Wouldn’t it be great to get customer feedback before there are even customers so you know what new features and products to prioritize?
Yes, we’re talking about gathering feedback from customers who don’t yet exist, for a product that doesn’t yet exist, to create a product that will perform better, sell better, and get rave reviews.
And it’s possible.
It’s just the opposite way Product dev usually works.
The usual way: When evaluating a new product, usually you present the product, a minimum version of the product, or a beta version of the product, to a group of users (or beta testers) and listen to their feedback (qualitative data) and look at their behaviors (behavioral analytics) to see where you’ve succeeded, and where you still need to pick out the bugs. But on brand new features and products that aren’t launched, knowing that customers want and need most is educated guesswork. A series of hypotheses and trials.
We’re going to show you how to leverage qualitative data to build better hypotheses to reach successful new products and features faster.
How can you leverage qualitative data when the product or feature doesn’t exist yet?
You have to talk to customers who don’t exist yet.
When preparing to create a new product or feature, your first task is to speak with potential future customers – people who are a good fit for the solution you’re thinking of building. If you have an existing user base and are planning to introduce a new feature, you can start there by finding groups of people whom you think are likely to need it.
Your goal is to check your assumptions against their real, qualitative feedback – and there is nothing like a two-sided conversation for gaining insights you’d never expect. Schedule calls with at least a dozen people you think will be a good fit, and ask:
- What goals, inside and outside of work, are you hoping to accomplish today, this week, and this year?
- Tell me about your work process – what do you do exactly?
- What frustrates and aggravates you on a regular basis – what are the hurdles between you and getting things done?
- What might make reaching your goals easier?
Then, present your product idea and ask if they think it could help them reach their goals and reduce (or eliminate!) the hurdles.
Of course, interviewing individuals doesn’t scale. So when you do have hundreds or thousands of users to poll about a new feature to your existing product, you’ll need to gather your qualitative data a little differently.
For new features on existing products…
If you’re looking for ideas of what to build next, or what to prioritize, you’ll want to cast a wider net for feedback.
Two groups of users tend to have the most valuable insights: Existing power users and good-fit users (users who should be successful with your product) who are showing signs of imminent churn.
The people who love you and the people who *should* love you, but don’t, both have perspectives you can use to prioritize what issues to fix first, and what features to build next.
We brought in some UX experts to tell us how they use qualitative data to point them in the right directions
“I have a system for analyzing qualitative data throughout the lifetime of the product, even after we have behavioral analytics! Qualitative data gives you things that quantitative never will. And vice versa, of course. Quant tells you what is happening. Qual tells you why it’s happening.
The first step is to gather qualitative data the right way. I prefer to do fairly targeted studies of specific types of customers or potential customers. I observe behaviors and ask questions around things like why people behave the way they do, what other things they’ve tried in the past, and what problems they’re trying to solve. I then take the insights from the studies and do some affinity grouping to help me see patterns in the data.
Of course, what sorts of affinity groups I use and what sorts of research I do initially varies wildly depending on what I want to learn and which parts of the experience I need to improve. Cindy Alvarez’s book Lean Customer Development covers a lot of good techniques for this in a lot more depth.”
When gathering and analyzing data from hundreds of customers, it helps to send a survey that invites open-ended responses, and have a more sophisticated system than a spreadsheet to track and sort the answers.
For example, Ancestry.com has been gathering net promoter score (NPS) and customer satisfaction (CSAT) survey responses using Qualtrics – but that still required paying a research company a fee to provide a one-off report to analyze the free-response sections. Now, they’re inputting their data into Wootric CXInsight™, which allows auto-categorization and sorting by sentiment.
As Laura Klein says, the goal isn’t to build the next feature – it’s to build the right feature. And opening up lines of communication with your customers is the only way to make sure you’re moving in the right direction.
“Don’t build a Franken-product by shoving every cool new feature you can think of into your product. Build targeted features that fill specific needs for your actual users, and try to do it so that you’re always helping your customers reach their goals as quickly and painlessly as possible.”
How to prioritize what to build first, and what to do next with customer feedback
Different product managers have their own preferred ways of using customer feedback, but we like how Luna May Shirley at Exponea, an e-commerce growth platform, brings their customers into the new feature development process almost from the beginning.
“Create a customer advisory board and have your customers discuss the issues that they are solving using your product with each other. You will be able to get fantastic insights into new feature development and their priority.”
Having a panel of customers ‘on tap’ to discuss issues and give their perspectives is a safeguard that not enough companies use.
Laura Klein, from Users Know, adds:
“The worst thing you can possibly do is to make a big list of undifferentiated features (we’ll call it a backlog!) and then sit around in a room with a bunch of people and just rearrange it into a giant list of prioritized features (we’ll call it a roadmap!) that you then build in order and without ever reflecting on what’s working and what isn’t. This doesn’t result in good products. It results in big collections of features, most of which users hate or ignore.”
Laura bases her ‘to do’ list on behavioral analytics and user feedback that indicate room for improvement. These become her “metrics I want to improve.”
Focus on the outcomes you want and then figure out which features are most likely to help you reach those outcomes. Make the decisions based on good research – both qualitative and quantitative, if possible – and then constantly measure to see if you’re improving things.
Working backward from her metrics goals, Laura looks at what user behaviors need to change to improve her target metrics. Then she generates a list of features that have the potential to change those user behaviors. From there, she says,
“I try to understand which is most likely to be successful and least likely to have bad secondary effects (for example, I don’t want to improve acquisition only to kill conversion by acquiring a bunch of the wrong users!).
I also try to understand what assumptions are baked into my assessment of each feature. Why do I think that feature A will improve my acquisition numbers better than feature B? What evidence do I have? Is there any research I need to do to come up with better hypotheses about which features will have an impact?”
Often, the solutions she uncovers to change user behavior for the better don’t include building something new at all.
“Whatever I’m building, I try to understand the expected impact on the user’s experience and how changing that experience will improve their outcomes and the outcomes for the company. Sometimes very small changes to existing features can make an enormous difference in customer success. Fixing bugs may not always be sexy, but it can be enormously powerful in terms of improving your product.
I admit, it can be tempting to always be biased toward building new features, since we often get rewarded for making new stuff, but often users are going to be happier and more productive if you do something like speed up the product or make the existing features work the way customers expect them to!”
Everyone knows that listening to your customers is important – but equally important is listening to your customers before they become your customers. Gathering qualitative feedback is a process that should begin well before there is a product and continue throughout the build, launch, and lifetime of the product. It’s how you’ll ensure product-market fit every step of the way.
Turn customer comments into data you can use. Get advanced text analytics from Wootric.