As a first pass, I like to talk to a few friends who I know will be brutally honest with me, and who fit the target user persona.

This is a great way to spot major mistakes in your thinking early on, particularly if you yourself are new to the market/industry.

How you proceed from there depends entirely on the type of product you're building.

For example, if you're targeting students, you can get excellent feedback on the cheap from online surveying websites that are heavily biased towards that demographic. If you're targeting niche specialists, you'll have to do away with a much smaller number of respondents, organizing long-form calls with a few industry experts. And if you're targeting a technically illiterate community in a remote Zambian village, you'd spend most of your time talking to organizations that have worked on projects in that location, and eventually observing the local community on the ground.

Each approach will of course come with its set of biases, and no amount of research will replace a small-scale trial or MVP.

The right people to talk to depends on the type of product or problem that you're currently solving.

Before conducting a user interview, it is helpful to do a user screening process where initial questions serve as filters for the type of respondent that you want to talk to. You can, for example, filter for demographics, role in the company, usage frequency of a particular product/service, brand awareness, etc.

Create a screener (check out the linked resource). Keep in mind that when potential respondents pass your screener, you need to do scheduling and usually offer incentives as a token for their time and insights.

As for where to find the respondents, there are several companies who offer respondent panels (e.g. usertesting.com, alphahq.com, etc.), but keep in mind that most of these companies are tailored more for B2C business model.

For B2B respondents, it would help if you already have a network of respondents and use snowball methodology to look for the next potential respondent. Costing would depend more on the incidence rate of respondents. The lower the incidence rate, the higher the cost — this only means that it's harder to find the specific respondent that you are looking for.

There are also different sampling methodologies available depending on how representative you want your sample to be and the type of study that you are presently conducting.

For face to face feedback chat to those who are articulate and understand the problem space well. Bear in mind that you also need someone who is willing to give you their time.

I usually start with surveying my networks, and go from there. Remember, it's possible that people you know may lie to you! Why? They don't want to hurt your feelings.

You might be better off installing a third party tool such as FullStory and trying to rope in a few users from public slack channels or twitter.

Just go out on the street and talk to people! You might have heard this extrovert advice before. It’s a great way to learn to deal with rejection for sure. But, it's not a time-efficient strategy to investigate a specific, product-related problem.

In product research you will need to adopt a targeted approach. Sure, it’s easy to “reach out to people around you”, but will you trust your PT’s take on a drag-and-drop restaurant menu builder? Probably not.

So, who is your target person?
Your end user. They are the one who are most likely to encounter the problem your product is solving. You already have a user persona built so you know the answer to that, riiiiight?

Where are they likely to be hanging out?

Think of your users’ online and offline activities. Are you most likely to encounter them Insta, Reddit, or in the local church?

Who do you already know?

Being the entrepreneurial busy bee that you are you are likely to already be connected to people from the same industry (LinkedIn anyone?). They don't need to be your end users, but simply more experienced people whose input you would trust.

But wait, do you always need to talk to people?

For quick tests, you can refer to user recruitment agencies, or use platforms like UsabilityHub, UserTesting. Depending on the service will help distribute surveys in the required demographic, and/or provide you with screen and voice recordings of how users interact with your product or landing page.

Don’t forget to buy them coffee.

In contrast to your fav triple-chocolate brownies, human beings are tough cookies. But, they like both brownies and cookies. If you have problems recruiting enough testers, offer them something in return. Amazon vouchers, free access to your product, a trip to the zoo together, whatever you feel would work.