How to evaluate the content of your UI?
Conducting content testing and investigations.
Why evaluate content
Think of this step as prototyping the conversation we should have with people using the product. So, test for:
- Establish a structure for our conversations with people, create patterns.
- Understand the words we need to use so that people understand us.
- Understand the information and emotions that interfaces need to convey.
- Understand if what we present in the interfaces is the information that needs to be shown (if there is too much or is missing) and how it makes people feel. Are they low on trust, for example? From this, we have a better vision of what and how to communicate.
What to test first?
Testing and measuring the effectiveness of content always happens from the macro to the micro.
In the macro stage, we will be thinking about questions like:
- Are our general keyword choices attacking the pain points?
- How does the product relate to the language people use? Is there anything else that Marketing emphasizes?
- Does the user have enough context to go through our main task streams
- Do people understand what we are communicating on each screen and what its purpose is?
In the micro stage, we will try to respond to:
- Are our CTAs clear, effective, and motivating?
- Do our commands provide enough information for people to pass through points of friction or trust in our task flows?
- Do users understand our labels, titles, instructions, and functionality?
Readability x Reading x Comprehension
We can investigate the effectiveness of a text in terms of its readability, readability, and comprehension.
Readability
The readability of a text tells us about a person’s ability to see your text.
The faster the decoding of characters into concepts, the better their readability.
Some factors will influence this ability, such as font, body, width, angle, style (bold, uppercase), color, the contrast of the letter’s shape with its background, the length of the lines, the leading, the measurement of spacing, and margins. All of this has enormous weight in the readability of a text.
Reading
Ease of reading will measure the complexity of words and sentence structure in a piece of content.
The assumption behind this metric is that complex sentences are more complicated to read than more straightforward sentences.
Comprehension
Finally, the densest level is that of textual comprehension, which will depend on much more than just literacy skills but also the level of education and prior knowledge. The ability to understand the text measures the assimilation of words and ideas present in the text objectively.
That is if a person understands the intended meaning of a text and draws the correct conclusions about it.
We also want them to perform the intended actions after reading the text in instructional or action-oriented content.
- All of these aspects must be considered when testing for content. What are you looking to investigate with your test? Ease of reading only, or understanding as well?
- It is essential to define the target audience for the test clearly, and that it represents the actual user. Avoid testing internally, as the load of prior knowledge can be too great and compromise the results, except for cases where doubt may prevent the finalization of the product taken for testing.
- Clearly outline the content to be tested. The smaller the scope, the better. You can even test just phrases. And always remember to contextualize the person who will respond and explain what is expected of them.
- Reinforce that answering “I don’t know” is possible. A guessed answer is not interesting for investigation. If the person doesn’t know how to answer or complete the task, ask about what they wanted to find on that screen, for example. Explore the moment.
Methods
Once you have defined what you intend to test, define the best method. There are several out there. I’ll leave some examples below of good articles about various testing methods:
Discovery
Explorations
Validation and Iteration:
Tools
I have used the following tools to apply content evaluation. Here are my highlights on each of them.
Qualtrics
This is the motherload of all surveys software. With Qualtrics, you will find a huge variety of question types, which will help you a lot in the evaluations. Its interface is not the friendliest, but it’s worth it. Especially when we talk about the Reports, it is unnecessary to open any excel file to compile the data and cross-reference answers.
It’s expensive, but it’s awesome for surveys.
Alchemer
Use Alchemer if you are on a budget and still need good survey software. You will find a good amount of question types specific for content evaluation. Their report is also robust. It gets the job done.
User Testing
I love working with User Testing. Despite their usability issues, it’s remarkable how quickly you can recruit and test prototypes. The video analysis is also excellent. You can make notes and clips of users testing your product. 10/10
Full Story
Fullstory is a platform that allows you to understand better the behavior of users and how their experience is going. It’s like Hotjar, but with steroids. You can filter sessions to get a more accurate perspective of how users interact with your pages or content.
Highlights
- UX Writing and UX Research MUST walk hand in hand. Research is an essential part of creating quality UX writing. So work closely with the UX Researcher of your team, if you have one.
- Remember: If you have data to back your UX Writing decisions, everything flows better. Evaluating your decisions is one of the best ways to prove the UX Writing ROI.
- Don’t rely on just one method of research. We have a lot of tools to help us experiment and bring results fast.
- Research allows us to create copy with the end-user in mind. Don’t write copy for yourself!
- Engage with your target audience. Build a list of users enthusiastic to test your prototypes and participate in early surveys. Trust me, recruiting is a crucial part of your research and a very exhausting one.
