Google’s In-Depth Segment: An In-Depth Review

Google Data Centre
11 Aug 2013
Michael Jarrett

After months of speculation Google finally announced a new search result segment: In depth articles. Right now this segment is only live on Google.com for English results. However, if Google sees a positive engagement from users expect to see these in the UK and other English versions of Google (such as Google.com.au and Google.ca) very quickly.

in depth search result example

This is a significant development for online content providers to gain greater search exposure; it means publisher content can now rank for highly searched for broad keywords like “credit cards” and “new york”.

Which keywords display in-depth articles?

Before reviewing the article level data it’s worth considering the type of keyword that displays the in-depth article segment. Google states it should appear when searching for “a person or organization name, or other broad topic”. Let’s see if this is the case:

We selected keywords by using popular terms appearing with Google Hot Topics, Google Trends and keywords we personally search for (like cake!). We then let Google Suggest guide us to a popular yet slightly longer-tail term. Below is a list of keywords that show the in-depth segment () and longer-tail terms that don’t ():

✔ “lego” ->  “what is lego”
✔ “kate middleton ->  “kate middleton baby”
 “facebook” ->  “growth of facebook”
 “new york” ->  “new york guide”
 “cake” & “cake recipe” ->  “chocolate cake” & “chocolate cake recipe”
 “women’s fashion” ->  “women’s fashion trends”
 “kanye west” ->  “kanye west yeezus”
 “credit cards” ->  “credit cards with rewards”
 “blackberry” ->  “blackberry z10”
 “flights” ->  “flights to florida”
 “weather” ->  “london weather”
 “cricket” ->  “cricket rules”
 “amazon” ->  “amazon security breach”
 “walmart” ->  “walmart careers”
 “xbox” ->  “xbox one”
 “real estate” ->  “real estate listings”
 “beyonce” ->  “beyonce haircut”
 “the great gatsby” ->  “the great gatsby review”
 “ford” ->  “ford fusion”
 “mexico” ->  “mexico city”
 “pizza” ->  “pizza recipes”
 “breaking bad” ->  “breaking bad season 6”
 “apple” ->  “apple iphone 5s”
 “earth” ->  “earth’s orbit”
 “barack obama quotes” ->  “barack obama nsa”

As you can see above Google are only displaying in-depth articles for “broad” keywords. As soon as the search query points to a specific topic, in-depth articles are no longer displayed. For example, Google displayed in-depth articles for “blackberry” and “the great gatsby” however not for “blackberry x10” and “the great gatsby review”, despite some of the in-depth articles actually being based on those topics:

indepthsearchresult2

This would hint that Google would rather see users revert back to their normal habit of scanning the regular ten links rather than reading articles it deems in-depth.

It’s also interesting to note topical and trending keywords don’t display in-depth articles. For example, there has been a lot of content created on the topics “kate middleton baby” and “barack obama nsa”(95,000,000+ pages to be exact) – surely there would be articles worthy of in-depth status? This shows us that the in-depth article segment isn’t designed to focus on news content. We’ll review this in greater detail later.

So what makes an article in-depth? We decided to dig a little deeper and review a number article metrics.

What types of domains do in-depth articles live on?

Our research found that the 75 in-depth articles we researched in this analysis appeared on 33 unique domains. Almost all domains can be classified as online publishers with the exception of ricksteves.com and wikitravel.org which are slightly different in their set-up. Below is a breakdown of the domains:

in depth articles by domain

Nytimes.com appeared within the in-depth segment for 10 of the 25 keywords. This was significantly more than the next most popular domains, which were businessweek.com with six, theguardian.com with five, and newyorker.com with five. It’s worth noting that most of the domains are US-based websites, which makes sense as these results come from Google.com. Once the in-depth segments appear in the UK, it will be interesting to see which domains Google favours.

Does a paywall impact in-depth articles?

The paywall debate has been hotly contested over the last 16 years. We were interested to see if Google ranked restricted content within its in-depth article segment.

indepth articles and paywall

We found that 12 of the 75 in-depth articles came from websites that have a paywall restriction in place. However, looking at this in detail, 10 of the 12 articles came from nytimes.com. It seems Google preferred to rank non-restricted articles.

Does the length of an article determine if it is classified as ‘in-depth’?

From the 75 articles reviewed we found that on average in-depth articles contained 3,950 words. The graph below shows that the most common word counts were 1,000-2,000 and 3,000-4,000 words.

in depth articles by length

Only two articles contained less than 1,000 words and five articles had 10,000+ words (now that’s what I call an in-depth article). The length of content appearing within the in-depth segment highlights that Google is favouring features rather than news or blog content.

Does Google prefer fresh content for in-depth articles?

Within our keyword review we already noted that Google didn’t seem to consider trending topics as a worthy choice for in-depth articles. We found similar results when reviewing the publication date across the 75 articles:

indepth articles and age

The graph above shows that none of the in-depth articles had been created in the last month and only one in the last three months. The most common period was three to six months followed by six to twelve. There were nine from 36+ months bracket and the oldest publication date was from a Washington Post article titled “Ford and GM Scrutinized for Alleged Nazi Collaboration” from 30/11/1998. Three articles had no publication date at all. This shows that Google doesn’t simply consider content from online publishers when creating its in-depth articles segment.

Does Google+ have an impact?

Next we considered the adoption of author and brand profiles. Interestingly the graphs below show two very different adoption levels:

indepth articles and Google+

Most in-depth articles came from websites that had Google+ brand profiles in place. 15 domains didn’t have a Google+ brand profile. This figure is slightly distorted by www.businessweek.com who appeared with seven of the 75 articles and don’t currently have a Google+ profile linked to their article content.

Author profile adoption was almost an exact opposite in terms of percentages. Only eleven articles had a verified Google+ author page. This was a surprisingly low figure. Congratulations to Mathew Honan who had two articles appearing within Google’s associated articles.

Does social engagement impact in-depth articles?

Has Google considered how users have interacted socially with article content? We found that on average an in-depth article had 410 +1s, 8,500 Facebook likes and comments, and 1,160 Tweets.

indepth articles and social shares

The most common social share count for Google+ was 10-50 +1s, for Facebook it was 1,000-2,500, and for Twitter the count bracket was 500-1,000. This data indicates that user engagement is a key factor when Google is selecting in-depth articles.

What does it all mean?

From the research above it seems content length, social shares and Google+ brand pages are all contributing factors for articles that appear within Google’s in-depth segment. It also appears that Google is focusing on established and trusted online publishers.

It’s interesting to see in-depth articles appear for brand terms such as “Apple”, “Walmart” and “Amazon”. If a user is searching on these terms then it is highly likely they are searching for the brand’s homepage. However, with in-depth articles Google have added an extra level of information to their search results page for each brand. Sometimes the in-depth articles don’t represent a positive brand message:

indepth articles and brands

What can a brand do to avoid negative articles appearing? Not a lot. At this stage it doesn’t seem likely that brands will be able to control this segment by trying to have their own content ranking because Google is focusing on online publishers.

How can I appear?

Google have given webmasters a quick breakdown of areas to consider:
https://support.google.com/webmasters/answer/3280182

Digital storytelling just got a little bit more in-depth! Get in touch – we’d love to hear about your own experiences and are happy to answer your questions.

1 Comment

Responses

Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>