The SERP feature is an umbrella category of design elements on search-engine results pages (SERPs) in which featured snippets, People also ask and knowledge panels all belong.
These features are heavily used, and they can strongly influence the search process. They can direct attention on the results page, help users modify their queries, influence their information needs, and even end tasks on the SERP itself.
The findings reported in this article come from a series of eyetracking and usability tests conducted from 2016 to 2019.
Definitions of the SERP Features
There are more SERP features beyond these three, and Google (and other web-wide search engines) constantly iterates on them. Google changes the design of its results page frequently, so the SERP-feature designs you see may differ from the designs covered here, depending on the time of reading.
There are two main reasons you should care about the design of these SERP features:
- SEO (search engine optimization) is a main source of qualified traffic for most websites. To succeed on the Internet, you must know how people use search and how they interact with SERP features, as shown by independent user research.
- Search on your own website (or across your intranet) will be heavily influenced by search on public search engines due to Jakob’s law of the Internet user experience, which says that people spend most of their time on websites other than yours. Specifically, users’ mental models for how search and search features work will be shaped by their experience with Google (or with the dominant search engine in their country), so your own site-search features should cater to the expectations created by the big search engines.
Featured Snippets
Featured snippets are excerpts of site content that summarize an answer to a query. They are presented in a card, and may consist of a paragraph, list, table, or a video. In some cases, they include a combination of those elements.
People Also Ask
People also ask is a list of questions related to the query, presented in accordions. Each question unveils a featured snippet along with additional questions. Essentially, these are collections of featured snippets for related queries — sneak peeks into parallel SERPs.
Knowledge Panel
The knowledge panel appears in the right sidebar on some SERPs. It shows attributes of entities (people, places, organizations, things), using information gathered from the web, structured databases, and other sources.
When researching companies that they were unfamiliar with, users in our study were heavily guided by the knowledge panel. These cards usually held high-level information about a company —things like the CEO, headquarters, address, and phone number.
Most knowledge panels contain important and relevant details — such as the dates of events at a concert venue or the stock price of the company.
The Roles of SERP Features
All three of these SERP features (and others not defined here) influence the user’s information-seeking process. They can:
- Act as signposts
- Direct the user’s attention
- Modify queries and tasks
- Provide quick answers
Acting as Signposts
SERP features — particularly the ones with images and video thumbnails — can take on the role that logos play on a webpage: they act as signposts that enable users to quickly verify that the results are indeed for the entity they are interested in.
Directing Attention
In the old days of search, when SERPs were just a predictable set of text links, users tended to scan all SERPs in the same way — linearly, starting at the top of the page and only looking at a few of the top results before making a selection.
SERPs aren’t predictable anymore — any given query could bring up a unique arrangement of text results and SERP features. As a result, people don’t scan linearly as often anymore. Their scanning patterns tend to bounce around the page, moving from SERP feature, to text result, to SERP feature, in a pinball pattern.
This new behavior makes sense for two reasons:
- SERP features receive special visual emphasis and visual weight on the page. Thus, they stand out and attract the eye.
- Exploring SERP features has a low interaction cost, but a high expected benefit.
In the process of information foraging, humans do a continuous cost–benefit analysis to decide whether the interaction cost of scanning or clicking on a result is worth their time. Users have learned that most of the time the SERP features contain the answers they’re looking for (thus, they have a high benefit). SERP features involve little interaction cost, because they don’t require an extra click, reading lengthy content, or leaving the current view.
In some cases, the SERP features received attention even after people had clicked a result. Several users in the eyetracking study continued to look at the knowledge panel after they had selected a result while they waited for it to load, a phenomenon we call post-click viewing. On some occasions, the user returned to the SERP and either read the text in the knowledge panel more thoroughly or clicked on the link inside the knowledge panel.
For example, one participant selected a link from the SERP and briefly looked at the logo for the company Tyco. She later returned to the SERP and then read the knowledge panel carefully.
Modifying Queries and Tasks
Queries are typically impoverished representations of an information need, and that’s why most queries are highly ambiguous. In a recent meta-analysis of 471 search queries, the average query length was 4 words.
We know that most users are terrible at generating alternate search strategies. But through SERP features, Google can present several expanded interpretations of the query in natural language and allow the user to explore those alternatives without leaving the page. It's like browsing the grocery aisles to find similar products. People don’t have to think up new ways of searching themselves; they can simply pick from the computer-generated alternatives, which is much easier to do.
The interaction cost of clicking on a question in People also ask is far lower than clicking on a regular result on the SERP or submitting another query. (Remember, the answer is simply expanded in situ, so there’s no extra page load.)
For complex searches that require multiple queries, the feature lets users rapidly move through variations related to the topic, all while remaining on the SERP. When one accordion is opened, Google loads more questions into the feature, usually related to the question that was clicked.
We observed multiple instances when People also ask features heavily influenced the user’s task. One participant was looking for the approximate cost of getting Botox injections for crows feet wrinkles. She noticed the People also ask feature on the SERP. One question read Is Botox safe for crows feet? This was a question she hadn’t previously considered, so she expanded the accordion.
Her click gave Google important feedback about her interest in the safety of Botox, and it loaded a new question, What are the bad side effects of Botox? After reading that featured snippet, she decided she might not want Botox after all. She then googled noninvasive approaches to crows feet.
By presenting related information and questions, Google can redirect users’ queries and entire tasks onto new targets.
Providing Quick Answers
In cases where users have a simple information need, SERP features can provide a quick and easy answer. If users want an address, phone number, or definition, there’s no need for them to click to through to visit a site. Good abandonment occurs when users finds their answer on a SERP and decide to end their task successfully.
Even for complex tasks users might be satisfied with a quick SERP-feature answer if they aren’t highly motivated or are pressed for time.
For example, we asked participants to find out why it’s important to take a full course of antibiotics, rather than just stopping as soon as you feel better. One participant googled why do you have to finish antibiotics; then he scanned the featured snippet and one of the People also ask questions. That was enough information for him to feel satisfied, so he ended his task there.
Consider SERP Features for In-Site Search
If you can’t beat ‘em, join ‘em — and you can’t beat Google.
Google led the charge with query suggestions, and now they’re often an expected feature for in-site search. Google has been experimenting with SERP features for more than ten years, so users are becoming familiar and comfortable with them. Some sites have started including their own SERP features on their in-site SERPs, and you might want to consider doing the same.
In particular, featured snippets are attractive to summary-reliant people and can work well for in-site search. You may be able to manually define the preferred result for common queries, and present those as featured snippets.
For example, Encyclopedia Britannica displays a top result that mimics Google’s featured snippet. Many participants in an eyetracking study scanned this featured snippet and selected it.
However, some people in the study skipped over this featured result, probably because they thought it was an ad. Ignoring it wasn’t a big deal, because the same page was also presented as a traditional text result directly below the featured snippet. Users who skipped over the featured snippet just clicked the regular result.
To avoid this type of banner blindness, follow Google’s lead in designing SERP features, as far as it fits with your site’s visual design. This is one of the rare cases where it may not be a bad thing to copy a big site’s design, because it will help ensure that your users will recognize what they are.
Share this article: