When you need quantity: web-based research 

“When you want to create a product or feature that people love, your goal should be to learn as much as you can before you’ve launched it. Before you’ve even written a single line of code or designed a pixel.”
— Hiten Shah 

Customer interviews are your richest source of insight. But they’re not always your best first step. 

In some scenarios, you’ll make more progress if you start with web-based research:

Note that while it may make sense to start with web-based research, it is never a substitute for customer interviews. In a pinch, web-based research is better than no research at all; but you’ll find it next-to-impossible to build a customer-focused business without a habit of interviewing customers. 

What does web-based research look like? 

Web-based research is leveraging the internet to gather insights from customers. It’s a scalable way to gather a large volume of data in a short period of time. 

Web-based research: collecting information about customers online through surveys, customer comments, and customer conversations

The more passive forms include mining forums, reviews, blog comments, competitor testimonials, and anywhere else your customers hang out. More active forms include sending out surveys and a wide variety of on-page tactics. 

In either case, the goal is the same as customer interviews: to learn more about customers, their context, and the JTBD journey. 

Which type of web-based research you use depends on what you’re trying to do. If you’re trying to get the lay of the land, identify high-level frustrations, or explore customer verticals, mining the internet is a great place to start. If you want to dig deeper into specific pain points or motivations, and pre-qualify customers for interviews, surveys are an excellent tool. 

Mining the internet for clues  

One group of people who have turned web-based research into a highly profitable artform is copywriters. Successful copywriters, especially those who focus on conversion, rely on something called “voice of customer data” (VOC data) to craft persuasive headlines, sales pages, and web copy. 

VOC data is the exact words, phrases, and framing customers use. To find this data, you mine customer “watering holes” or anywhere customers voice their opinions online. Here are a few places you can start looking:


Pull from multiple sources to find repeated phrases, words, and framing. Make sure you collect the data in some form as well. Screenshots and copy/pasting what you find into a database are easy ways to do this. (More on that in the next section.)

Voice of Customer (VOC): the exact words, phrases, and framing customers use

However, one major caveat here is not all of these web-based customers fit the bill for your product. Internet mining delivers useful data that’s easy to gather with a bit of time and effort, but it’s data you want to take with a grain of salt. 

For a more targeted approach, you’ll want to leverage surveys. 

The humble survey and how to use it

Hiten Shah is a co-founder at FYI, started Crazy Egg and KISSmetrics, and is an advisor or investor in over 120 companies. And every time he starts something new, he assumes, “I don’t know anything, and neither do you.” 

To get over this knowledge gap, Shah starts every project with a type of survey he calls the Early Access Survey. It goes out to his newsletter subscribers and gathers information about their business, how they’re solving a problem, what’s challenging, and whether they’re available for a follow-up interview. In return, he receives valuable information about customers and pre-qualifies a segment of them for interviews. 

If you don’t have a robust email list, there are plenty of other ways to leverage surveys: 

And that’s by no means an exhaustive list. When Shah was debating features for FYI, he’d test a feature idea by adding a new button for it in the product—before building any functionality. When people clicked the button, they’d receive questions about what they thought the button would do and their motivations for clicking it. 

...keep in mind the old maxim, “garbage in, garbage out” is especially true with surveys. 

Whichever survey method fits your product hypothesis, take extra care in crafting it. Keep in mind the old maxim, “garbage in, garbage out” is especially true with surveys. Twenty thoughtful responses will be more valuable and take up less of your time than 300 junk responses. 

To optimize your surveys for quality not quantity, weigh each question carefully and tell your participant why it’s worth their time to respond. A note or subtext as simple as, “please share as much detail as possible because this helps us create a better product for you” can substantially boost response quality.  

Finding signals in all the noise

Here’s a hard truth: just because you write down 100 things people said on the internet doesn’t mean you made progress. To leverage customer data, you need to learn from it. 

Ideally, you discover every customer or potential customer has a similar problem, no obvious solution or a very suboptimal solution, and a clear outcome they’re trying to reach. 

This is ideal, but rare. 

To make sense of more confusing data sets, try tagging the pieces of information. Remember, one of the goals of customer research is to understand the customer and their context—their JTBD journey. Use tags like the following to organize that picture (revisit 2.3 for more detailed explanations and examples of each):

As you organize your research around these tags, you’ll start to see patterns. Or, just as helpfully, you’ll start to see where you don’t have patterns and need to follow up with more substantive research like customer interviews. 

Which begs the question, how on earth do you organize this information?