One of the best ways to use user testing to test particular aspects of features or even an entire feature itself.
It's often not necessary to wait until you have designed an entire flow. The trick is to understand what it is you're trying to gain insight for or validate.
What do you do when you get stuck on designing a particular feature?
There are many ways to approach this and get over the hurdle. My natural instinct has always to go back to the problem and better understand the constraints or the problem. Another strategy, especially if you're short on time or if you simply just want to figure out the best combination of elements, you can simply test.
You have to be careful that you don't end up doing A/B/n tests on a small number of participants. User testing is really not the ideal situation to try and validate on some quantitative level whether or not to go with a particular design.
A far more effective approach is to use some combination of design studio with pseudo A/B testing backed by qualitative probing. What do I mean by this?
Don't simply ask your potential users, "Do you like A or B?" and then tally up the answers and go with the one with the most votes. DON'T DO THIS.
This can be an effective method when working with fellow designers/team members who may already be familiar with good design patterns and understand why you are approaching the design the way you are. (This is essentially dot voting.)
With users who may be using the interface, you have to understand that YOU the designer know more than the user. You have been thinking about this design challenge. You have interviewed far more users than any user of your product has ever done.
My approach focuses on framing small design challenges, getting users to express potential solutions, and then presenting them with my solutions.
This is often a fun method because the solutions I come up with in advance mirror what the users suggest - which makes me seem like a wizard. It's also a helpful method because it shows that I'm on a similar train of thought as my users.
However, the true value of this approach is that users can quickly point out the flaws in your design against the backdrop of their own recommendations. This helps the designer understand why the user is thinking this way. Most importantly, as you near the deadline of a project, it is helpful to begin optimizing what you have rather than exploring something completely different. The combination of brainstorming, and pseudo A/B testing with qualitative probing reveals new directions and insights within predefined constraints.
The goal here is not to receive design direction from your users. The goal is to understand how they're interpreting and responding to your design so that you can resolve underlying issues and pain points that surface when your user meets your interface. Rather than taking your user's suggestions point blank, you then need to reflect on the learnings and decide how to iterate and modify your design.
If you're interested in learning how this method might work for you, leave a comment below!