Google SERP Eye-Tracking: 2005 vs. 2014
Update: Mediative’s (@Mediative) Chris Pinkerton (@ChrisPinkerton) shares fresh insights on the 2014 eye-tracking study at SMX East 2015. His insights shared at SMX East 2015 complement the SMX West 2015 session on eye-tracking presented by his colleague Gord Hotchkiss. This article now combines the insights from both these conference presentations with our original coverage of eye-tracking research at the bottom of this article.
Chris Pinkerton, vice president of North American client development for Mediative, shared fresh thoughts on Mediative’s 2014 eye-tracking study as compared to the 2005 study. The study sheds light on what a top organic placement means in terms of capturing traffic.
2005: The Start of SERP Eye-Tracking Research
Research found that users spot-scanned the listing as they went down the page and focused on the upper-left corner. The green and blue indicates peripheral vision, and the entire scan lasts 2.5 seconds. Pinkerton points out that in 2005, the SERP was a much simpler and more straightforward landscape (10 blue links). The study was conducted with 30 people. (Pinkerton points out that 20 people are considered statistically relevant for an eye-tracking study.)
“Google’s algorithm and search results pages have changed significantly since 2005. This means people have changed the way they look at, and click on, the results,” Pinkerton says.
2014: Updated SERP Eye-Tracking Research
53 participants of mixed age, gender and search savvy were taken through 43 different search tasks conducted on a desktop using Google. Metrics measured:
- Time to first look
- Time spent looking
- Percent of participants looking
- Number of clicks
Example search task: “Imagine you are moving from Toronto to Vancouver and are looking for an apartment. Use Google to find apartments for rent.”
There’s no longer a triangle. Because people are consuming so much content today, they scan much more quickly. They now process the SERP in 1-1.5 seconds.
Note that the way the information is scanned is now vertical – Pinkerton points to our use of mobile devices as the cause of this new propensity toward vertical scanning. He also shares that in 2007, Mediative did a similar eye-tracking study in China with Baidu and found the same vertical scanning tendencies.
Navigational vs. Informational Search
When a user is conducting informational search, they’re more open to various sites – there are four times as many clicks below the fourth organic listing that in navigational search. In navigational search, by contrast, the searcher is looking for a specific site and they tend to find it right away.
- As you dive deeper into the organic space, there are more clicks in 2014 than in 2005 – which is good news for online publishers. You can rank lower down on page one and still get a click.
- Queries are longer and more nuanced. We started to see in the late 2000s a lot longer query length – as businesses and marketers we need to consider what types of questions people are asking when they consider purchasing.
- Rich snippets increase time of gaze – they cause the eye to move that location first. When an image is located in a listed, humans like to see other humans – natural human behavior will make your eyes go to that image first.
- Click behavior on paid listings remains nearly identical in 2005 vs. 2014.
- Whenever there’s functionality on the page, people either condition themselves to avoid it or interact with it – and it depends on their intent. We are so fast at scanning, we immediately disregard anything that’s irrelevant – which is part of the reason why some people have trained themselves to avoid inorganic listings entirely.
- In 2005 there was a lot of peripheral activity, but not as much in 2014. They don’t even really pay attention to the right side of the SERP.
- Businesses that are lower on the SERP (especially positions 2-4) will see more click activity than they used to.
What does this mean for businesses?
- SEO is no long one-size-fits all. Branding opportunities exist, and you need to be visible where there will be the biggest impact on branding and traffic.
- Optimize your results. Front load content and take advantage of meta data and Schema markup.
- It used to be about local, local, local. Now it’s about intent, intent, intent. First listing vs first page depends on the searchers’ intent and your brand.
All data from the Mediative eye-tracking study is available here: Mediative.com/smxeast
What About Mobile?
Mediative is currently undergoing a mobile eye-tracking study. Register here at “http://www.mediative.com/smxeast-mobile/” to get the results as soon as they become available.
EDITOR’S NOTE: The following is the original article about Mediative’s eye-tracking study as shared at SMX West 2015, published on March 3, 2015. Above is the update from SMX East 2015.
Mediative Chief Strategy Officer Gord Hotchkiss and Business Development Manager Matt Agtarap share results from their company’s 2014 Google SERP eye-tracking study in the aptly titled “How Users View and Interact with Contemporary Google Search Results” at Search Marketing Expo (SMX) West 2015.
2005 Eye-Tracking Results
In a 30-person case study in 2005, participants’ eyes automatically focused on the upper left. Mediative called this area the Golden Triangle.
“When searching for information, we use the same strategy we use when foraging for food. We’re looking for something on the page that corresponds to the idea in our mind. We want to find something that promises we’ll find what we’re looking for on that page — these decisions are made in split sections,” Hotchkiss explains.
Searchers will look for clues, represented by the links that are shown on the page. The first place searchers look is in the hyper link title — then the eyes will go to the left hand side. Searchers typically only look at four different results and compare them at any one time — so they chunk the search page into consideration pieces. In 2005, 100 percent of people would scan the first four listings, and 80 percent will make a click within those four.
But SERPs don’t look like this anymore …
2014 Eye-Tracking Results
Videos, local listing, images, hotel listing, carousels, direct answers — today’s SERP is much different from the SERP of 2005. Accordingly, where searchers have been conditioned to look has also changed. Last year, Mediative conducted a new eye-tracking study (with 53 participants).
- Users scan the whole SERP quickly — trying to identify the “chunk” of results that is most relevant to their query. (Click to Tweet)
- There is less horizontal scanning then there used to be.
- Average duration on a SERP in 2005 was 14 to 15 seconds. In 2014, it took 8 to 9 seconds. (We’re scanning more real estate, but we’re finding something we’re confident in clicking on twice as fast.) (Click to Tweet)
Some other takeaways from the 2014 eye-tracking study:
- Less than 10% of users look at ads listed on the right side. (Click to Tweet)
- Businesses that are lower on the SERP (especially positions 2-4) will see more click activity than they used to. (Click to Tweet)
- The fourth organic listing earns 26% of clicks. (Click to Tweet)
Hotchkiss shares more facts:
- 50-60 percent of our feelings of a page are created before we even click — the stars, the position, etc. influence our feelings pertaining to a listing.
- We’re naturally programmed to look at images — thumbnails will attract attention.
- When searchers are looking for relevant content, there is more scanning and four times as many clicks below the fourth organic listings.
Implications for SERP Success
- It’s not always about getting to the No. 1 organic result.
- A solid, user-focused content strategy is much more robust than chasing the algorithm.
- SEO is no longer one-size-fits-all. Branding opportunities exist and you need to be visible where there will be the biggest impact on branding and traffic.
6 Replies to “Google SERP Eye-Tracking: 2005 vs. 2014”
Very interesting reading.
I wonder how the new snack pack is affecting this?
As google keeps toying with the layout of SERP pages, eye tracking must change as well?
Thank you for the study. I have 2 questions – how does the findings relate to actual clicks on SERPS? Can we say that Page title and meta description plays an even more important role?
Very interesting study.It turns out sometimes it’s better for a website to rank on the fourth position rather than rank on the third one. However, there is a statement we can’t agree with “50-60 percent of our feelings of a page are created before we even click”. Is it possible to measure somebody’s feelings?