Google Ranking Factors Explained and Recreated: Part III

Now it is time to look at the simple arithmetic that was used to recreate Google ranking factors for our case study.

In part 1 of our in-depth local SEO case study we took a look at what we could learn from the collection of 12 search results, with 3 pages of results for each search.

In the second section of our local SEO Case Study, we looked at how each different company fared across all of those searches, examined what the local search ranking factors are and how they work. We also established that we have been able to get metrics to recreate an estimated 75% of the local search ranking factors for these companies from publicly available information using a number of specialized SEO tools.

This article explains how each of those Google ranking factors was weighted in order to approximate Google’s search results.

I have shared all of my data along the way to help illustrate all that went into this. The next sections will explain all the different equations that were used to calculate the point values for this graph:

You can skip to the end if you would like to skip the explanations of the equations used and just see the spreadsheet of the original Google ranking factor data and the adjusted data after the equations.

The Equations to Recreate Google Ranking Factors

We will go through each of the local search ranking factors that we talked about in our last article and show the equation and explain briefly why each was weighted as such. We will then show the final spreadsheet with the weights and the final graph again. If you need more explanation of any of these Google ranking factors please see the previous article of the SEO case study that goes into more depth or if you’d like to learn about all of them (not just those included here) take a look at our ultimate guide to local SEO.

These are in order of the importance of their category, as listed in the previous article. Percentages are from the Moz Local Ranking Factors Survey of 2017. Some factors were used for more than one area, but they are only described once.

With each ranking factor I also included a statistic or the statistics that I thought would best give an idea for what was normal for that category. That way if you look into any of these factors for your own site you have more of an idea of how your site stacks up!

1. Link Signals: Incoming Links (29%)

These are all measuring the external links that are linking to the site, also known as backlinks.

1a. Domain Authority (DA)

Domain Authority is on a logarithmic scale of 1-100 that is essentially measuring the importance of the domain relative to other URLs. Due to the logarithmic nature of this scale this is one area where a more complicated equation would probably achieve better results. Most of the Domain Authorities are also rather low, which means they will not be as important.

Despite these challenges, we were going for the general gist here and not perfect recreation. And we got it.

Equation = Domain Authority x 3

Average DA score before adjustment = 19.13

1b. Page Authority (PA)

Page Authority is also on a logarithmic scale of 0-100. The same issues are present here. Page authority is only measuring the importance of the single page, which makes it less important. It is also easier to improve and get a higher score.

Equation = Page Authority x 2

Average PA score before adjustment = 24.25

1c. Referring Domains

A referring domain is a domain name such as wwww.multiversemediagroup.com. As discussed previously, the number of backlinks a site has can be very misleading. Referring domains is a better metric, but by no means perfect. Quality of domains is far more important. Both Domain Authority and Page Authority include measures of quality as well as quantity, but adding this separate metric allowed us to have another data point that served to increase the weight of backlinks in general. Since we were not looking in depth enough here to assess the quality of each linking domain, we used the simple system below to give different weights to different amounts of referring domains.

As you can see, getting a lot of referring domains is not easy. That is why I also included this as a metric beyond DA and PA.

2. On Page Signals (24%)

These are the signals that are actually coming from the page itself. This is also a 0-100 scale, but it is a linear score.

2a. Multiverse Free SEO Audit Tool

This is a free SEO Audit tool offered by Multiverse  that evaluates a number of important on-page signals. It is on a 1-100 linear scale. It is much easier to score a high score on this than it is for either DA or PA due to a few major reasons: just about everything for on page signals is actually under your control (where backlinks are not), the linear scale is much easier, and the factors included are much easier to achieve in general.

Equation: Score x 2
Average score before adjustment = 67.84

2b. Domain Authority (see above)

2c. Name, Address, Phone Number (NAP) on Page

Yes, this is that important. Put it in your footer if you want your business to be found in Google’s local results. If the search was done in their home area those companies received 50. It was devalued significantly to have anything less than the full NAP because it is important to have it all throughout your site.

Why were companies graded down if they had the proper NAP information but they were in the wrong city?
I graded companies down that had their NAP done properly but that were in a different city because we were looking at rankings for this search being conducted in Jacksonville, Florida and since these searches would normally turn up local results this is something that they had to overcome to get ranked.

2d. Bonus: Keywords in URL

This was not included in the previous post, but it is important. It is something that is looked at in the on page SEO audit, but I thought it was important enough to make a point of mentioning here. Adding this factor to the study probably has minimal if any effect on our results, but I wanted to emphasize it because it is important enough to make sure you think about it for your own company and URL naming structure. DO NOT use naming systems that use numbers or other meaningless names. Wordpress can do this if you let it. Make sure your URLs are named after whatever they are talking about. Don’t make it spammy and overdo it, it should make sense and be on topic for the page.

*Sidenote: don’t use dashes in your webpage name. People don’t remember. Also make sure to use .com as your ending, not .net or anything else. It’s harder for people to remember other endings when they are used to remembering .com addresses. Make it easy on them and you’ll have more visitors.

3. Behavioural Signals (11%)

No signals used.

4. Personalization (9%)

No signals used.

5. Citation Signals: NAP Consistency and Volume (8 %)

5a. Moz Local Score

This is something that you can determine for your own site by going to Moz Local Business Listing Search and putting in your domain. It is a score of 1-100 that is on a linear scale.

Equation = Moz local score / 2

Average score before adjustment = 57.59

While this is important, I didn’t want this to overwhelm other metrics that were harder to obtain. Relatively speaking it is much easier for companies to score highly on this than on DA or PA. This is also only supposed to be about 8 % of search results, which calls for a bit of a scaling back.

5b. On Site NAP Evaluation (Same as Above)

This was weighted heavier to account for being relevant in both categories.

6. Google My Business (GMB) Signals (7 %)

6a. NAP Correct on GMB

This was already an important part of the Moz Local score, so I didn’t do a separate analysis or score for this.

6b. Location

This was extremely important. As discussed in a previous article, for the Google local 3 packs location is the #1 priority. In local Google ranking factors it isn’t quite so high, but is still quite important.  I did experiment a little bit with the results before settling on how much to amplify location, as I considered this to be one of the factors that would be hardest to pinpoint exactly. Holding everything else constant this was closest to the best fit.

6c. GMB Category

This is something that has quite a bit of significance. As mentioned in a previous article, best practices for GMB categories aren’t always completely clear. For our particular searches the measurement criteria is more clear. There is a GMB category called ‘general contractor’ and the category actually includes the city of the company, so ‘general contractor in Jacksonville, Florida’ is clearly the ideal category for this set of searches.

6d. Keyword in Business Title

In the days of Google, this is something to consider. Your business name is generally going to be the easiest search result for you to show up for, especially if you have a URL that uses part or all of your business name. If you only get one set of keywords don’t you think it should be a good set?

A Quick Note About if this is a ‘Spammy Practice or Not’

Some people consider this spammy, but I don’t think it is. This just means that it pays to have your name be clear and to the point. Don’t get cute and don’t just use your personal name. Those types of names generally aren’t helpful for Google nor are they for people. While some SEOs view this as a flaw of Google and other search engines, I don’t think this is going change. At the end of the day if your business name is ‘Jacksonville General Contractors’ how is Google supposed to know if someone is searching directly for your company or for a general contractor in Jacksonville? Google will likely just have to guess unless the searcher has other things in their search history it can base that decision off of.

7. Review Signals (7 %)

7a. Number of Reviews

For simplicity’s sake and to demonstrate how much more important Google My Business (GMB) reviews are than other reviews these are the only reviews we looked at. Some of the meaningful goals were broken out in the last article, and those were used as the basis for the scoring here.

As you can see, I didn’t value this a whole lot, but wanted to include it because this is such an easy thing to do. If you don’t have any reviews please make a point of reaching out to some happy clients and asking for GMB reviews by sending them a link to your review page!

Even just one review is so much better than 0 (technically infinite times better!).

If you can, try and make that jump to getting at least 5 reviews. Asking for reviews should be something that is built into your process for after a job is completed, but make sure your client is happy first.

7b. Google My Business Average Review Score

These days a 4 star review is a bad review. Keep that in mind when you review other businesses. If you have concerns do the business the courtesy of telling them so they can address the concerns before leaving a bad review about it. Also be proactive about addressing concerns that people bring up about your business. Find ways to let people know they can give you feedback. If things are left unaddressed it may come out in a review.

7c. Bonus! Google My Business Hours

While these are not crazy important, they do matter. This increases the credibility of your business by having your business hours that people can visit you. However, be careful. If you post business hours Google expects people to be able to come into your business and interact with someone. Google is now crowdsourcing certain information, so if someone wants to meet with you and finds you aren’t there and they report this to Google it could cause your account to be affected.

Posting honest business hours is the best way to go.

Social Signals (4 %)

No signals used.

The Grand Finale: All of our Google Ranking Factors Put Together

First I have published the master ranking sheet which has all of the original data with the Google ranking factors before the equations. Then I have posted the master sheet with values converted to point values. For both I have color coded them accordingly:

• Green = Good
• Yellow = Mediocre
• Orange = Subpar
• Red = Bad
• Bright red = Really bad

Keep in mind that these are color coded this way relative to this specific set of searches. Some of these factors are good in general for all searches and some are only good for this search in particular. Hopefully at this point if you’ve read the other articles you know which are which!

The first row and column are sticky on the original so the company name always stays put, but at this point this functionality is not retained when embedding spreadsheets. Hopefully one day it will just start working due to someone’s hard work and innovation at Google…until then you’ll just have to keep track of which row and column you were looking at when scrolling.

The first three categories of this spreadsheet (traffic, keywords, and backlinks) are actually extra categories that were not used for the Google ranking factors for this study but that are included to give some extra insight into each of the brands and their online presence.