Semantic Differential Survey – Validating How Users View a Brand

When stakeholders complain about a design being boring, dull, or not communicating the desired messages, you can address these concerns by preparing upfront and conducting tests. The first step is to agree on brand words with your client before starting the design process. Tools like BrandDeck can help stakeholders select words that describe the desired brand, such as calm, individual, or futuristic.

Once you have a set of agreed-upon brand words, you can test whether the design delivers on those keywords using a semantic differential survey.

Conduct a semantic differential survey:

1. Log in to Lyssna and create a new study within your project, naming it “Branding Survey.”
2. Select “Desktop Only” and choose the “Design Survey” option.
3. Upload your design (e.g., the Frontend Masters website).
4. Set the question type to multi-choice and ask, “Which of the following words do you associate with this design?”
5. Input the brand words and their opposites (e.g., futuristic vs. historic, calm vs. busy).
6. Include any words mentioned by the stakeholder, such as “dull” or “boring.”
7. Optionally, allow people to suggest their own words.
8. Randomize the order of the choices to avoid bias towards top-listed items.
9. Save and continue, choosing either Lyssna’s Recruitment Panel or a shareable link for your audience.
10. Share the link with participants and have them complete the survey by selecting the words they feel are most appropriate.

After participants have completed the survey, log back into Lyssna and view the results within your project. The results will display the votes for different answers and highlight the most popular choices. This information can demonstrate whether the design is delivering on the intended brand keywords and if the client’s concerns are representative of the audience’s feelings.

The Value of Semantic Differential Surveys

Semantic differential surveys are a powerful tool for validating design decisions and addressing stakeholder concerns. By establishing agreed-upon brand words upfront and testing the design against those keywords, designers can gather objective data on how well the design communicates the desired brand attributes. This approach helps to minimize subjective opinions and ensures that the design aligns with the intended brand message. Incorporating semantic differential surveys into the design process can lead to more effective designs that resonate with the target audience and satisfy stakeholder expectations.

Semantic Differential Surveys Video Transcript

So every once in a while, you will come across stakeholders complaining about the design that you’ve produced for them.

They might be saying it’s boring or that it’s dull or it doesn’t wow them or whatever else.

And they complain that it doesn’t communicate the messages that they want to communicate.

So in such situations, you can deal with that if you do a little bit of preparation upfront and a little bit of testing.

In terms of the preparation upfront, what you need to do is you need to agree with your client what your brand words are going to be.

So I typically use this tool called BrandDeck, branding.cards.

And essentially, it’s a deck of cards, either digital or physical, with different words that could describe a brand, words like calm, individual, futuristic, etc.

And so essentially, what your stakeholders can do before the project begins, before you do the design, is agree a set of words that you want the design to communicate.

And once you’ve got that set of words, then essentially, we can test whether or not the design delivers on those keywords or not.

And we can do that with something called a semantic differential survey.

And it’s a very fancy name for a very simple thing.

And we can do it by going to Lyssna.

We can log in to Lyssna, go into our Frontend Mastersproject, and create a new study, which we’re going to call our Branding Survey.

We’re going to select Desktop Only, as I seem to always do on these tests.

And then we’re going to go for a Design Survey.

And we’re going to take our design, whatever that is, in this case, Frontend Masters website, and we’re going to upload it.

And then we’re going to ask the question, which…oh, first of all, we’re going to set that it is a multi-choice question.

Which of the following words do you associate with this design?

Okay?

And then we can put in our brand words.

So our brand word might be futuristic, and it’s opposite, historic, right?

Or another brand word might be calm, and its opposite would be busy.

So you see what I’m doing there?

I’m putting in the brand keywords, I’m putting in their opposites.

But I can also put in any words the stakeholder mentions.

So if they say that they feel like the design is dull, we can put in the word dull.

Or if they say it’s boring, we can put in the word boring.

Okay?

And we can even allow people to suggest their own words if you want to.

But we do need to randomize the order of the choices because otherwise people tend to select things at the top of the list.

Then once that’s done, we just hit Save and Continue.

And we can either use Lyssna’s Recruitment Panel to get people to complete the tests for us, or alternatively, we can get a link that we share with our audience.

So we’re going to get a link for now.

We’re going to skip all the various options that are available to us so that we’ve just got that link, which we can copy.

And then essentially, we can paste that in, go to that link, start the survey, and there you go.

You see the design, and then people can select whichever words they feel are most appropriate, and then click Continue.

And that is submitted.

Job done.

Now if we go back to Lyssna, and we log in, go into our project again, and branding, and we can see our results here.

So what this will do is this will show me all of the votes for different answers and which ones turn out to be the most popular.

And that’s a great way of showing that you actually are delivering on the keywords that the brand was supposed to communicate, and that the words that the client used are not representative of how our audience feels.