Why Every Startup Should Conduct User Testing

Imagine working insane hours on your product for months, only to see some people signup for the free trial and never upgrade to paid.

Surely, the user is the problem! You must not be targeting the right user persona?

Well, maybe. But, maybe the product experience just falls short of expectations?

Maybe the person on the other end of that onboarding flow is stuck? Maybe they can't figure out how to use your product? How do you account for that?

Most early stage startups don't. They operate blindly and hope that someone will report an issue or refer to the FAQ when they can't figure something out. After all, who has time for support when you are a solo founder or a small team?

If you are nodding your head, you are probably thinking about this all backwards.

Let me explain...

User testing is a power-up move that every startup should pursue as they start to gain traction. The only exception is if your product is so basic, that it really doesn't need much explaining. A text box is a text box and a button is a button. You type into a text box and press a button - there's not much to it.

But, let's assume your app is not that simple. Maybe it has different menu areas, reporting screens, settings, configurations, integrations - you get the picture. It may seem obvious to you as the founder or daily user of your own product, but others may not have the same confidence or experience. This is where user testing comes in.

The reason I call user testing a power-up, is because it shortens the time it takes to get the product experience right. Every churned user is an opportunity cost. Even though user testing may not be exactly cheap on a startup or indie budget, it is worth every penny if it saves you from churned customers. Plus, a churned user can mean a lost positive word-of-mouth opportunity, and at worst - a bad mention online that creates further lost user acquisition opportunities.

What is the difference between QA and user testing?

The other day I asked on Twitter if people are doing user testing on their products. A few people replied. One person discussed in some detail how they have a dedicated individual on their team who performs quality assurance on the product to catch 99% of the bugs.

I went on to explain that user testing is not QA, but I'm unsure if my point was received.

I think developers may confuse QA and user testing because they are in that debugging mode and may think that user testing refers to the ability to use the product without bugs.

This is not what user testing is about.

User testing aims to understand how people navigate your product and try to accomplish whatever it is that the product promises. You can have a bug-free user experience and still have a poorly designed product.

User testing aims at reducing frustration and making it as easy as possible for people to accomplish their goals.

For example: you could have a perfectly functional app that lets someone design a social media banner. QA would never catch this because it's bug-free. But, what if the user is frustrated that they do not have a starting point? Maybe they need a template? This is where user testing makes it apparent that something is missing.

Ok, but what if you have a starting point - this must mean the app has great UX, right? We'd need to ask the user to see if the starting point is easy enough to customize to publish the image - then we'd know if the UX achieves what it set out to do.

How do I recruit testers for my product?

If you understand opportunity cost, then you'll quickly realize that user testing is actually a cheap way to do yourself a huge favor. So, how do you then find users to test your product? There are several ways to find and recruit testers:

#1: Hang out in communities where your potential customers spend time and try to talk them up into testing your product (with no strings attach, no sales pitches).

This is a great strategy for scrappy founders. Find communities where your potential customers hang out and talk to them. These may include Reddit subs, Slack channels, Facebook groups, online communities.

The key is to find actual prospects who are likely candidates for your product and not some random person online. This will guarantee better feedback because the people are much more likely to be familiar with the industry.

Some ways to recruit people:

1. Offer value to the community and get to know people by establishing good rapport.
2. Offer a service for a service - if you really need to, you can offer a service in exchange for a user test.
3. Offer to pay for their time - there is a good chance they won't even want your money, but it's a good gesture out of respect for their time.

#2 Use Facebook ads to target by interests

If you are savvy enough to run Facebook ads, you can target audiences to recruit them based on their interests.

1. Set up a Tripetto form to collect their information.
2. Make it conditional so if they do not qualify for a question, they cannot proceed.
3. Collect the qualified person's information and contact them
4. Instruct them to leave a video review of them using y our product or website (whatever you want tested). They can use an app like Tella to make it easier.

You'll want to create your questions ahead of time. Keep them open-ended or task-driven and make sure to ask follow-up questions. For example, you can ask users to create a project, to sign up for an account, or to find a specific feature in your product. After the session, ask them to talk through their thoughts about each of the tasks and how easy / difficult it was to fulfill them.

#3 Not recommended, but you can ask your close network to test for you

All User Testing professionals will advise you against asking people you know to test your product. This is because the results you get will be skewed no matter what you say to that individual. People do not want to hurt other people's feelings and even if they speak the truth, it'll be sugarcoated one way or another.

However, if you have absolutely no way to test your product with an outside perspective, you can use this strategy to a limited degree to get at least an initial overview of the problem areas. An actual user or potential of user of your product would be the best bet, so this is the last resort until you have enough resources to get qualified testers into your product.