Are all link issues Penguin related?

[almo136]

We have had an issue with our organic traffic for over a year now. Our organic is at about 25% of what it was.

I got a few messages from Gary Illyes from Google and he said we should take a look at our backlinks as "some of our algorithms seem to trigger because of some weird links". Once cleaned up we should give it some time.

We did this and cleaned up anything that looked remotely suspicious by adding it to our disavow. The problem is our organic traffic has not improved at all.

My question is if links were the problem (we didn't recieve any manual warnings) does it mean it would have been penguin related? Therefore we need to wait for this to be refreshed.

Or can link penalties (drop in origanic traffic) be related to the core algorithm that is running all of the time.

I am trying to figure out if we just need to wait now or if we need to do even more cleaning up.

[anneconq]

How about your strategies or techniques?

Your target keywords how's the performance, are you targeting the right keywords?

[Wilburforce]

if links were the problem (we didn't recieve any manual warnings) does it mean it would have been penguin related?

Possibly, but

can link penalties (drop in origanic traffic) be related to the core algorithm that is running all of the time.

The algorithm is under constant development and refinement, so that is also possible.

You need to do two things:

1. Monitor your backlinks frequently, and update your disavow file with anything that looks suspect. Particularly suspect are large numbers of exact key-term links.
2. Wait.

There isn't any clear evidence that frequent changes to the disavow file have any negative effect, but I would personally advise against repeatedly adding and removing any particular site or page.

If the problem is Penguin-related, you may not see any effect until the next Penguin re-run at the earliest. However, Penguin is a weighting, not a penalty, so can be at least partially offset by other factors (and other factors can add to its effect, or create negative effects of their own). Don't confine your attention to backlinks: keep looking at the whole site.

[martinibuster]

The following are my opinions, not statements of facts. It seems there are some spammy practices that appear to be going unpunished but in fact there may have been an algorithmic action that devalued their links. Furthermore, these spammy tactics are widespread and there is a mistaken assumption that they're Google friendly.

Near as I could tell one competitor site I looked at was ranking because of other quality inbounds. Another competitor site I reviewed wasn't ranking at all. The point of which, for the first site that is still ranking is that it appears that the spammy links were devalued and but the site was allowed to rank based on the remaining backlinks (no punitive action for spam other than a devaluation of the spammy links). Thus the first site keeps ranking despite their spammy link building. The second site never had quality links so they simply don't rank.

That second site fits into the category of site I call, "Ranking where it's supposed to rank." That means that once low quality links are discovered and devalued the site is ranked where it should be ranking. The solution is to get to where the first example is, to obtain better quality links.

The problem as I see it is a mistaken notion of what a quality link is. It has close to zero to do with third party metrics like DA. A link can have low DA and still be a high quality link. I know this to be true because I have ranked websites with low DA inbound links. In my opinion, relying on third party metrics for your business decisions may not be efficacious for ranking a site. Take the time to truly understand what makes a link quality. Quality has never been accurately revealed by Google's PageRank on the toolbar. I suspect the same is true for third party metrics.

There are other kinds of confusion that enters, such as the use of semantic relatedness as a measure of whether is a link is useful. Finding semantic relatedness in a link is sometimes a game of self-deception and wishful thinking.

[tangor]

My question is if links were the problem (we didn't recieve any manual warnings) does it mean it would have been penguin related?

Maybe ... maybe not. As noted above, Penguin does not run full time. How old is the site? Are the links the same as they have always been?

I ran into a situation were an older site (six+ years) had links, both internal in external, that were built using 1990's thinking (keywords stuffed everywhere!). While no penalties had been applied the traffic was disintegrating. Went through, cleaned up all the links to something more normal, and three months later the site began to recover..

[Planet13]

@ almo136

I don't understand this:

"My question is if links were the problem (we didn't receive any manual warnings)..."

And then this;

"I got a few messages from Gary Illyes from Google..."

Do you mind how you got Gary Illyes to contact you? usually googlers don't just reach out to people affected by Penguin.

Was this on the google product forums?

[Linxv5]

It was a gradual traffic decline over the year?

[almo136]

To answer a few of the questions:

- The site is about 6 years old.
- Around 2012 we received a warning in GWT about links. These were created by an SEO company we had hired but were no longer working with. We cleaned everything up, added a disavow file and filed a reconsideration request and got the warning removed.
- From that point forward we kept an eye on backlinks and added to the disavow periodically.
- Organic traffic hasn't been a steady decline. We noticed a sudden drop in October 2014, one in January 2015 and one in March 2015.
- After the first drop we checked all links thoroughly and added to the disavow again. Most of these were really old links that I guess slipped through previously.

Interestingly I got talking to Gary Illyes on Google+ in July 2015. I asked him if we had a Panda Penalty (we thought this may be the case as we were confident backlinks weren't an issue). He confirmed we didn't have a Panda Penalty at that time. He said he would need to speak to the ranking team to find out the cause of the issues.

He then got back and said "We looked into it and we could fine tune something that, if I'm not mistaken, increased the traffic to the site 2-3x. For the links issue, you will have to wait it out. If you cleaned up your links, then you will see its effect "

Sure enough our organic traffic did get a boost after this but still only back to about 25-30% of what it was.

There have been no signifinant increases/decrease since then.

[JS_Harris]

Try this: Post a quality article to your site and IMMEDIATELY go to your webmaster tools account and "fetch as Google" and then submit it to Google search. Next immediately go to Google, make sure you are logged out, and search for the exact match title. Your page will be in the results within moments. (DO NOT perform a search before you submit the article! If you do you're going to get a cached version of your search for up to an hour afterwards)

When your page is shown in results after doing this it hasn't been fully evaluated yet and hasn't had time to "settle" and one of several things will happen. If, for the exact match title, it ranks within the top 10 the article should settle somewhere in the top 20 based on what it deserves after evaluation. An instant top 10 is indicative that the page is approved for that title.

If, on the other hand, you find the article ranked in the 50-100 range, or the 500 range, that article is highly unlikely to move into the top 10 when it settles. The article is instantly not seen as worthy for that title. There are exceptions to the above but they don't happen often. There are no guarantees but checking your pages like this works as a barometer of sorts, just don't go changing titles or text immediately after submitting the article or it will get knocked down in rank quickly, it's what Google does to articles that seem to change for SEO reasons.

You get 500 such submissions per month, seeing what Google instantly feels before finding any incoming links or having any traffic or social metrics is fun. It also reveals which site(s) up top of serps are untouchable for any given title. Also, I fully expect Google to be deceptive with this and to rank some pages well that don't deserve it or to rank some pages poorly that deserve better and to give both an adjustment period where they EVENTUALLY end up where they belong... but only by a few spots, not by 50-500 spots.

Tip: If your article is informational but all the top results for a title are shopping related you can remove that article until you change it sufficiently. Why? Because Google is compartmentalizing results and informational pages rarely return for shopping terms, they languish at a low rank with big brands up top. The reverse is true, big brands struggle to rank for informational requests more often than not. There are many such "compartmentalized mini groups" but the informational vs shopping is perhaps one of the biggest dividers.

Important: Quality pages rank for a wide range of keywords, the more comprehensive the article the better the exposure it is likely to receive. The above only seems to work for exact match titles, not individual keyword searches, and given what I just said I don't recommend you look up specific keywords and then change things just for that keyword or keyword phrase... you've got to take a whole page approach in 2016.

The above should tell you pretty quickly if Google doesn't trust your domain. I'm a firm believer that what other people put on their domains should not be any concern of yours, especially if it's taking up your resources/time to track them down when you could otherwise be improving your site. Hopefully with RankBrain and machine learning now being active on all searches we're going to see Google's reliance on backlink metrics expire.

[JS_Harris]

Short version: RankBrain is going to eat Penguins and Pandas soon, once it incorporates what they do into it's own algorithm and becomes RankBorg.

[tangor]

That's ... well .... rank! :)

Yet again the google webmaster will jump through hoops. (sigh)