made a difference.
Marley: First example: When I joined, the top driver of negative feedback was sellers not knowing what accessories to include with their gear. I hammered this home week after week using Chattermill data.
Any successful business case needs the what, why, and how big - Chattermill helps us uncover the what and why, then we quantify it using tools like Tableau or Fullstory to show the problem's size.
When we showed decision-makers the actual revenue loss, they immediately sought a solution. We implemented a simple checklist of expected accessories that sellers can check off. This increased our NPS for our sales score and improved conversion rates – a win for everyone.
Second example: We partnered with Evri, a UK courier known for being cheap and fast but with a poor reputation for delivery issues. Knowing it was risky, I set up daily Chattermill alerts for any Evri mentions so we could monitor customer reactions in real-time.
If customers were really unhappy, we were prepared to roll it back to protect our reputation. The alerts helped us stay on top of the situation.
We've since iterated to be transparent about delivery options - customers can choose Evri for a lower cost or opt for other carriers at a higher cost. It's about providing both our business partners and customers with the information they need to make informed choices.
Q: What would be your advice to someone who wants to connect customer insights to the ultimate business outcomes they're looking to achieve? How should they start?
Marley: First, set a benchmark - understand what metric you're looking at and where you're at now before starting new work. Too often, companies just ship without knowing how to measure success.
Take a step back and define why you're doing the work - to reduce return rates, increase conversions, or whatever the goal may be - and understand your starting point.
Even if you don’t have tools like Fullstory or Chattermill, you can simply send an intercept survey asking customers to rate their confidence in the journey on a scale of one to ten.
Collect your baseline data, implement the new work, and then assess whether anything has changed. There are really low-fidelity ways to do this, but it's crucial to know whether your work actually affected the metrics. Otherwise, you're just taking shots in the dark, and that doesn't help make good decisions.
Getting started with UX research
Q: What's one thing you wish more product, UX, and research managers knew about using customer feedback effectively?
Marley: Don't take everything at face value. One customer's lived experience, while true for them, doesn't necessarily scale to everyone. The trap is putting efforts toward what's actually a small problem, even if it affects some customers significantly.
That's why mixed methods are important - to understand the attitudinal side through qualitative data, then back it up to ensure you're impacting the most people in the most significant way. With limited time and resources, we have to make difficult decisions about which project will create the biggest impact.
Q: What piece of advice would you give to a UX manager or researcher just starting out with Chattermill?
Marley: Click everything. Just move around the site as much as you can. So just playing around with the tool, you uncover so much and you don't need to know all of it at the beginning. You'll learn as you go.
Q: If someone was on the fence about using Chattermill, what would you tell them?
Marley: Having used Chattermill, I'm a big fan. It makes creating reports really easy and sharing findings really simple. We've recently been challenging ourselves and other teams to cite our sources, using Chattermill to show the evidence behind the decisions we've made.
This allows us to have really informed conversations with colleagues about whether we're making the right decisions or need to see things differently. Those evidence-based conversations are the best part of the job, and they ensure we're doing the best we can for our customers.
Inside the mind of a UX Researcher
Q: What would you say is your secret superpower as a UX leader?
Marley: Stakeholder management. I can make friends with anyone.
Q: What's the first thing you check in Chattermill each Monday?
Marley: Top drivers of negative feedback.
Q: Which feedback channels matter most to your team?
Marley: NPS and Fullstory for on-platform behavior.
Q: What would you say is the most underrated feedback source?
Marley: Customer support tickets.
Q: One metric you're obsessed with?
Marley: Return rates.
Q: Biggest UX myth you wish would disappear?
Marley: That some research is better than no research. That's not always true.
Q: What's your go-to way to get buy-in from leadership?
Marley: Understand the metrics they care about. Tailor your pitch accordingly.
Q: What's the best compliment a customer has given your team?
Marley: "I'll never shop with your competitors again."
Q: Favorite Chattermill feature and why?
Marley: Segments - saves me time.
Q: Any trends in product or UX you're paying close attention to in 2025?
Marley: AI generally, and specifically, how it helps UX research workflows.
Q: One thing product and UX teams should stop doing immediately?
Marley: Starting work without a clear measure of success.
Q: What's the most rewarding part of your job?
Marley: Hearing about how I help my team become faster, better, and more confident.
Q: The best piece of career advice you've ever received?
Marley: Within your first 90 days, make a driver tree for your organization.
Q: If your team had a tagline, what would it be?
Marley: Test before you guess. I have a sticker on the back of my laptop that says it, which I got from UX Insight this year, and I love it.
.png)
