Poorly Performing Content - Should I Deindex it?

[Vantelli]

Hey guys,

I have a blog with 350 posts and some of them attract a lot of organic traffic, but many of my posts are stuck at the bottom of SERP and barely generate any visits.

Everything was fine until the mid of March and since then there is less and less visitors every week. When the site was on it's peak in early March there was 6000 visits a day and yesterday 3000 only. I'm trying to figure out why my traffic is slowly dropping down and I'm thinking to create a list of posts which perform poorly and to deindex them.

Do you think that would help or I can't improve my site on this way?

Thanks

[getcooking]

I'd suggest analyzing those posts to look for ways to improve them as a first step, rather than noindexing them. Make sure they are useful and thorough. Make sure they are well written (no spelling errors or grammatical issues) - hire someone if you need to or at least have a second set of eyes look at them. Small improvements can make a lot of difference. That doesn't mean only do small improvements, but content can have minor issues that will have a big impact ranking-wise.

I would use noindexing sparingly to start if you do go that route or you could harm your traffic even more (unfortunately I speak from experience).

[Vantelli]

I agree with you, but I really can't figure out what's wrong with those posts with little or no traffic. They're written by the same guys, are not different from posts which are ranked in the top 10 in Google. I cheeked for duplicate content and everything and nothing seems to be bad/different from other posts.

Right now it looks just like a lottery. I blast my site with several articles and some of them perform great, others are waste of time and money. There are no rules (or I can't figure them out) at all.

I've selected articles with 30 and less visits from Google in the last 30 days. Will noindex them to see what's going to happen.

[JS_Harris]

Sorry, I have to say stop before you do irreperable harm.

The approach you describe is often used by webmasters to clean up a site but I have to point out there is a rather lengthy process to go through before using noindex. You've skipped over all of it.

For starters Google has a patented system of ranking articles in positions they will not ultimately settle in, over time the articles will migrate, perhaps even in the wrong direction, before settling. Also, articles are all interconnected and even if one does not pull direct traffic it's part of the whole and supports the others, or hampers them but some of that is expected etc.

A better idea, look at each page from a visitor's perspective and ask yourself how useful the page is. Next ask yourself how different the page is from other pages on your site. If multiple pages are very helpful in solving search visitors actual queries but overlap a great deal, such as being overly specific about brand or color or location etc when generic is enough, then consider condensing these into one.

If your site basically repeats what is available on other sites, even in your own words, then you can expect less initial traffic until it is deemed authoritative enough to gain more. You've got a whole lot of different factors to look at in between from internal link structure to duplicate content issues(ie, snippet on index, in categories, under multiple tags, on page etc) to a lot of other things to go over first, imo.

You can't expect all of your pages to be heavy hitters either, in fact you will likely find that for each heavy hitter you will have many that are not, it's to be expected. Support pages are taken into consideration, a good widget fixing guide will have other pages related to different aspects of the fix, for example.

Put away the axe and dig into other things first, you don't want to be sending a message that a lot of your pages are junk either.

[aristotle]

You should look at the site as a whole. Do all the pages fit together, and complement and support each other? Pages that don't fit in could actually hurt the site by confusing the algorithm's evaluation of the theme and/or purpose of the site.

[keyplyr]

Instead of noindexing pages, there's always the option to lower the priority of the page(s) in sitemap.xml, which suggests to SEs that the page(s) have less importance compared to the priority level set for other pages: [sitemaps.org...]

[Totalx]

I would go back and see what you did before March. Also, see if you recently received any bad links.

[rdbseo]

I agree with what everyone is saying but I thought I'd jump in and share my experience with this.

We got hit hard by the updates in March/April and we lost a lot of visibility and leads as a result. After rumours of the updates being around quality we took a step back and did a full audit of the site, we decided to categorise all our content into three different pools - update, combine and delete.

We updated any post that was highly relevant to what we do.

We combined any posts into one bit guide so if we had 3/4 posts on the same topic we combined everything, updated the content to make it s$%£t hot and republished.

Finally we no-indexed any post that wasn't relevant or couldn't be updated, we deleted and 301 redirected some posts where it seemed logical to do so.

All in all we overhauled over 200 pieces of content and it's done our site wonders. We recovered in May and saw our visibility return better than what it was previously. Articles that were never ranking are now ranking. It was a good opportunity to make sure all then on page SEO was done properly (I joined after the two writers).

I don't think no indexing everything is the answer but consider updating, and combining stuff. Use it as an opportunity to overhaul your site.

[JS_Harris]

That's an excellent way of taking inventory, rdbseo. Instead of focusing on traffic numbers you looked at it more from a human point of view and made the pages better(or removed them). All webmasters should do such site cleanup regularly. Better content > more content.

It's sometimes hard for a webmaster to remain unbiased about his/her content quality. Can I do better than this? Is there better out there than this? Being as honest as possible goes a long way in self evaluations. I'm glad it worked out for you.

[tangor]

This is an editing question. All authors think their word children should live, and sadly, in the real world, that is not the case. They are there, but die on the vine with no friends (visitors). Those children can be culled, reborn, educated (pick your analogy) and become upstanding and loved. A website is a living entity, or is should be. Evolution is what moves forward, static is the past.

I can't recall how many times I've gone through sites and deleted, combined, expanded ..,. and always with the user and what they want in mind. INFO sites don't face this problem quite as often as ecom. Some info is evergreen and it is and there's no need to change it ... but the front end, HOW things are found on the site can be excited time and time again as the audience evolves ... as it always does.

NOINDEX won't change things much as the SE never forgets a url it has met. That part is done and over with the first time it is indexed and before you decide to noindex. Mitigation can be done, but that doesn't change much.

There's the OTHER possibility that the page is perfectly sound, just not popular. In the scheme of things (the site) it might be quite reasonable for the page to exist... just will not have that many hits. In the over all scheme of things not all pages are equal, if they were they'd all be #1 in the serps (as if that will ever happen!)

Don't over think this, and also don't have great expectations for every page. Do the review (as suggested above). If the page has a reason to exist, keep it ... after all no skin off nose if it is not popular and MIGHT, from time to time, if monetized, pull in a half cent or three. Otherwise, edit (combine or expand). Que sera sera.

[DelyStain]

What about keywords? Did you research them for each article? Maybe you overoptimized a page? What keyword density have you used? Relevant pictures and videos on a page?

[rdbseo]

Great reply @tangor

Worth noting... we de-indexed as we didn't want to have a load of re-directs pointing to say /blog and up the bounce rate. No indexing posts that were getting a load of traffic but served no purpose has lowered our BR too!

In a few months once they've 100% dropped out of the SERPS we will delete. We're tying this in with a blog migration from WP to our own CMS.

Another tip - take full advantage of Fetch as Google in Webmaster Tools :)

Good luck!

[mboydnv]

We did this just after fred update. Removed 22 posts. Deindex from google those posts and strengthened some others. We lost 70% of our traffic and have never recovered.

The posts we removed were horrible written by a terrible travel writer. I refuse to bring them back. We bumped up our guide content. ANywho it crippled us. We are pretty much done. No way back. They have us on some weird filter that no matter what we do, we head to the gallows. =(

[seoskunk]

mboydnv ever thought of rewriting the posts as quality content and then bringing them back?

[tangor]

In some niches, no matter what you do, thin content (or poor content) will never perform. If you are currently RANKING AT ALL think twice about fiddling with it in the chase for more because it will be treated as "new", "start at zero", and the performance will be even worse if it is in a niche that g wants to play (travel, transportation, lodgings, et al). Same goes for ecom (pricing, comparison, etc).

LOOK AT YOUR CONTENT from the USER'S viewpoint. If it works, keep it, and don't worry. Else, if if makes sense, combine/expand as REVISED (not "new") and keep that working.