Abstract

Correlation does not equal causation, friendly note.

When Google migrated Google Places to Google Plus Local, several business took advantage of this business opportunity and integrated their websites with the new social platform. It is been suggested in the Search Engine Optimization community than Google assigns different values of relevancy to Google Plus Pages than those used to rank organic search results. While many SEO researches have gathered a lot of data about what Google might be using to rank organic listings, Google Places (Google Plus Local) information has been lacking the same amount of scrutiny. This article analyzes 100 Google + Local listings and compares several factors to help draw correlation conclusions about what might affect Google rankings. Jump to the perfect website for Google Plus Local listings.

History

A massive effort to help SEO consultants understand what is relevant information for high Google Plus Local listings was made by David Mihm who surveyed dozens of SEO experts for their opinion about what might affect Googles rankings (Article about G+ Local Ranking factors). The article was published in mid 2012 when Google Plus Local was barely getting started. I took some of the rankings factors mentioned in the research and expanded a few of the ones that I thought will be relevant for 2013.

Method:

Three computers with no search history, cookies, proxies, cache or stored internet files were used to gather the results. The results were compared for accuracy in three browsers: Mozilla, Chrome and Internet Explorer. All results were geo targeted and allowed for Google to use HTML5 to gather geolocation information. Keywords used for this particular study were in the niches of “internet marketing”, “plastic surgery”, “mental health”, and “website design”. All search queries included “keyword” + “location”. Only the top 5 results appearing in Googles organic search listings under the Google Places section were used. This research compares 100 websites and 20 search queries. Because not everybody has claimed or migrated to Google Plus Local, Google Places might be used interchangeably in this article.

Justification of methodology: the reason why this article focuses in the four mentioned markets is because there markets have history of high competitiveness overall. It would be interesting to see what other SEO researchers find on regards of other niches and markets. I suspect that the algorithm adapts intelligently to specific, exclusive, categories.

Design:

Eighteen different independent variables were used to create a correlation matrix in which high rankings (100 for the first position, 90 for the second, 80 for the third, etc) were defined as the dependable variable.

• Indexed Pages: Number of indexed pages as reported by Google
• Bing Backlinks: Number of links reported by Bing. [inbody:website.com -site:website.com]
• Page Authority: As reported in SEOmoz Page Authority
• Domain Authority: As reported in SEOmoz Domain Authority
• Backlinks: as reported in Majestic SEO
• Keyword in Tittle: Having the keyword exactly as the query in the tittle tag of the Google Page
• Pagerank: Page rank of the linked domain (official website of the business)
• Alexa: Rank for the main website
• Responsive: Mobile friendly CSS OR mobile redirect using Iphone User Agent and screen width of 300 px
• First Archived year: As reported in Archive.org
• Loading speed: time it took for the website to load
• Facebook: Facebook likes
• Tweets: Tweets and or retweets
• Plus: Plus ones on the Google Listing site
• Plus Rank: Page rank of the Google Plus Local page
• Exact Category: Having the exact search query in the category section
• Description: Having the keyword in the description exactly as the search query
• Num Reviews: Number of approved Google reviews in the Google Places page.

Results:

Among of the most prominent determinants of the top rankings inside Google Plus Local was having the keyword in the title of the Google Page which had a POSITIVE correlation of 0.35, the highest in this particular research. This translates into having the desired keywords in you Google Plus Local page correlates positively with having a high ranking in Googles organic listings. The highest NEGATIVE correlation was having the keyword in the business category in which having the EXACT search keyword in this area is associated with lower rankings in Googles organic search results (correlation of -0.35).

The top three positive correlated results are: Keyword in title (0.355334527259351), Page Authority (0.228456371946415), and keyword in the description (0.0816496580927726).

The top three negative correlated results are: Exact category in title (-0.353553390593274), Loading speed (-0.246351063511426), and Plus Pagerank (-0.22237479499833)

Conclusions:

It came as a surprise to me to find out that perhaps one of the strongest ranking signals was just having the keyword in the Plus page tittle, it almost felt unfair because there are local businesses that use their brand names than they had from decades ago and sometimes a new business with SEO in mind would legally register with the department of State “Blue Keyword Keyword City” and get higher results. Something that I found shocking was the fact that Google dislikes having the exact keyword in the “category section”, having a broader category is much more preferred. For example if the target keyword is “plastic surgery in city”, instead of having “plastic surgeon” in the category, a better practice would be “medical clinic”.

Something that I predicted was the fact than Google will prefer fast websites in Google Places than slower ones, and I was partially right. Actually Google seems to “put down” websites than are slower/heavier (having a negative correlation of -0.11). It seems logical than faster websites might have a competitive advantage because bounce rate is usually lower in fast websites  and they can usually be opened easily in tablets or cellphones  Also, having too many indexed pages has a negative correlation with high rankings in Google Plus local, perhaps because it dilutes the ranking power of the domain. I was expecting to see responsive/mobile design being a positive factor but at this point all evidence seems to suggest a minimal correlation (-0.11 perhaps do to the fact that very few sites have actually mobile optimized content).

Something that came to me as a surprise is the fact that most local website rank very high without social signals. As a matter of fact, having social signals correlates negatively with high rankings inside Google Plus Local at least based on the results of this experiments. This does not mean that social signals harm Google Plus local listings.. Perhaps it is a design error or perhaps the pages than got social signals had “fake ones”, I do not now. But at this point, I would recommend only social signals from authorities in your niche, not just any customer with a recently created social profile. (just my opinion)

The perfect website for high rankings in Google Plus Local 

The ideal website for a high Google Places/+Local ranking will have the following features based on this correlation study:

1. Have a claimed Google + Page with the keyword in the tittle
2. Have a high landing page authority (SEOmoz)
3. Have the desired keywords in the description
4. Have a decent page rank, at least of one (not too much of a concern at this point)
5. Have only authority pages indexed (add noindex in the robots.txt to pages that do not “help you rank”)
6. Have a broad category inside Google Plus Local, not so much a laser target keyword.
7. Have few or no social signals, social votes ONLY FROM niche authorities.

Further research:

As Google continues to change and adapt, it will be interesting to see what factors will receive more weight as other ranking signals become more relevant such as Google authorship and other emerging social platforms become more widely spread. Again, this is a correlation study and DOES NOT imply causation. I am in the process of developing an experiment in which some factors can be isolated to hopefully create a cause-effect relationship with some of the variables. Of course, Google will change but human psychology will most likely stay the same. By understanding what makes people happy, we might become more attuned to what makes Google happy.

Evolving research: Other niches/markets should be taken into consideration outside to four mentioned here. This study is meant to be just an introduction to further research, a bigger sample size is required to increase the accuracy of the correlation factor. The next step is an actual SEO experiment in which factors are isolated to help draw more accurate conclusions about local SEO rankings. Please feel free to ask questions or join our SEO community to stay informed about the results future experiments.

Recommended sources for Local SEO research:

I recommend business owners to read these articles and follow these Local SEO leaders in Google Plus to stay informed about SEO in Google Plus Local Listings.

• Dave Cosper, EZ Local article: Salsa A recipe for local SEO, Follow +Dave here
• Mike Blumenthal, Categories for Local SEO, Follow +Mike here
• Dave Ashworth, Raise of Author Rank, Follow +Dave here