Google changing page titles and description
0:50 “[…] I’ve got one question about page title and description. […] a few days ago, we optimized our page title and description. […] after […] changing, we see the title and description changed when using the site to search in the Google. But after a longer while, the title and description has been back to the one in the first place. […] does Google think the former title and description are better than the one we optimized, that’s why it changes, or any other possible reason that may cause this?”
John replied, “I wouldn’t necessarily assume that if Google changes it to something that Google thinks it’s better, and that you should use that title too. It’s more that our systems have selected a different title.
And usually, the title is picked on a per-page basis. So that would feel a little bit weird if we show one title first and then shift it to a different title later on. It might also just be that we have shifted to an older version of the page for indexing and processing in general. And then we’ll shift back to the new version as soon as we can process that properly again.
So from that point of view, I wouldn’t necessarily see it as a sign that what you did was not perceived as being good by Google. But it’s just being shown like that. And one of the things you can do, specifically with titles because they’re based on the page itself now and not on the query, is you can submit that page for reindexing and reprocessing.
And then pretty much as soon as that is reprocessed, you can look at the title with a site query to see, what is the current title? And you can use that process to fine-tune the titles that you give on your pages and see, how does Google actually show the title? Does it change the words? Does it separate or make a difference to parentheses or brackets make a difference? And essentially, optimize the title for what you would like to have shown for that specific page.
And since you’re working on a larger site, probably you have templates that are across lots of different pages. Sometimes you can update those templates as well to use the same things that you learned from optimizing a single page title […]. So I think in a situation like this, what I would do is try to see, is Google maybe accidentally switching back to an older version of a page in general? And will it switch back to the new version to separate out the aspect of the page title changing and then going back? And then really try on a per-page basis to figure out, what is the best way for you to show titles so that they can be picked up and shown in search.”
What to do when your articles are considered duplicate content
15:44 “First off, we have a news publishing website since 2009. We post articles related to recipes, health, fitness, and stuff like that. […] we have articles that are considered per an SEO tool as duplicate content. But what we tend to do is recreate another version of this recipe […] or maybe tweak around maybe sugar-free or salt-free or everything related to that. So it’s still considered duplicate content. What this SEO tool suggested is that we remove it, because none of the duplicate content are being ranked or indexed by Google. So what is the solution for this?”
John said, “[…] I would be cautious about blindly following any SEO tool just as something to start with, in that SEO tools tend to make assumptions with regards to what Google will do and what will happen. And sometimes those assumptions are OK, and sometimes they’re not correct. And I think taking this kind of feedback from SEO tools is useful, because it’s still something where you can take a look at it. […] you have to make a judgment call. […] you might choose to say, well, I’m ignoring the tool in this case. And I’m […] following the guidance in a different case. So just to start with, if you’re seeing something even from a very popular SEO tool that tells you, oh, you should disavow these links, or you should delete this content, always use your judgment first before blindly following that.
Take advantage of link risk management to reevaluate your backlink strategy.
Take advantage of link risk management to reevaluate your backlink strategy.
With regards to duplicate content like that […] it wouldn’t be duplicate per se because it’s slightly different content, it’s something where you always need to consider the strategic aspect of how you want your pages to be shown and what the competitive environment is like in the search results. And sometimes it makes sense to have one article split up into different variations because these individual variations are unique enough that people are actually searching for them. And sometimes it makes sense to combine an article and say, well, I have three or four variations. I will make one page out of it. And usually, what happens there is that one page is a little bit stronger in the search results. So that means if it’s a very competitive environment, having fewer pages that are stronger is usually a good idea. Whereas if you’re already doing really well in those kind of queries, then maybe sometimes having multiple variations is an option to better target the specific intents that people have. And that’s not something where I would say, there’s always one clear answer. You should have fewer pages, or you should have more pages. It’s really something where if you see this kind of situation popping up, you can make a judgment call. And it’s more, almost like a strategic decision rather than an SEO decision […].
So I think in a case like this, I would check to see, what are the queries leading up to these pages? Are you even getting enough visibility for these pages individually? And if you’re seeing that they’re ranking very badly at the moment, then maybe it makes sense to combine them. Maybe it’s also the case that the competitive environment for those queries that you’re targeting is so strong that even if you combine it, it doesn’t make a difference. So it’s like, yeah, you have to look at all of those sides.”
Google Search Console notified you about indexing problems due to duplicate content? Read our guides and fix:
- the “Duplicate, Google chose different canonical than user” status,
- the “Duplicate without user-selected canonical” status.
Ranking on international markets
23:14 “ Currently, I’m working on a website that is based in India and we get leads from all over India. As we can provide our services all over the world […] I want to rank my service pages in USA to get more leads from there. So can you help me know what […] I can do so that I can rank on top of my competitors that are […] in USA?”
John replied, “It’s hard to say without knowing your website […].
So usually, the thing that I think you need to watch out for is, on the one hand, geotargeting. In that if you’re going from a country-specific website to something that is more global, then it helps to make sure that from a technical point of view, your website is available for that. So in particular, something like using a generic top-level domain instead of a country-specific top-level domain. That can help.
The other thing to also keep in mind is that any time when you go from a country-level website to a global-level website is that the competition just changes completely. And it might be the case that, for example, in India, you’re ranking really well for these queries. But in the US, the competition is just so much stronger that even that “really well” level from India doesn’t translate to something that is reasonable in the US. And that […] is not something that you can just fix with technical issues. You really have to work on that for the long run to make sure that your website is actually much, much stronger and that people see it also as something […] relevant for them in their country. And I don’t think there’s one technical trick that you can do to make that happen”
If you’re looking to internationalize your business, check out our Ultimate Guide to International SEO.
How to deal with subdomains
27:18 “We’re working with an eCommerce client. Their eCommerce is on open source, which is an online store management system. […] their blog is on WordPress. The main URL is example.com, whereas the blog is blog.example.com. What would be the best approach for this client to get credit from the blogs?”
John said, “So they already get credit from the blogs in the sense that probably these sites are cross-linked between each other. And that’s kind of what we’re looking for to make sure that the two aspects are linked between each other so that we can pass signals from one side to the other.
From our point of view, we don’t say that subdomains or subdirectories are preferred in any particular way. Some SEOs have very strong opinions about subdomains and subdirectories and would probably want to put this all on the same domain. But from our point of view, you could do it like this as well. […] from our point of view, this setup would be fine. If you did want to move it into the same domain, then practically speaking, that usually means you have to do some technical tricks, where essentially you proxy one subdomain as a subdirectory somewhere else. And you have to make sure that all of the links work and all of that. Sometimes that’s quite complicated. So that’s one thing to watch out for.
The other thing I’d say is that sometimes eCommerce systems also have some limited page creation ability as well. I don’t know in this specific case if that’s true, but sometimes being able to create individual pages within the eCommerce site is something that you can do. And sometimes, that’s enough for the basic-level blogging that people do on an eCommerce site. It’s definitely not the case that we would say WordPress blogs are the best, and they will always rank number one when it comes to blogs. So if you can create the same content within a more limited eCommerce text page creation system, then probably that’s just as good. So that’s, I think, another option that might work out for you.”
36:42 “ I have images with some text on them. In that case, should I use alt text?”
John: “I think ideally, if you have text and images, it probably makes sense to have the text directly on the page itself. Nowadays, there are lots of ways to creatively display text across a website. […] I wouldn’t necessarily try to use text and images and then use the alt text as a way to help with that. I think the alt text is a great way to help with that, but ideally, it’s better to avoid having text and images.
And the question goes on, should I write alt text for products for an eCommerce site since there’s already text beneath the product that describes the product?
So from a more general point of view, the alt text is meant as a replacement or description of the image. And that’s something that is particularly useful for people who can’t look at individual images, who use things like screen readers. But it also helps search engines to understand what this image is about. And if you already have the same description for product around the image, then for search engines, we have what we need. But for people with screen readers, maybe it still makes sense to have some kind of alt text for that specific image. And in a case like this, I would avoid the situation where you’re just repeating the same thing over and over. So avoid having a title of a product be used as an alt text for the image, but rather describe the image in a slightly different way. […] that’s the recommendation I would have there. I wouldn’t just blindly copy and paste the same text that you already have on a page as an alt text for an image. Because that doesn’t really help search engines and it doesn’t really help people who rely on screen readers.”
When should the last modified date be updated?
41:34 “[…] our CMS automatically updates the last modified date for every change even when adding a few words, fixing a spelling mistake, or adding an internal link. Is this a good SEO practice, or should the last modified date only be updated when there is substantial changes to the content?”
John: “[…] I don’t think from an SEO point of view it matters much. […] when we look at things like sitemap files, where there’s also a last modified date, the sitemap team ideally wants to see that date updated whenever there are any changes on the page when it makes sense for us to recall that page. And that includes things like a random link in the sidebar, for example. So from that point of view, people do it in different ways. And it should just work out. With regards to the last modified date in the structured data itself, usually I think it makes sense to have something aligned with when the primary content actually changes. But if they’re for technical reasons why you have to do that immediately or why your CMS does it immediately, I don’t see a problem with that. When it comes to dates on the page that we sometimes show in a snippet, what happens there is we try to find multiple mentions of that date on the page as a confirmation. And then we use that as something where we say, well, this date is probably associated with this page. And it’s fairly certain that it’s a reasonable date. And if we were to show a date in the snippet, we would probably choose that one. We don’t show a date in the snippet for all pages. And it’s really something where you can help to guide that as well by making sure that maybe there’s a visible date on the page as well. So if you have a news article, for example, and there’s a visible date on the page and the sitemap file has a slightly different date, then probably we will give that visible date a little bit more weight. So that might be another way that you can control that a little bit.”
Getting background image indexed
53:54 “So we have, typical on home pages, a hero image. And the hero image is embedded not with the image tag but rather as a background image. With that, we’re lacking the possibility to add an alt attribute to describe the image. There are maybe some technical aspects why the agency has done that approach for us. But I was wondering, what is the recommendation if we want to have the image indexed? So be available on the image search on Google, but also to understand, how is Google judging if the image is embedded as a background image, be relevant for the page? It also detects that we can’t put behind the image.”
John: “ Now, so I think for the most part, if you’re using something like a CSS background attribute to display an image, then we would not pick that up for image search. It might be that that has changed in the meantime. But at least in the past, it […] we only focused on the image tag, the source attribute for the image tag. And if you had a picture element, then that’s still kind of within the same thing. And the other option was if you link to an image file directly, then we would also be able to pick that up for image search. But if it’s purely a CSS background image, then we, at least in the past, did not take that up. And it sounds like that’s still the case. So it’s tricky for me to say how you can best embed that. One thing you could potentially do is make sure that it’s at least linked on the page itself if you care about it for image search. It is also something where sometimes it’s not critical for that image to be findable in image search. So if you have something like a decorative image of, I don’t know, your company’s headquarters or something, then in image search, it’s going to be rare that someone actually searches for that. So that’s something where you also have to make a judgment call of, well, technically if like we want traffic from image search, this is the way to do it. But in practice that particular traffic from image search is not critical for our website so we don’t necessarily care about that.”
56:48 “OK, I understood that one. And about not being able to describe– I mean, the alt attribute and the title attribute could potentially be used to describe the image. Would the title be enough? We’re probably using a div tag, and then we have CSS. We style it. Would the title be enough to describe the image so that Google understands, OK, this is the image that we have? Or we have something here, and the title would give more context, OK what’s here?”
John: “Yeah. So for image search, we use a number of different things. The alt attribute is the most straightforward one, but we also look at the text around the image itself. We look at things like the title tag if you have a link to an image file. So things like the headings on the page, the captions below the image, even that will be taken into account for image search. So title attribute is usually the easiest way to give that information, but it’s not the only thing that we use.”