Welcome to the Q&A Forum

Peter is correct - your search, admin and user pages are common pages to block for Magento. What you block is up to you, though. Don't forget that a page that is blocked by robots.txt can still be found by search engines, so if it's a page that will contain private information you should protect it with a password.

Hi Garrett,

I agree with Dave - getting some inbound links and other press will help. Bing and Yahoo tend to rate keywords being present in the domain/URL more strongly than Google does, which may be why you're seeing it come up there. It looks like https://feello.com is now ranking on page 1 for "feello" - an incognito search I just did saw it at position 8, under the image carousel.

One thing you might want to do, if you haven't already, is use Organization markup in schema.org to try to improve Google's Knowledge Graph entry for the business. That will help Google understand that Feello is a branded term, and that that brand is associated with your website. You should also use the "sameAs" schema.org property to link to your LinkedIn, Twitter and Facebook pages, since Google is currently ranking those higher than your domain - that will help Google understand the relationships between those pages. Good luck!

Not sure how you'd cover this in cheat sheet form - but somethign on using JavaScript/AJAX for faceted navigation would be helpful.

Category pages are useful to help users browse deeper into your site; however, I doubt you are seeing significant SEO impact from not having your category folder in your product URLs. In general, it's usually better to keep a page at the same URL rather than to move it, and having a slightly more "SEO-friendly" URL wouldn't provide enough SEO benefit to be worth the risk and hassle that moving all of your product pages would take. I think it's fine to leave it how it is.

Hi Pashmina,

It looks to me like the problem is that Google doesn't yet understand that "Bar & Cocoa" is a brand in and of itself. Until they receive the signal that "Bar & cocoa" (or "bar and cocoa") is a different query intent than "bar cocoa", this will continue to be an issue. Focus on building your brand online; in particular, I recommend researching the Knowledge Graph to understand how Google catalogs entities. Some things to try:

• Roman's suggestion to implement Schema markup is a good one; I actually recommend implementing markup both inline and via JSON-LD in your header (you can also use Google Tag Manager to implement JSON-LD schema markup; check out this awesome post by Chris Goddard for more on how to do that: https://moz.com/blog/using-google-tag-manager-to-dynamically-generate-schema-org-json-ld-tags). Make sure you use Store markup on your home page to mark up your business name, logo, contact information, and other important details about your business, and mark up as much information about your products as you can.
• Spend some time doing some online-focused PR (that's public relations, not page rank); the goal should be to get mentions of your brand name alongside relevant words about your products (like "chocolate shop") in reputable news outlets. It would be ideal if this coverage resulted in links back to your site, but news coverage without a link will still provide something of a brand signal.
• Start planning your content calendar; think about what's unique about your products and create some content around it that you can then, hopefully, promote to earn links back to your site.
• Make sure you're behaving like a brand online in other ways. Build out your social media profiles, and start trying to build a following on e.g. Facebook and Twitter. With a highly visually-appealing product like chocolate, you might have success on Pinterest and Instagram. Make an effort to engage with your followers - don't just talk about yourself, but really start a conversation and respond when people talk to you. Having a robust, active social media presence sends search engines and users alike the signal that you are a real, reputable business. Claiming social media profiles under your brand name will also provide additional pages that use your brand name in conjunction with your business.

The good news is, all of these activities are worth doing to promote your business online anyway; the bad news is, they will probably take some time to really establish your brand. In the short term, it might be worth investing in some paid search ads to make sure your site shows up for your brand terms.

I can't speak to why the other websites are using data-href; it could be for better tracking, or it might just be a function of their CMS that their links are coded that way. For affiliate links, I strongly recommend using rel="nofollow", as that is the best way to disclose, as it were, to Google that these links are paid for, and not editorial. Using data-href may functionally serve the same purpose, in that data-href links typically don't pass PageRank (to my knowledge), but they don't carry the same weight in terms of demonstrating to Google that you're not trying to hide the nature of the links. It may not matter much one way or the other but rel=nofollow is the best practice.

Hi Matt,

I think setting up a 503 HTTP code on the holding page and then using 302 redirects to point all pages to that page is a viable option. You could also consider having every page return a 503 error (make sure your robots.txt page does, as that will keep the search engines from continuing to crawl). The pages on the site will most likely fall out of the index while you're returning a 503, but that's OK since there won't be anything for your users to find anyway.

The key here is to add a Retry-After header with the GMT date and time your site will be available. That lets Google know when to come back and that the site isn't actually down/returning a 503 forever. Yoast has a great post on this at http://yoast.com/http-503-site-maintenance-seo/ which I'd recommend checking out.

Hmm! Well, it could be a couple of things.

If your full crawl ends up crawling the whole site, it's possible that the preliminary crawl timed out on a request to your site. As long as you continue seeing the full crawl, you should be good to go.

If your next crawl continues showing just a few pages, I would check to make sure you're not blocking robots with either the meta robots tag, a robots.txt file, or nofollowed links. If you don't uncover anything out of the ordinary there, it could be an account- or campaign-specific error on our end; please email help@seomoz.org and our super-awesome Help Team can help get that fixed.

Canonicalizing all these URLs back to your main page will drastically reduce the total number of pages that will drive any kind of traffic. At the very least, I would look at which type of URL is attracting the most organic traffic (location, listing type, breed, etc) and work to make these pages real landing pages for these search terms - other than the listings, what other information could you provide on these pages that would make them more unique? Striving to make the pages more unique from each other will give you more long-tail opportunity than canonizing everything back to the main page.

Ultimately, I recommend you rework your navigation using a combination of facets and filters, in order to reduce duplicate content while still having mid-tail level landing pages. Hannah Smith wrote a great piece for Moz on this a couple of years ago that's still useful today: http://moz.com/blog/information-architecture-faceted-navigation-duplicate-content-oh-my

Hi Todd,

Using the Pragma no-cache tag means that no cached content from that page will be used when a user next accesses the page; the request will go back to the origin server instead.  It's useful when you have a frequently-updated page that you don't want users to end up seeing older versions of.  It's generally considered to be a bit of an outdated tag - you may want to explore using your HTTP headers to expire content instead - but won't hurt you with the search engines (or prevent the Cached version of your page from showing up in the SERPs).  Hope that helps!

Hi Oliver,

To clarify - you're not only seeing one result total, just one result from your own domain out of the results for that term, correct?

It is rare for Google to show more than 1 or 2 results from the same domain on the first page of results.  It sounds to me like you've got a whole bunch of pages on the same domain that are all cannibalizing each other to try to target the same keyword - instead, try targeting different keywords with each page.  If you want all of your hotels to be found when people search "Hotels in Bournemouth," creating one page that highlights all of the properties and targets that phrase would be a way to do it.

In terms of Google Places - properties showing up in the 7-pack doesn't have much to do with what's displaying in the regular organic results (Google certainly wouldn't exclude your Places pages just because you show up in the organic results).  I would recommend further optimizing each of your Places pages, and making sure that address/contact info is easily crawlable on your own domain as well.  Good luck!

effectdigital is correct. If you're blocking pages via robots.txt and still seeing them in the index, it's likely that Google is encountering links to these pages, and is indexing them that way, without updating its crawl (since your robots.txt says not to). Your best bet is to:

• Add a meta robots noindex tag to each of the pages you want removed from the index;
• Remove the disallow directive from robots.txt;
• Wait for Google to re-crawl the pages (using "fetch as googlebot" in GSC may speed this process along
• Once the pages are no longer in Google's index, re-add the disallow directive to your robots.txt file.

So that's really what I meant was "as long as you're seeing the full crawl," was that if your next crawl found the whole site it would probably be fine. It sounds like that's not the case, so I think the Help Team is your best bet - they're far more able to diagnose campaign-specific issues than I am

Custom Search Result Filters are specifically for Google Custom Search - they'll appear in internal search results on your site if you're using Custom Search, but not in the overall Google search results. So those are more for user experience than anything else. "Unspecified type" usually comes up when the structured data testing tool is misidentifying a bit of code as structured data, and can't figure out what this is. This can happen a lot when you are using link rel properties such as rel="alternate." You can usually just ignore it.

So neither of these are going to be of much help SEO-wise, but you probably want to keep the Custom Search result filters in place if you're using Google Custom Search on your site, since it helps you have better, richer custom search results.

Adding "noindex" to your page doesn't just keep link juice from flowing to it - it also means that it won't appear in the SERPs at all - what about your users who are actually trying to find your page? I wouldn't noindex anything unless you're sure that you don't want anyone to be able to find that page via organic search.

When you're structuring your site, it's better to focus on having a structure that allows search engines to crawl and find every page (no orphan pages, no content that search engines can't crawl, no duplicate content) and that allows users (you know, the people who give you money!) to easily find content.  If you want to conserve link juice on your pages, try to make sure you don't have a ton of links on every page, but don't try to nofollow/noindex content based solely on the idea of link juice - it won't work.

From everything you're saying, it looks like search engines will be able to crawl the content either way (as long as you make sure the dropdowns are using search-engine-friendly code like HTML or CSS and not JavaScript or Ajax # tags).

Once the site launches, you can double-check that the content's being indexed by doing a site: query in Google that contains a large chunk of the text contained in the dropdown.

Here's a video from Google Webmaster Tools confirming that the meta keywords tag is not used for ranking: http://googlewebmastercentral.blogspot.com/2009/09/google-does-not-use-keywords-meta-tag.html

The only thing anyone might use a meta keywords tag for these days is your competitors scraping your tags to see which keywords you're targeting on which page!

To Peter's point, I wouldn't worry too much about Twitter share counts since those are going away anyway. You mention what OSE says about links to the two versions of your sites, but what does Google Search Console say? Is it recognizing inbound links to your site?

If your URLs are redirecting to their correct versions, you're likely not losing too much link juice, and if you've made all the necessary changes in Search Console to set your preferred domain, Google should understand and count those links as all pointing to the same domain. OSE will continue to report on links to the www- and non-www versions separately because of how it reports on subdomains, but Google should get the picture loud and clear.

The biggest thing you can do to show Google "this is the real version of my site" is to start building new, high-quality inbound links. Over time, the volume of inbound inks to the preferred version will overtake the volume of links to the old version, which will reinforce for Google what the correct version is.

Law and real estate are two good examples where it can actually make sense to have the kind of state-based content you're talking about; since practitioners are only licensed on a state-by-state basis, and the ins and outs of the practice do vary based on the state you're in, having different content for the different states in which you're licensed to practice makes sense. The approach you've described is how I would advise someone to go about this: create content that is highly unique to each state and its individual practices, make sure it would be useful to people searching for info about that state, and avoid e.g. a find-and-replace approach where you're switching out the state in boilerplate text.

Nothing you're describing sounds like it would qualify as duplicate content to me. It's important to note that duplicate content isn't something that would garner you an actual penalty from Google; the risk you run is that pages will compete with each other in search, or that Google will remove duplicate pages from the index or not rank them well for anything. It doesn't sound like that's the case here. I could see the approach you've described possibly incurring some risk for a doorway page penalty, in which Google penalizes sites that create a ton of different thin-content pages, each optimized for a different single keyword or phrase, that all funnel users to the same place. From what you describe, though, it doesn't sound like what you're doing qualifies as creating doorway pages.

I don't share your SEO consultant's concerns about crawlability. 10 pages per type of real estate is still only what, 100 pages? 200 pages? Google can easily crawl sites of many times that size. You may want to take a look at your site's log files, and at your crawl stats in Google Search Console, to verify that Google isn't having difficulty crawling your site, but since it sounds like your content is ranking well and getting traffic, I doubt that's the case. You'll want to be intentional about how you're approaching site architecture, making sure your navigation is intuitive and that you're linking internally from page to page well. You may want to conduct an internal link audit to see how site pages are passing internal link equity to each other; JR Oakes has a guide on how to do so at https://github.com/MLTSEO/MLTS/blob/master/Tutorials/PageRankandCheiRank.ipynb.

Since I haven't seen what your SEO consultant is seeing, it's entirely possible that there are additional factors in play that are influencing their advice, but from what you've described above, I don't think anything you're doing risks a Google penalty. I wouldn't e.g. create 10 blog posts for every topic you write about - that would be overkill - but if it's just your core site pages, talking about what services are available in different states, I can't see why that would be an issue.

If you don't want them to appear in the index but still want them on the site for user reasons, noindex, follow is one way to do it - you could also use the rel=canonical tag to call out the original post as the canonical one.