For well over a year now I have been meaning to test this: can the rank of your Google Places (Google+ Local) page be improved, by building links to it. I’ll admit, I had some half ass attempts in 2012 where it seemed like a moved the needle, in fact, Darren Shaw was even crazy enough to include it in his MozCon Slidedeck.
I’ve wondered, why haven’t we heard more about this? It’s been a year+ since it seemed probable that this might work, and few have spoken out on it. Is it possible that everyone but me did a test? Made it work, and now is keeping it to themselves? Jerks!
So, once and for all, I am going to put this sucker to rest.
- Why do I think these things can be ranked? Two reasons: unlike in the past, the Google Places listing is now indexed. Two, most all the data on the page is also indexed. If I didn’t know better, this is starting to sound like a website…check it:
- More or less, I have been using the same ranking techniques for Google Places for 2+ years. I am bored, and it’s time to go back to the drawing board and shake things up.
Jacob King & Adam Steele
Do links built to the Google Places page, increase the page’s ranking in local search results?
We will build links to 4 Google Places pages. 2 pages will be merged, 2 will still be non-merged. Links will be built for anchors that trigger traditional pak and blended. An even amount of links will be built to each page.
- run a Whitespark citation benchmark for the listing, as well as run the listing through Ahrefs;
- rank trackers from MicroSiteMasters and Whitespark will be used. The “https://plus.google.com” URL will be tracked;
- choose a listing that: has no site attached, is not for profit (ie. a high school) and appears dormant;
- log all links built, so as to compare against Ahrefs 30 days later to make sure we were the only ones promoting the listings;
- rank track 4 additional listings, with the same makeup as our test subjects, and compare movement against eachother; and
- choose businesses in the same city. Difference in SERP landscapes might throw things off
The study will run for 30 calendar days, after which time, if at least 2/4 listings have positive progress, we may have gained some insight into whether this has any influence.
Next steps are picking the listings, the keywords, setting the benchmarks, and getting under way. All of which shouldn’t take too much longer than a day or two. I figure we will update this post every 10 days, explaining what has been done to date, and any observations we have made.
I would appreciate very much if anyone had something to contribute – a control that we perhaps missed.
Please, for those of you nay sayers out there, PLEASE keep an open mind. This is meant to be a fun experiment – that is all.
I hope you enjoy and I look forward to this.
Update: Just in from local genius Nyagoslav, “you might want to check the rankings of the rest of the business listings, too, because if many of them change at the same time, then it might not be an effect caused by your activity, but rather because of an index update.” Absolutely. Not sure exactly how yet, but I am going to figure something out.
[UPDATE] August 19th
We are now three weeks into the campaign, and the following has been completed:
- listings chosen: this was way more difficult than I had anticipated. We had to find listings at a majority of stages, raw to semi optimized, and hardest of all, they needed to be listings no one was going to cry about…ie. I wasn’t going to possibly negatively affect someone’s livelihood. So, we chose 4 listings, all close to home (Vancouver, BC Canada), all of which are out of business, but not marked as such. I have chosen not to share the listings with you yet, as I don’t want anyone messing with them. They must remain solely at our mercy – for now : )
- keywords chosen: also tricky, I didn’t originally find these listings by searching specific keywords. I searched various queries such as “closed for business vancouver bc” and qualified them from there. Then I had to make sure they ranked for something. We aren’t doing anything more than blast them with links, so I wanted listings that already showed directionality, and just required a push. I didn’t want to have to try and rank something for a term it possibly wasn’t relevant enough, or optimized enough to rank for. Sure, I can do this with an organic listing, but I didn’t need the extra hurdle for this. For some we have to go 20 pages deep, but eventually we found listings that met the criteria.
- rankings: each listing has been manually benchmarked, while also thrown in Whitespark rank tracker. We had to dump MSM as we forgot that it wouldn’t work beyond page 1 for local rankings. We will be updating rankings periodically. Per Nyago’s suggestion, we have also chosen to track for additional listings, which share similar SERPs with our chosen listings. The “why” is above.
- link building: it took us a fair bit longer to get all this organized, and as a result, link building was delayed. Jacob began building links on August 8th, and they are of the automated variety. So far, layer 1 has been completed, with other layers/tiers en route. I will cover anchor text and the rest in later updates. Also, I have run an Ahrefs report on each of these listings so as to make sure no one else is building links to these puppies.
We have been tracking our keywords since the first week of August, and so far see nothing that would suggest what we were doing is having any affect whatsoever. With only tier 1 delivered, and many links remaining to be indexed, I believe it is too early to call it. One notable worth mentioning is that we aren’t seeing many of these links being picked up in Ahrefs for whatever reason. Anyone have any idea why?
Jacob and I plan to get a bit more aggressive all around, and hopefully, in a couple week’s time we see something.
[UPDATE] 9/19/13 – Results
After much work, and 1000s of backlinks the result seems pretty clear…building links to ones Google+ Local page does not appear to improve rankings.
We followed everything above to a tee, and saw no substantial, stand out loss or gain in rankings – stagnant in fact. For those curious minds among you, here are the keywords we tracked, their rankings before (benchmark taken Aug 14) and a few days ago.
Rankings for dates in between are much of the same. Rankings were collected via manual means (local IP, etc.), and Whitespark rank tracker. We did not include the latter because results returned from the tracker (which is generally bang on) were a bit wonky, and not useful.
Jacob and I did everything we could to push these listings, and had hoped for the opposite. Nevertheless, I am happy I finally gave it a real go, and can put this “myth” behind me. I will check the rankings one more time for good measure in 2 weeks, and if nothing, will put this puppy to rest for good.
If anyone has any questions, or would like to see more data from this study, please let me know.