How to tell when a directory backlink or other backlink is worthy of disavow tool? Especially when a keyword is not ranking passed where it should.

[Deacyde]

Hello,

I jumped aboard as SEO for a client, who seems to of had been hit by panda and penguin back in 2012 of April, the panda part I feel I've fixed by creating better content, combining pages that were same topic into one, basically creating a better content experience that relates better to search terms users are searching for.

Once the site was redesigned and relaunched all keywords improved minus one, the main keyword they want to rank for. Created a landing page for it, that is very nicely optimized for that keyword and it's brothers and sisters, however that page isn't used by google since it's brand new with a PA of 1.

Doing a backlink audit I found 102 links out of 400 using the same anchor text as the keyword they want ranked for, they also have synonyms anchor text for other links too but not quite as much.

Most of those 102 domains using the main keyword anchor text are directories, in my opinion I'd declare all of them spam, however there are a few with DAs higher than 50, making me little more nervous to disavow, since I want to make sure we get out of the penalty if we were hit by penguin but also don't want to ruin the ranking for other keywords we're doing better with, since they are longtails and short, but very relevant to users.

How is the best way to determine if a site / directory is spammy enough that it's penalizing you and how could I approach the anchor text issue with backlinks? 99% of these links I cannot have changed, since they're directories I doubt many have had a human mess with them in a while.

Sidenote* If you're going to post a link as a response, try to summarize what that link will be about, as many times links are giving as an answer but end up not really providing the meat we were seeking.

Thank you!

[Deacyde]

Hello Robert,

Hope you're doing well,

Seems I just didn't include all the info I needed, making posts as if you all have access to the data I have in my head is a mistake

What I meant is I have 500 urls, after doing a full link audit, 153 of those urls are trashy directory or comment spam backlinks with very spammy site templates, also they often times not had anchor text with main keywords of ours.

Now yes, 153 out of 500 urls isn't that bad, however the 153 urls are more than 3 years old, and our total backlinks has grown from 160 ( a year ago ) to 500 ( today ) meaning at one point in time, we did have a majority of backlinks that were from trashy directories and comment spam, in fact last year ( 160 - 153 left us with 7 possibly ok backlinks )  was the most recent, so it's a very high possibility these urls were the reason we got hit by penguin ( have data that suggest this as well ).

It's easy to get fixated on one SEO keypoint or another, but when you do, just remember to follow the data trail and make sure to see the other SEO footprints as well during that search, you can find a good bit.

Even answering questions from others here helps me find further keypoints to highlight and go through.

[RobertFisher]

I don't believe you can just say we have x links from directories of which __% are no followed and __% from followed use KW as anchor text; therefore, the links are the problem. It is simply not that simplistic.

Secondarlily, in your original comments you said this: "...however that page isn't used by google since it's brand new with a PA of 1." I can show you tons of pages with PA of 1 that rank on page one of Google.

I caution always that you should not get fixated on one issue with SEO as it can lead you down a rabbit hole. I do not think in this instance based on what you have here, that you are suffering due to the links. Frankly, you are looking at a page with a PA of 1 and then pointing to a site with 400plus links and saying the other pages are increasing in rank. The anchor text being what it is, I don't see how it is affecting this one page.

Best

[Deacyde]

Thanks for the reply, I agree with your points but also felt I didn't explain the rest of the backlinks quality that well either.

they have around 450 backlinks, 102 use that keyword we want as main as anchor text, the rest of the links aren't all good either.

Most of the 411 backlinks are nofollow it seems, with around 150 of those links being follow. Many of the follow links are under 50 DA and majority of those are directory listings or comment links.

This is why I think they've been hit, since the the backlink anchor text for follow links are using our main keyword and aren't to very high quality sites.

Couple that info with the fact I've had google index that one landing page 3 separate times since our redesign launch a month ago and only that main keyword fails to gain rank, It did make a big jump when we launched the redesign from 72 to 58 then to 41 but 2 days after ranking up to 41 it dropped back to 62 where it has stayed for the week, moving up to 64 and holding.

This is while every other keyword has improved by 20+ ranks, just this one keyword isn't going anywhere, I have around 160 more links to define the DA, follow or nofollow and anchor text if follow, then I will be grouping those follow links into a best to worst list and disavow 10 of the worst urls at a time every week to see how things go.

What do you think?

[RobertFisher]

Deacyde

First you said, "however that page isn't used by google since it's brand new with a PA of 1. " I suggest using Fetch as Google and then once it shows as complete, click submit to index.

As to when is a specific link or group of links worthy of disavowal tool, that is a bit broader answer. You have roughly 400 links and roughly 100 or 25% are spammy directory links. IMO you likely are not being bothered by penguin IF the other links are at least reasonable. Also, with rare exception, most directory links are not doing you much good so losing them is not going to do much, if any damage. Think of it this way: Your client is a plumber, doctor, lumber store, etc. On a directory most of them have low value pages and then link out 20 to 100's of times. I have seen very few even higher value directories that are not either sending out a ton of links per page OR blowing it by using sitewide links, etc.

I think you can always experiment. Do the disavowal and leave in the few you think are higher value. For each of the questionable directory domains disavow the whole domain to save time assuming they have multiple links. Also, a strong caution here is no matter what link tool you used, also go look in GWMT (search console) to be sure there aren't any really dastardly ones you have missed  as many directories now block ahrefs, majestic, moz, etc. 

Once you do that if you don't gain traction, resubmit adding additional "good" directories to the disavowal. With what you have as a fairly low number the disavowal should be fairly straightforward and easy to get done. Nothing like dealing with a few hundred thousand bad links!

Let us know if this helps,

Robert