UX Research: finding your users


Part one

In order to ensure that your UX designs meet the needs of the user, research and testing should take a highly important role in your work. If you don’t include them, you risk creating a product which hasn’t been verified, and therefore doesn’t address their needs. However, if you work on a project where you aren’t given lists of users to work with, what should you do?

I came across this scenario on a previous project I worked on – Springer Materials. Due to the way in which the company sold their product, the people who bought the product were not the same as the people who used it, and often, the two groups were not in contact. This presented a series of problems:

  • The needs of the people buying the product were different from the needs of the people using the product
  • The people buying the product often didn’t know in-depth details about the product they were buying.
  • The business listened to the buyers, but not to the users
  • The user base was made of a mix of different groups, all with slightly different needs.

From the points above, you can see it was important to ensure that we included the voice of the actual users in the conversation. However, if your project stakeholders can’t give you lists of people to interview, where do you go?

Research your user base

Part two

Firstly, you will want to get an understanding of just who your users are. By knowing who the company are selling to, you at least have an idea of the institutions where your users come from. In the case of Springer Materials, we knew that the product was being sold to academic and corporate scientific institutions. Of course, the product wasn’t going to be used in some parts of those institutions – it’s a fairly safe bet that Springer Materials wasn’t being used by the students in the Performing Arts section of the universities it was being sold to*. We began to research the departments which might well be interested in its use – materials science, chemistry, engineering and the like.

(* although it’s worth noting that this, in itself, is an assumption. Perhaps it was being used in that department as research for prop and set building? Always make sure you note down and challenge assumptions such as this in your research.)

Building your user base

Part three

In our research, we began to collect details of departments, as well as individuals, who may be able to help. Some work with a search engine can yield contact details of heads of departments, key lecturers, or their administrative assistants.

We then tried another approach – leveraging professional networks. In our case, there were many different professional associations and groups for materials scientists, so we managed to find details of their contacts online. At one point, we even tried using LinkedIn as a resource, searching for individuals and associations inside its social network. We did find that this was somewhat limited in opportunity, as being able to contact them meant having to purchase a subscription to the network.

Ready for contact

Part four

To prepare to contact them, we made a list of these contacts. This would be the centre of our efforts – providing a record of everyone we had contacted, and whether we had been successful. It also helped us ensure that we didn’t bother people who has previously declined. We would be sending out a lot of messages, inviting people to engage in user research, so it was key to draw up an email template. With some experimentation, we found some useful guidelines for this template:

  • Keep it concise: Make sure you state clearly what you’re doing, and what you’re asking from them, but don’t waffle. People are less likely to read a long email from someone they don’t know.
  • Offer an incentive: If you can, ask your stakeholders if you can provide something to get the users interest. You may not always be able to, but the offer of a voucher or a product in return for their taking part may well help improve your success rate.
  • Tailor your message: Don’t just put lots of email addresses in the To field and hit Send All. Try including parts in your message which are tailored to that individual. This could include showing where you got their details, or, even better, a contact you share in common. Whilst this makes the process a little longer, the personal touch may well help influence people better.
  • Check, and check again: Before you hit send, make sure everything is right. Is the spelling and grammar correct? Have you got their name and position right? What about other details? Getting a poorly worded email is often a guarantee the reader will bin the message.

Through our outreach, we did find that success rates were somewhat low – often only one in around twenty people contacted would response with interest. Discussing this with friends who work in marketing fields, I understood this to be a common problem. Busy people don’t respond well to a cold approach. It was important we explored other avenues.

Industry events

Part five

The Springer stall at the ACS Fall Conference

The Springer stall at the ACS Fall Conference

One of the most successful efforts in this exercise was to attend industry events, such as conferences and promotions. The company would often have a stand at these events, which meant we could take up a corner with a small table. Here we asked passers by to take part in on-the-spot user testing sessions, or to sign up for user research sessions later. Whilst we did get a good number of people, we did find that there were a few issues with the format:

  • We only had a short time. People would get distracted if we questioned them for too long, and want to explore the conference further.
  • The people we were working with weren’t always our users. Attendees may not have seen our product before. This wasn’t a problem in some ways, as getting a clean impression of our work was a good thing, but it did mean that they didn’t have an previous context with the product.

We therefore found that industry events were useful for short user testing sessions, looking at a particular feature or filling out a survey. This helped us get larger numbers of less in-depth answers. This is worth bearing in mind when designing your research for a situation such as this.

Photo of user research session at a conference

Conducting user testing at our booth at ACS 2017

Taking it to the next level – establishing relationships

Part six

Once you have a list of people who are willing to take part in your research, it’s important to explore their connections. Ask those who have signed up whether they know anyone else who would be willing to help. Do they think their head of department could assist? At one point, I provided posters to our research participants with the words “MATERIALS SCIENTISTS NEEDED” across it, for them to put up around their campus, or pass out to interested parties.

The ideal here is to establish a rapport with your candidates and their institutions, so that they not only come to you once, but can keep providing people for your research.

Epilogue: that sounds like a lot of work!

Part seven

You might be reading this and say; “how can I do this, and keep up the day-to-day UX work?”. This is a valid point, but the trick here is not to do this alone. Involve your stakeholders, and your team. Product owners and business analysts should be able to help, as well as people in sales or marketing. Speak to your design lead, and see if they can get help for you. Most of all organise your time, so you can spend just an hour or so on this task each day. This will prevent the work you are doing on finding research candidates from interrupting your other, everyday work.

As the start of this post stated, research and testing is paramount to your work. In order to produce designs that address user needs, you need to be able to conduct research sessions with users. During the production process, you need to regularly test stages of development with users. It is all too easy to put off user research as too much work, but don’t fall into the trap.

Unverified assumptions lead to insufficient products. It is only by testing our assumptions that we can be sure our product is right for the user.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This site uses Akismet to reduce spam. Learn how your comment data is processed.