Massive drop in rankings alert!

[SeoSheikh]

Hi fello SEOs. I've been working for a large UK artificial grass company since last November. They we're converting badly but ranked no.1 for the keyword artificial grass. Now they convert well but have slipped down to 18. That's one hell of a drop!!! The website is www.easigrass.com

But;

Before I came on board they had their main .com website with about 25 subdomains for each major city in the UK. I switched that in February 2016 to sub folders and the rankings for "artificial grass" have drastically dropped in the SERPs but they now have many more pages being indexed by Google.

I'm simply lost at what has happend. Am I trying too hard to rank or is there a problem that I''m not seeing. Many of the pages have great grades and scores with MOZ, most in the 90's so the on-page SEO doesn't seem to be a problem. I'm wondering if there is something that I'm not seeing.

HELP!!!!

[RobertFisher]

In SEO much of what we do is an experiment. We think we KNOW things but we are experimenting. We look at a website page for kumquat vendor and realize the brand is before the keyword in the Title Tag and go "AHA!" as if changing that convention to the acceptable Kumquat Vendor | Kathy's Kumquats is going to shoot us to position one over all the other kumquat vendors who have the same exact thing wrong because we KNOW the answer. (My use of caps is not yelling; simply emphasis.)

If we think we are not experimenting I have one question: Why do we constantly hear others ask and often ask ourselves: Why is that site outranking mine when my SEO is of the highest and best variety? We don't ask this before we made the changes we ask it because we made the changes and nothing or little happened; thats experimentation followed by question and it's all good. But, it's still experimenting.

In science we learn that for any experiment we must be careful with endeavoring to test two variables at once. Yes, there are ways to do multivariate testing, but that does have limits. If I am understanding you, your client had 25 separate subdomains in each major UK city. bob.grass.co.uk, joe.grass.co.uk, wilbur.grass.co.uk, bethany.grass.co.uk, moz.grass.co.uk, smoke.grass.co.uk, etc. Each of these sites in some way ranked for terms related to artificial grass.

If the term is "best artificial grass" and in Manchester you had each one of the above subdomains ranked on page 1 and then you turn these into sub directories on the main site, your ability to rank each one of these as they did as subdomains is highly limited. Could you, yes, in a vacuum, but IMO not in reality on Google or Bing, etc. Subdomains are treated differently than subdirectories by search engines. My guess would be you thought if the subdomains all had some weight and you moved them to the main site that weight or the link juice would transfer; did you use 301 redirects for the change?

But the other issue is this is not dissimilar to what we used to do with "micro-sites." In highly competitive verticals we would have 50 websites that were all around the same item or service: dating, plumbing, attorneys, etc. Since search engines would not allow us to have 9 out of 10 positions on the SERP with one domain, we would just cheat and put up all these others as if they were not related and Voila! Then the world rotated on its axis and, bummer for us, our tactic no longer worked.

So, your best hope is to begin putting them back as they were and hope that you can continue to have each rank as if it were a separate site for a while. You might instead try to 301 the urls from half and see if that impacts the rankings for your main site. You have to either go back to exactly as it was OR you have to experiment, because when you change 25 variables it is nearly impossible to determine exactly what happened.

Hope that helps,

Robert

[EGOL]

but they now have many more pages being indexed by Google.

This is a comment that caught my attention but is not obvious during a quick look at the website.  Adding a lot of new pages can alter the ranking ability of the site.  If there is a really lot of pages and their quality is low, that can cause a Panda or other problem that can tank the rankings of the entire site.

[SeoSheikh]

Instead of the old subdomains, I created landing subfolder pages or "micro sites " which the sub domains are forwarded to. Then each micro site will have pages added to them to assist with local targeting such as testimonials, directions etc.

[EGOL]

Are these "micro sites" on different domains?

[SeoSheikh]

Thanks Robert, I think I will try reverting with some of them. I completely agree with you that we are ALL experimenting with our SEO efforts. There's so many SEOers that recommend switching from sub domains to sub directories, this is why I made the change.

Thanks for your detailed input.

[SeoSheikh]

No, they're just sub pages. So maindomain.com/uk/cityname1 etc...

[SeoSheikh]

See graph of indexed pages...

indexed-pages.png

[SeoSheikh]

You can see the changes in position almost switch over night. I honestly think that this is the best format for future roofing the website though. The old one was a completed mess, with very poor site navigation and architecture.

the-drop.png

[RobertFisher]

Personally, I am fully confused at what you have done. I suggest you slow down a minute and let's clarify what you had, what was done, and appropriate next moves. 
You said you created subfolders in one place which would be mysite.com/subdirectory/page-for-something.

Now you are saying you created micro sites? which would be microsite.com/page-for-something

What did you have first and what change did you make?

[SeoSheikh]

Apologies Robert. The owner of the company still calls the subfolder pages "Microsites" and it appears it has stuck in my brain!!!

There are no microsites, just subfolders (maincompany.com/uk/city-location/ which used to be the subdomains to house each major city franchise of the company. Back then when it was all subdomains (city-location.maincompany.com) had masses amounts of duplicate content as they had basically a Wordpress multisite and just duplicated each website but changed its target city keyword.

Am I making any sense here at all? Basically, it's all in an orderly fashion NOW, but the rankings have dropped.

[BradsDeals]

After reading through this entire thread, my big question is if the content for each city is still roughly the same thing or if all of them are substantially different after the move to subfolders. Because I agree that moving content off subdomains into subfolders SHOULD generally be a positive move... unless all of those subdomains were essentially identical, in which case you've just handed your root domain a big duplicate content issue.

[SeoSheikh]

Hi Rebecca,

Yes the content on each city is unique (not great, but unique). When they were on subdomains they were absolute replicas of each other apart from city name switches and maybe some short intro text, but other than that, complete copies.

Now those old subdomains get forwarded on to each new subfolder page so;

city-location.maincompany.com gets forwarded to maincompany.com/uk/city-location/

Alex