Andria Tay has focused her career on the convergence of entertainment, technology and ecommerce. She contributed to the success of the iconic MTV cable channel in the 90’s, and led the team at EMI that launched Norah Jones, whose first album was certified Diamond in the US and sold nearly 30 million worldwide. She went on to lead teams at Zazzle.com, MediaZone, Akira, and Walmart.com before her current role leading communications for Narvar, the post-purchase customer experience platform.
Narvar helps more than 800 brands such as Sephora, Patagonia, Levi’s, Bose, Warby Parker, Home Depot, LVMH, and L’Oréal ensure every touchpoint along the purchase journey engages consumers and enables emotional connections—from pre-purchase to in-store experiences and beyond. Under Andria’s direction, Narvar has committed itself to thought leadership. This has helped them create the post-purchase customer experience category, and enabled them to speak to C-suite executives in a way that makes Narvar a strategic partner for their clients.
I recently spoke with Andria to understand the process behind their recent State of Returns study, as well as how she approaches research in general. Below is a summary of our conversation.
Tell us about what you were aiming to achieve with this research
I believe the purpose of content in general is to pave the way for the conversations we want to have in the future. We look 6-12 months down the line, think about where we want to be as a company, and figure out how to lay the foundation for those conversations. Leveraging data is an important component of that strategy.
High value thought leadership is provocative, providing a fresh perspective and recommendations on how to guide strategy moving forward. You can use third party data to bolster your point of view, but there’s a lot of value in gathering proprietary data. If you do it right, it creates additional credibility and authority.
Research helps us better understand the market we’re serving. As an example, returns are a huge part of what Narvar does, which not everybody realizes. Some think of us as the package tracking people, but reverse logistics is just as important, if not more so. Returns continue to be a pain point for both retailers and consumers, though it’s a growing part of the experience and retailers are trying to find better ways to manage the complexity.
In order to do that, retailers need to understand consumer attitudes and consider how they balance those with what makes business sense. We wanted to talk to our retailers and say, “Here’s what we’re asking consumers around returns beyond just, ‘Make it free.'” There’s no such thing as free. That cost has to be covered somewhere. So what are some of those other things that consumers care about, and how does it really impact their overall buying behavior? That’s where this research started.
The most challenging part of producing valuable research is trying to be objective yet get to the information you really want. You obviously want to corral what you’re asking, so that you get the kinds of information you’re looking for, but not to lead people. It’s this delicate balance of how you construct your questions and your questionnaire. You have to take a step back and think, “Okay, yes, I understand the space that we are working in, and what we want to get out of it, but at the same time, I want to be an objective reporter.”
When you think about other reports you’ve seen from companies, in some cases you think, “This is a thinly veiled sales pitch,” versus, “This is real authentic information that will help us all make better decisions.” Obviously you want to be on the latter side.
Let’s talk about your question development process. You mentioned that there’s a lot of mystery around returns for the retail industry. How did you identify that?
We’ve done this survey for four years now. The line of questioning has shifted a little bit over time, but there are certain things like “bracketing”, which we suspected was happening up front and we continue to track every year since we were the first to define and quantify it.
We personally know that a lot of us in our own shopping do this, where you end up buying multiple things, whether it’s different sizes, or colors, or styles to try on at home, and then you end up sending most of them back, if not all of them. Ufortunately, that’s just an artifact of e-commerce shopping.
What we wanted to do though was quantify it, because we couldn’t find that out in the market anywhere. So we asked that in our very first survey and were surprised at the scale of it. Even in that first survey, something like 40% of people,said that they bracketed at least occasionally, and last year it jumped up to over 60% since people were shopping more online.
Being able to track bracketing over time has become something Narvar can hang our hat on. But, obviously, things change, so you adapt sections of your survey to explore different aspects of the issue. Also, sometimes you’ll find in a survey that what you think is going to be really interesting, isn’t. Or there isn’t anything sort of statistically significant, so you move on.
With this last one, obviously with the pandemic, that’s where a lot of new areas of inquiry came in. We wanted to know things like, “Have people’s behaviors changed in these key areas, like are they using more alternative locations? Are they more willing to pay for certain conveniences?”
Who were your closest collaborators on the project? What roles did you find were most helpful in working with to develop these questions?
I’m lucky in that I work with a fantastic team, and a lot of the folks at Narvar have deep retail industry knowledge. One of the people that I work with pretty regularly is our Head of Customer Success. He comes from a very different perspective, and of course, he’s very close to our customers. He sees a lot of the trends happening in the market, and can share what our customers are asking about.
Another close collaborator is one of our lead designers. She’s been here at Narvar even longer than I have. I work through a lot with her, since she also comes with a knowledgeable perspective and helped with visualizing the data for our first few reports Her role has evolved over the past few years to be more on the innovation side, so I tend to involve her early in the strategic process, and work with another of our fantastic designers on the data viz, which is a critically important part of the equation.
More than 1,000 people took your survey. How did you get that many participants?
On the actual survey execution and recruitment side, we use SurveyMonkey. It’s just easier. You can buy an audience through them and pay per respondent, with cost being dependent on your parameters. They have add-ons, so if you add a qualifying question, it costs a little bit more. And because you know that that qualifying question may disqualify 50% of the people they invite, you do have to factor that in as well to be sure you’re getting the sample size you want.
For example, we’ll specify, “We want an audience of people who shop online for personal purchases as a baseline. And we want them between certain ages to corral it a little bit.” And then we ask usually a qualifying question, especially when it comes to returns, because not everybody has tried to return an online purchase. And we don’t necessarily want to pay for people who have no experience or limited experience.
So we usually ask, “Have you returned an item that you purchased online within the last twelve months or six months?” We’ve gotten down to six months, because we want that recency of experience. That helps us disqualify at least a chunk of folks who are not going to be able to give us meaningful answers.
SurveyMonkey has gotten really fast with this process. When we first did this, it took a week or two to get answers. Now, we usually get them within three to five days.
How did you create the content once you got all the raw data back?
The content strategy of it starts before the survey,visualizing what you are aiming for. What’s your ideal headline? What are the key insights you’re hoping to yield? You may or may not get it out of the survey, but once I get the results back, those are the first things I look for. Does this data support these hypotheses or not?
Sometimes counterintuitive results are okay. It’s like, “Okay, we assumed it was going to be like this, but actually, this is what people care about.” And that helps you also craft the content that comes out of it.
Then you start slicing and dicing. There are times when, when you look at something just in aggregate, nothing really stands out.
But when you start looking at, “Okay, is it different for millennials versus everybody else? Is it different for people whose last return was to Amazon, versus anybody else?” And that is where a lot of these different, more interesting insights surface. And that can trigger additional content ideas. Like, “Oh, I didn’t realize that people who returned to Amazon are perceiving this as a better experience, even though it isn’t.”
With one of our previous studies about communications between retailers and consumers, it was really interesting to see the percentages of people who were Gen X or Boomers and had absolutely adopted chat and some of these other new channels, but they all wanted email on top of it. Millennials want a little less. So there’s these nuances that come out of it, and help you tell a story.
To me, it’s very much a situation where you put your reporter hat on and see what’s interesting.
Can you give me an example of something you found interesting?
In this last study, one of the things that we asked that was new, was not just about some of the specific services that people had used, but would you be willing to pay for it? Because again, we know there’s no such thing as free returns. It’s a little bit tone deaf to tell retailers, ‘Just make them free, that will solve your problems.’ That’s not simplifying returns, and it’s not sustainable from a business perspective. So again, thinking about how we can dig that level deeper into what could be more valuable to retailers. And what are we hearing from consumers that they may be open to, is where we ask them these questions. So it wasn’t just, “Have you used curbside pickup or drop-off of returns? Have you used a kiosk or locker?” But also “would you only use it for free? Would you be willing to pay for it as part of a subscription? Would you be willing to pay up to $5 or over $5?”
And it was surprising how many people said, “Yeah, we’re willing to pay to get somebody to come and do a scheduled pickup of a return from my home.” And it may be that you’ll still see 50% of people say, “Yeah, I like it, but I only want to use it for free.” But it’s those nuances of how people answered these, and what they prioritize. And we found pretty clearly that they’re willing to pay for convenience and predictability.
Findings like this are what helps inform how we look at our product and services and also how we advise our retailers.
How did you think about distribution for the report once it was published to maximize media coverage? Did you talk to journalists and create a PR pitch around it?
Absolutely. That’s actually one of the best ways that we’ve used our research, because we don’t do a lot of paid demand gen, really. We’re really relying on organic traffic, whether it’s through organic social, PR, using our internal teams to amplify.
We usually gate our full report as the primary asset. We create other blog content and infographics to help provide a preview that drives people to this asset as well. We publicize it through newsletters and emails.
PR is probably the biggest thing. For that, we usually end up teasing out multiple angles or storylines out of the data.
Honestly, most of the time there’s too much data to work with. We end up having to not use half of the stuff that we get out of the survey. If we had more time and bandwidth, we’d go back and dig into some of those things a little more. There is so much there. It’s a really rich source of information. It fuels everything. For example, when we were doing events, it was helping to catalyze what we’d speak about. It feeds into so many pieces of our strategy.
Any recommendations for someone thinking of using surveys in their content program?
Definitely keep it small to start, and just do it. Research is a heavy lift, but you don’t have to make it unrealistic. Don’t try to answer all the questions at once. Think about it as, “You know what? Let’s try this in a manageable chunk.”
But don’t skimp on the audience side of it, because that can really skew your results. So still try to get to statistical significance with a decent size audience, but maybe chunk it down into, “We really want to answer these three questions” so that you can get something that’s usable.
You also don’t have to do all this yourself. In the past there have been cases where we decided to hand this to an outside company that manages this whole process. This is especially true when we needed research in multiple countries and languages.
It sounds like the real power here is using this across the company – everything from trade shows to product and service development. This needs to be a strategic initiative for the company, and not just a marketing project.
Exactly.
—
Download the State of Returns research here, and connect with Andria here on LinkedIn.
Watch Andria speak more about this topic to the Bay Area Content Marketing Meetup last July.
