Insider Tips from Shutterstock: Turning Subjective User Feedback into Objective Insights

with Sarah MacDonald, Senior Product Manager

Hey everyone! Francis, here ✌️

This week's article marks an exciting first for the Product Pantry newsletter. I'm thrilled to introduce Sarah MacDonald, Senior Product Manager at Shutterstock, in our first-ever interview with a Product Manager.

If you're not familiar with Shutterstock, they're a stock photography, video, and music company that empowers creative professionals around the world to tell their stories. Sarah has a wealth of experience in building recommendation AI products, and at Shutterstock, she works on Shutterstock’s discovery and recommendation features.

In this article, Sarah will walk us through her workflow to surface customer insights, how she filters noise and finally, how she prioritizes the different types of insights to make sense of it all. User feedback is subjective, and Sarah's workflow is great at turning this subjective feedback into objective insights she can use to improve her product.

Sarah, we're eager to learn from you and see how to make our processes more objective and effective.

Let's get started!

---

Hey there, everyone! Excited to be part of this newsletter.

Today, I'll share a topic near and dear to my heart: how to surface customer insights to inform your product decision.

One of the most crucial aspects of building great products is understanding the needs of your customers. Let's dive in.

1. Look at Data

First things first, I start by looking at the data. This always proves what's going on and where user flows are breaking down for whatever reason.

I keep an up-to-date map of user journeys I care about and log baseline metrics of these flows. This way, I can watch and react promptly to changes. This also helps me find opportunities, places where the funnel constricts, etc.

If I see an area for improvement, that's where I will focus on the following steps. It's like the saying goes:

"Data doesn't lie; people do."

2. Look for Feedback

While data is important, it doesn't always tell the whole story.

That's why once I've identified areas of improvement from the data, I take a more hands-on approach to gather tangible insights on what's causing the issues.

I'll often walk through the user journey myself and ask coworkers to do the same, taking note of any friction points or confusing parts.

In addition to this, I'll also turn to our customer service crew to see if there have been any conversations with customers about the issue I'm researching.

I'll also check out the answers to the NPS survey we have for more insights. The NPS survey is offered to customers via a "Send Feedback" link after they conduct a search. The survey prompts customers to provide feedback on whether their experience was positive or not, what they were looking for, and suggestions for improvement. I'm lucky to have a lot of active users, so I always have stuff in there that may corroborate my findings or provide additional insights into why a particular part of the flow isn't working as intended.

Once I have all of this information, I use it to form a hypothesis about what's going wrong.

It's kind of like putting together a puzzle - each piece adds to the picture until you can see the whole thing. By combining metrics data and customer insights, I can get a clearer understanding of what's happening and how to improve the product.

3. Speak to Actual Users

Then, you just gotta go straight to the source. I also like to plan interviews and speak to actual users.

1. Set Goals for Your Interviews

This is where researching data beforehand becomes super handy. It helps to formulate a hypothesis before writing questions for the customer interview. By clearly understanding what you want to achieve, you can design interview questions tailored to your specific needs.

For example, you looked at the data about how people sign up for your service and noticed that many of them are leaving when they get to the part where they have to choose how their order will be shipped. After testing it yourself, you realize that all the fast shipping options are really expensive, and you assume that's probably why people are leaving. Your hypothesis is that customers want shipping that is both fast and affordable, but that's not an option right now. So, you'll ask questions about that to get more information.

2. Ask Users to Show You How They Do Things

But don't just take their word for it. When interviewing customers, ask users to show you how they do things.

People tend to lie to look good or forget things, so to better understand their actual behaviour, I find it's more beneficial to ask them to show me how they do things instead of just relying on their self-reporting.

Have you heard the story about the measuring cup company? They asked people how to measure a cup of water, and no one mentioned looking at it from eye level. But when they actually did the task, they all lowered themselves to eye level. It just goes to show that sometimes people don't realize what they do until they're actually doing it!

Here’s my simple process for validating a problem through user interviews:

• Ask the user if they use that portion of the site and how often.

• Ask how often they use the platform in general.

• Ask them to share their screen and show you how they use the site to complete a simple task (e.g., find an asset for x, y, or z).

• Ask them to talk through their process and record what happens.

• Point out or ask about anything they did that was unexpected or took them a long time.

• Boil down those points into themes in a Miro table. These themes may change, but examples of themes could be "Friction when checking out" or "Search results not relevant to query" - something that other users may be experiencing.

• Add the feedback on sticky notes and organize it on an Impact X Effort scale, which can be shown in the Miro link above.

So, ask them to show you what they're doing. It's like getting a backstage pass to the user's world.

3. Ask the Key Questions

I always like to ask the key question: "What is your favourite thing about website/product/process/service]" or "What is your least favourite thing about …?"

When you ask a user to identify their favourite or least favourite thing, it forces them to think critically and express their true feelings about the topic. It's like taking the user's temperature.

4. Record and Analyze

After collecting all the feedback and data, the next step is to document and analyze it.

If I require quick feedback on an existing feature, I typically approach my colleagues who are users themselves. It's easier for me to work with them as I have several colleagues to choose from, each with different job roles.

In some cases, I even ask them to demonstrate how they would accomplish the same task on a competitor's website.

If it's a significant new change, which is rare, I turn to the UX team. They use UserZoom to gather feedback. We then organize the feedback in the same manner as when working with my colleagues. It allows me to collect user feedback and insights and organize it.

Alternatively, I talk to users directly and record the conversation, which I then map on Miro, a collaboration and whiteboarding platform. Miro helps me visualize the feedback, which makes it easier to analyze and identify patterns.

To make the process even more efficient, I plan to use ChatGPT, to summarize the feedback from my next recording. I will give it the transcript, and it will provide me with a summary, making it easier for me to log the results.

Sharing drives 90% of our newsletter growth. If you enjoyed this, please share it!

That’s it, folks

And there you have it, folks! A step-by-step guide on how to collect and analyze data to build better products.

A big shoutout to Sarah for sharing her expertise with us.

Remember, understanding your customers' needs is key to building great products. By using a combination of data, tangible insights, and direct user feedback, you can gain valuable insights to inform your product roadmap and make your customers happy.

Until next week, keep on building great products!

Francis