CRO | Saturday December 26, 2020
Google is constantly conducting all kinds of experiments to determine which websites are most relevant to a search query. But also to determine how they can best display those search results. This week we saw a striking experiment pass by.
I wondered what was behind it. In this article, I’ll explain a little more about the psychology behind online experiments.
Search via Google, it is impossible to imagine our lives without. It is probably the most widely used tool on the internet. Google’s right to exist is the user-friendliness with which you can find information that you are looking for.
In order to make the search engine even more user-friendly and to ensure the most relevant search results, Google is constantly experimenting and improving their platform. Earlier this week, we noticed that Google has started an experiment in displaying search results.
Unfortunately, this experiment is not yet accessible to everyone. So you have to be lucky to see it pass by. Very often opening an incognito window and doing a search there can help too.
In this experiment, you will see more information directly in the search result so that you can either immediately find the answer to your question and no longer have to click through or you will see a confirmation that this is the website you want to visit.
An icon is displayed in front of the meta description of all search results. When you then move your mouse over the search result, you will first (briefly) see a loading bar appear. You can then see the following:
Seems like Google Search is testing a new UI… I like how smooth it is.
It will now show image previews if you hover your mouse over a search result/link.
— Ishan Agarwal (@ishanagarwal24) December 16, 2020
Google’s vision has always been to display information that is as relevant as possible for every you search for. In order to make that information increasingly relevant, they are constantly conducting experiments.
Such an experiment begins with a hypothesis. That is a proposition that you will test in your research. I assume that this is also the starting point for this experiment. I suspect Google has formulated a hypothesis like:
By adding more context to the search results with images and extra text, visitors can filter the information faster without having to leave the Google search results. This provides a better user experience and a saving of time for the users.
The next question is why they chose this particular experiment. To find out more about this, we need knowledge of how our brain works. Much research has been done into what moves people when they use websites and online tools. And that is precisely one of my areas of interest and expertise.
I have done a lot of research into how existing psychological phenomena apply to online behaviour. And I think these two are behind the experiment:
The image superiority effect refers to the phenomenon where pictures and images are remembered before words. This effect has been demonstrated in many experiments with different methods.
In this case, visitors can better estimate whether this is what they are looking for through images. Let me explain that with an example:
Suppose you have a plant in your home that does not thrive in winter. You are looking for what you can do to get the plant back in better condition. When you search on Google for “How do I take care of [name of plant] in winter”, and you would see an image in the search results of the approximate state of your plant, you can more easily recognize whether the content is relevant to you. If there is also an image showing the plant is completely healthy again, that is even better.
Did you know that experiments have repeatedly shown that people tend to test their thoughts in a one-sided way? They (unconsciously) search for evidence consistent with their current thoughts. Instead of searching through all the relevant evidence, they formulate in their head the very questions that ensure that they find an affirmative answer to support their theory.
Such confirmation bias can be annoying when searching on Google. You are often looking for an answer and you already partly have the answer in mind. Currently, Google is showing a selected part of the text with bold words (what you’re looking for). It is difficult to put the text in context. Because Google shows more text with this experiment, you can better estimate which search result matches your search query.
When everyone sees this kind of extra information, more people are likely to find the answer to the question they have on the search results page. As a result, fewer people click on your page.
In recent years, Google seems to want to answer more and more questions directly on the search results page. Although this provides a better user experience, it also reduces traffic to the sites that provide the information.
The worldwide known search engine optimization specialist Rand Fishkin already researched this in 2019. His research found that the number of queries that didn’t lead to a click had risen to almost 50%. With searches on mobile devices, even more, than 60% of searches no longer lead to a click.
While this has raised questions in the US Congress, as an individual entrepreneur or marketer, there’s little you can do about it. At most, you can ensure that you make smart use of the opportunities offered for extra visibility. And then we hope and keep your fingers crossed that it positively contributes to your brand awareness and still yields more clicks.
Google is and will always be looking for ways to show you the best search results. At least you can assume that. In order to do this, they continue to experiment. That does not always mean that they will actually put it live. There have been enough experiments that turned out not to work.
Have you seen this experiment yet? Do you think it is a valuable addition? If you read this article at a later time, it may be that the experiment has already ended.
Are you interested in collecting data for your own website as Google does? Maybe this is a good time to try scroll depth tracking using Google Tag Manager and Google Analytics.