Should We Be Afraid of Google’s Core Web Vitals?
Let’s figure out this whole Google’s Core Web Vitals mess that we’re going to have to deal with here. Should we be scared?
(00:02:50) No, actually, you should not be scared. I think actually everyone should see this as an opportunity. If you’re on this call with us, you know that this is important and you’re going to have a competitive advantage because you’re ahead of the curve. We still have time. This is an opportunity to basically improve your site in a way that will help you with your search rankings. I think instead of being scared, it’s better to see it as an opportunity.
What Are Core Web Vitals?
(00:03:35) Here’s the thing. The reason Google came up with these Core Web Vitals is site owners were tired of having performance gurus tell them what they needed to measure about quality performance. Instead, Google’s decided, we’re going to do this for you and try to dumb it down as much as we can. The Web Vitals Initiative, as it was called by Google, is an attempt to make this landscape less murky, a little simpler for everyone on the call. That’s hopefully what we’re going to do, freebase, kind of break it down.
Our goal is to allow you to focus on what Google considers the most critical metrics, these Core Web Vitals at least in their opinion, that you can track and optimize over time because that’s what we want to do. We want to have a benchmark and we want to show what those benchmarks are and then we want to try to get as close as we can to those benchmarks. Google introduced these Core Web Vitals back in May of 2020.
They are a set of metrics related to speed, responsiveness and visual stability. There’s three of them. We’re going to go over all of those today in detail. Those metrics in no particular order are largest contentful paint, which is the time it takes for a page that’s made content to load, so an ideal LCP measurement would be around 2.5 seconds or faster. Again, simplest way to say, largest contentful paint measures loading. Okay. Then we get into metric number two, which is the first input delay.
First input delay is the time it takes for the page to become interactive to the average user. An ideal measurement there would be something like less than 100 milliseconds. First input delay measures interactivity. How long does it take before the user can interact with the page?
Then we get into the third metric, which is the one that’s really cost much consternation and not only with everyone on this call, but in all of our clients particular is cumulative layout shift or CLS. Cumulative layout shift is the amount of unexpected layout shift that kind of happens on visual page content.
When you opt… An ideal measurement of that is like less than 0.1, which measures visual and self. You could say the cumulative layout shift measures visual stability. I can’t wait for Andrew to jump in on this. I know he has a lot to say. But these layout shifts specifically happen whenever a visible entity changes in a starting position.
If I’m… Like top and left position is the default writing mode, between frames and that stuff moves. If I open up a page and you have a logo and that logo is not sized correctly, there’s some movement as the page loads, especially as it renders on a mobile device and that causes a shift and that shift to Google is a very poor user experience.
We’re getting a lot of that. The CLS metric specifically is impacted by a ton of issues for those of you on the sites running ads. Specifically, we have… If you use fonts that are improperly displayed and I know
Andrew is going to get into this a little bit. He’s going to talk a little bit about FOIT versus FOUT, basically flash of invisible text versus flash of unstyled text, which can lead to fonts jumping around. Many of you suffer CLS issues because of embeds or just because of banner ads or IFrames. We’re trying to minimize those as much as possible.
Probably one of the biggest reasons that CLS happens is an inability to specify dimensions of images, which is something else that we’re going to talk about today. Or, of course, any content that is dynamically injected on a site.
What’s an example of content that’s dynamically injected on a site?
(00:07:26): Ads. Imagine that. Ads are one of the biggest things that you’re going to see that are going to cause issues with CLS. Now, some ad companies are better than others in trying to fix this. So we’re going to review that today.
Of course, one of the other biggest things that go into CLS or actions that are waiting for a response from the network, before updating the DOM and the DOM, of course, is the document object model that makes up the page.
I know many of you are like, “Oh, my god Casey! Acronym soup. My eyes have already glazed over, what are we going to do?”
Again, we’re going to try to break this down for you, these three metrics, these three Core Web Vitals and then we’re going to talk about how these three Core Web Vitals are going to be grouped with four other metrics to make up what’s called the user experience algorithm that Google has already announced is going to be live next May. Hopefully, we’ll get into all of that today.
How is this update different from previous Google Updates?
(00:08:35): Because they announced it. Usually Google doesn’t come out and tell us that they’re doing an update.
Usually, it just happens and then they let us know, post factum like, “Oh, we just did this update.” Typically, Google lets us know about these types of updates or changes to the way they evaluate websites when there’s something that we can do about it. Like mobile indexing. I can’t think of another one right now.
But basically stuff that you should be probably getting your site ready for. Stuff that’s within your control. How frequently does this happen?
Google changes its algorithm very frequently. I think Mueller said to Casey like, “Once a week, if not more. Almost on a weekly basis, they update their algorithm.” But this, this is something that they’re giving us time to prepare for. They’re giving us instructions.
But again, this is nothing new. This is stuff that we’ve been kind of preaching for a while and now this is one of those things where Google again, just like what we saw with the November update, Google’s coming out and saying, “Okay, so you guys haven’t been listening. We’re now making this a part of the evaluation of your website.”
They’re giving us time, just like with mobile indexing. They’re telling us it’s going to come into action next year. So now’s the time to start paying attention and really, really following the advice that we’re putting out for you guys.
How does Search Console validation work or Core Web Vitals work?
(00:10:19): Most of you probably have access to your Google Search Console already. Usually, when it sends out an error or warning, you can click the little validate fixes button and then it says… It does like a pre-validation for about a minute and then if it passes that it then says, “Okay, validation process has started.”
Then in a few hours or a few days, it kicks out an email and says whether you passed or not.
For these, it works a little differently. You can’t just validate like that because it needs to look at real world values from the Chrome User Experience Report. If you’re trying to validate an issue with your cumulative layout shift for example, and you click validate fixes, that actually starts a 28-day clock.
And so it starts tracking from that point on what it’s going to look like. Then after 28 days, it may say that your CLS scores have gone down and are much better or may not.
When you’re doing optimization, you don’t want to constantly be clicking the validate button. You want to do all your optimization, get it as good as you can and then click the button to re-restart the test and then evaluate from there.
That’s not the only tool for evaluating CLS. That’s just in the Search Console. There’s been a lot of confusion about that in terms of validating that one.
Google Core Web Vitals Report
(00:11:40): Just to very quickly kind of add in on what Andrew has said there, Google has a page, and I’m going to go ahead and paste it over. It’s called the Core Web Vitals report. And there’s lots of little snippets on that that I think all of you would find of value here.
If you scroll down through that, there’s a couple things that you need to be aware of including, of course, as Andrew said, when you click validation on this. We’ve had clients come to us and say, “Well, Casey, I was in the account, I saw some errors and then I checked next week and there weren’t any errors and I didn’t do anything.”
Andrew, if you actually run across that where we’ve actually seen issues saying, “Okay, there were red errors there. But I didn’t validate. And now those are yellow.” Or “Oh, hey. I had some yellow errors there and now they’re green or they’re not there anymore.” I’ve only run across it a couple of times.
But basically what’s happening there is Google is helping you out. It’s possible that you’ve made borderline changes to your pages without having to do anything on your part to let Google know about it, but they found those changes.
When that happens, they specifically say that there are examples where we will go in and we will rerun reporting at our end without your notice. And if we see that you’ve made these changes, or you’ve made a big change, like for example, something switching their networks.
Here’s a good example, Arsen, if you were on Wix, and you switched from Wix over to WordPress and you made a lot of changes to your page speed that were very noticeable, especially for Google with Chrome UX history that they pulled from your search got from your analytics, you would most likely see big changes in data collection in Search Console.
If you go in and take a look at this report here and I think I’m going to make sure that I paste it over here again, there’s lots of finite examples where Google will say, “Hey, we’re going to try to do some of this heavy lifting for you. If we see that you’ve made changes, we’ve seen that you’ve made dramatic changes to the site architecture itself or we’ve gone in and reassessed your site, you might see us, especially if you’re a borderline case, we’ll give you the benefit of the doubt there.”
How big of a ranking factor is Core Web Vitals going to be?
(00:13:52): They’re going to be super important. I can’t speak to two of them being a ranking factor.
Are they a quality signal? Absolutely. Google has come out and said, I’m quoting, I just pulled this up, “A good page experience does not overwrite having great relevant content.” Google understands that not all sites are created equally. Not everyone is on WordPress. Not everyone has a CMS.
We’ve worked with some content management systems especially on the E-commerce side that still don’t give you a place to add a canonical to a page. You’re still very limited. And Google understands that.
That’s why things like schema, having schema on your page is not a ranking factor. Having schema on your page is not going to help you rank higher on Google. It’s going to help Google understand what the page is about.
It’s going to provide a better experience in the search results by putting out the little image, the thumbnails, the snippets, the star ratings and all of that. But you can still rank at the top of page one without any schema on your site.
Google understands that not everyone has access to do all this cool stuff or stuff that’s necessary to be done on their website. I don’t think they’re going to make this such a strong ranking factor.
But then again, I really don’t know. But as a quality signal, super important because again, this all boils down to user experience and Google wants to make sure that whoever’s performing a query on their platform, the top results that they’re going to engage with are going to be quality, that they’re not going to provide a bad experience because then they’re going to start losing market share to other search engines who are doing that.
Keep that in mind. It’s good to do this, it’s important to do this. But fixing this will not make you number one on Google.
(00:15:48): I believe that, again, these Core Web Vitals are going to be grouped with four other ranking factors and they are going to be part of a larger user experience algorithm and that user experience algorithm is going to function more like a tiebreaker.
Very similar to what happened individually with the HTTPS, the mobile friendly things, all of those.
Now, this is important folks, because I know a lot of you on the call are in the food and lifestyle niche. You could be more impacted by this than the vast majority of other niches because your content is so similar.
How do I know that? Because today alone, I was targeted on Facebook with three Christmas cookie recipes that were exactly the same. From three different bloggers. They were literally the exact same recipe, little bit of difference here and there. But they were very specific.
I’d even have to go and take a look at it. But it was a very specific kind of Christmas sugar cookie that they were doing. This is where this user experience algorithm may affect you. Because if you’re competing against another site for the same query and you’re relatively equal on a lot of other metrics, maybe they get a boost if their UX experience is better than yours.
That’s something to be aware of. We definitely want to make sure that we get this styled in.
What’s the absolute top must fix item publishers need to focus on fixing before May?
(00:17:25): The full quote that Arsen was talking about was, “A good page experience doesn’t override having great relevant content. However, in cases where there are multiple pages that have similar content, page experience becomes much more important for visibility in search.”
This is directly from the announcement that Google gave when they were pushing out the page experience algorithm. This is then telling you that again, your quality of your content, how you’re presenting is still of the utmost value.
The rest is a tiebreaker. I think my advice, especially if you’re looking to concentrate on one thing, is taking the rest of them aside. I’m hoping that everyone on the call knows that you have to focus on these seven metrics. Hopefully all of you on the call are already mobile friendly.
You’re already doing safe browsing, if you don’t have any safe browsing notices in your Search Console. You already moved to SSL. So you’ve got the HTTPS security dialed in. You’re not using intrusive mobile interstitials.
Okay. So if you’re on the call and you have a mobile interstitial that’s popping up on the first clip from Google, you need to cut that stuff out immediately. A lot of incorrect information about that. All mobile interstitials that come up on the first click from Google, even if they scroll down, guys, that’s not correct.
If you go back to the original announcement from Google, it doesn’t matter if there’s a delay there. It’s on the first click from Google. So if you’re going to be using mobile interstitials, they definitely need to be non-intrusive, specifically.
That aside, let’s say that you’ve dialed in your three Core Web Vitals, you’ve dialed in these other four metrics here, the mobile friendliness, the safe browsing, the HTTPS and you’ve eliminated all the mobile interstitials, now we come down to the most important kind of the meat of the issue which is the quality and relevancy of your content.
Now, the general view is that all the elements of the page experience are important, but it’s still the best content that Google wants to rank. They’re going to prioritize over that over even some subpar issues on the individual metrics that go into the UX algorithm.
Just the bottom line, a good user experience doesn’t outweigh the quality and relevancy of the content for that query for that user at that point in time.
Can slow site speed be one of those issues Google is checking for?
(00:20:06): Absolutely, absolutely. Yeah. We’ve all been there. We’ve all tried to load a very slow website trying to get a piece of information and it was very frustrating. I get this all the time.
Dealership websites are notorious for their speed issues. A lot of recipe bloggers, unfortunately, are still on all of our audits that we’re doing, we’re still catching page speed issues.
But yeah, you definitely want to work on that. That’s one of the things that you should be looking at. I would focus on, if possible and if you’re with Andrew, this is already probably being handled for you.
But you’d be moving on to say, third party scripts, setting up better hosting, lazy loading, removing large page elements or watching the size of your images, not the actual physical size, but the memory size of your images will all help.
Every little bit helps. Again, it’s not the North Star. You should still be working on this. You should be working on this but it’s not something that you should drop everything else that you’re doing and focus on optimizing for speed.
Where should you go to analyze your metrics?
(00:21:46): It’s a great question because these are just testing tools. So they run a simulation and then give you a bunch of information so you can go fix your site.
So they all work slightly differently. I like to rely on Google PageSpeed Insights first, because that’s Google, you’re going to the source. It doesn’t mean it’s the only tool to use. I think it’s important, especially for things like cumulative layout shift, using multiple tools is really important.
For example, if you’re using Clickstream on your site, which a lot of people are using, it’s the filmstrip that appears at the top of the recommendation engine… It recommends other posts.
That actually appears late in the process. Carl, and his team have been really good about not interfering with page speed. But in the real world, that can be tracked as a CLS shift or a layout shift by the Chrome User Experience Report. But the GPSI testing tool doesn’t flag it.
When you’re using GPSI, it’ll show the field data at the top and that’s actual real world user data that’s from the Chrome User Experience Report. That’s in actual browsers.
If you do that and then you look at the lab data, lab data might say 0.0 in all green in the CLS section but if you look at the field data, you see that there’s a shift there. We see that a lot with Clickstream, in particular.
PageSpeed Insights doesn’t catch Clickstream. There’s ways to fix that one. You basically have to put in the spacer so that when Clickstream loads, it’s not pushing the page down because you’re shifting layout preemptively so the user doesn’t see that.
The filmstrip at the bottom of PageSpeed Insights, if you look at the little thumbnail screenshots, if you compare them, you can see what’s happening during the load process. They’re really tiny. So they’re kind of hard to see.
But if you see something where your text comes in and then the next thing, it’s lower, that’s a layout shift. Those thumbnails can be really helpful for identifying what they are.
WebPageTest is another great tool. You can record video with WebPageTest and actually play it back in slow motion.
So it’s kind of like watching paint dry, because it’ll be tenths of a second. But that lets you really see what’s happening because it slows it down enough to sort of give you that insight. So you can see that this loads and then this loads and that pushes things down.
So you can see what’s really going on. The short answer is use multiple tools. In terms of what’s important, I’d say the Chrome User Experience Report is pretty darn important, because that’s what Google’s looking at in terms of rankings. That’s not necessarily going to tell you how to fix it.
What is the Chrome User Experience Report?
(00:24:27): The Chrome User Experience Report is an example of a Big Query report that you can set up and run on a monthly basis and it will pull in all of your historical information from analytics showing how your site is graded on a monthly basis on these various Core Web Vital metrics.
Now, for you to make use of this report, you’re going to have access to a Google account and a Google Cloud project account. But they’re very easy to do. As a matter of fact, if you click on the link I’ve given you, it will go ahead and prompt you to create that Cloud account. And then you can go ahead and use the same link to enter your URL and start pulling your own historical data. This is something that I, Arsen, Andrew, we include regularly with our reports to clients. How these Google Chrome accounts work is there’s always a month lag.
So clearly, right now, the newest data we can get is up to the end of October. Then Google runs new Chrome UX data on the second Tuesday of every month thereafter. When I’m doing a mini audit, or a full audit, I will pull the Chrome User Experience data over several months. I can look through it and see, “Oh. Look, here’s where they worked with Arsen.” And “Oh, here. Look. Here’s where we changed hosting from Bluehost to BigScoots.” And, “Look at the jump in the Core Web Vitals.”
We have that historical record so that we can start looking at these metrics over time. The problem with these Core Web Vitals and the problem with these user experience reports is that Google changed how they measured them from April to May.
There’s a lot of lost metrics. People reached out like Casey, my Core Web Vitals and everything were great in April and then they literally fell off a cliff in May or June and that’s because Google changed how they started to measure some of those metrics.
So understand that if you’re going to run a historical report, kind of just ignore everything before May and look at May, June, July, August, September and start benchmarking from there. That’s a more solid and valid benchmark to use for these Core Web Vitals specifically.
Is this all the same information in just a different view?
(00:26:52): Let me also just mention, this data that Casey’s talking about, it’s the same data that’s showing in GSC, it’s the same data that’s showing as the field data in PageSpeed Insights. It’s all the same information, it’s just different ways of viewing it.
(00:27:04): It’s just visually, you can actually understand it. It’s a chart. The chart shows you, here’s a month view of your first input delay. Here’s a monthly view of your cumulative layout shift. I’m like, “Oh, my gosh! Only four people out of 100 are having positive CLS issues, whereas previously, maybe five people out of 100 were having a good experience.”
Again, it’s a visual way for you to kind of see a breakdown of improvement overall. If you’re working with someone, maybe you’re signed up for a hosting plan or not even a hosting plan, but a blog support plan and they’ve told you, Core Web Vitals are not important, you should get another blog support plan.
Or if you’ve asked them about these user experience metrics and they don’t know what you’re talking about, you should probably get a new blog support plan. I bring this up, because I’ve had this happen over the last several weeks.
We’ve got those people to switch and they’re very happy they did. Be aware of that. This is a business. You want to pay for the best support you can possibly get. Lots of options out there. If you need some suggestions, we can help you on that.
How long should you wait to look at the data after you’ve made a change to your website?
(00:28:31): Well, we certainly won’t be able to see it. For example, and Andrew will tell you, we can make changes for Google Core Web Vitals today and we could literally have 100% improvement the very same day. The only way we’re going to be able to visually see that is for us to run the Chrome plugin or to use Lighthouse or to use the Google PageSpeed Insights.
But we won’t see the aggregate data for at least 28 days because that’s how long it takes for it to run that collected information because Google uses an aggregate model.
Their goal is to pull and aggregate values across all of your URLs. But that makes sense as well, because it could take 28 days for us to get most of our URLs reindex and re-processed anyway, so it makes total sense.
With the holiday period and bloggers publishing a lot of content at the moment, should they wait until January to address Core Web Vitals?
(00:29:39): I’m going to give Andrew and Arsen a chance to weigh in on this. But whenever it’s a technical issue, we want to make those changes now because technical issues are one of those things where we can have huge ranking swings on the positive if we make technical improvements because we’re making it easier for Google to crawl and algorithmically score content.
If your page speed is so horrible that you’re timing things out or let’s say you’re on the call and you’re running AdThrive and you haven’t reached out to that drive to turn on deferred loading or little things like that, you’re going to be shocked at how much better you do by doing little things like that. Those are technical fixes.
Or if you know that you’re failing Core Web Vitals because of issues with your logo or because now you know, kind of how to check those, making improvements and all that right now as we go into the busy holiday season is absolutely only going to help you. So in that respect, yes.
Can it be harmful to update the theme of your website?
(00:31:11): Changing a new theme is a migration because you’re moving to a whole new everything. The code is different. How your content is organized on the page is different.
Different scripts, CSS, JavaScript, unless you really, really, really, really have to. We do a crazy amount of migrations per year at TopHatRank.
For recipe publishers to very large brands and we always say unless you really, really, really, really, really have to, don’t because if it’s working for you and it’s making you money and you’re ranking at the top of page one for a lot of your queries, this will all shift when you move to a new theme. It’s going to come back if the migration is done properly but you’re going to see a fluctuation in rankings.
Google has to recrawl the new site, has to get to know what the new organizational structure is and all that. If you’re considering moving to a new theme to address Core Web Vital issues, you’re probably on a really crappy theme right now that you want to switch to a new one to address this. I can’t speak to a theme. Skylar’s themes are amazing. A lot of you guys are on them.
He puts in so much work into them. Every time we do an audit, there’s less and less stuff for us to find just because Skylar is implementing a lot of this stuff into the theme. It’s such a weird thing to do just to address a handful of issues that can be probably fixed with hiring a very smart developer and maybe dumping a few hundred dollars to fix certain things like getting rid of JavaScripts that are not being used on your theme.
You probably have a theme that has a shopping cart JavaScript somewhere in there and you don’t have the need for a shopping cart, but it’s still there and it’s still being injected into your header. I don’t think moving to a new theme is a good idea unless you really, really, really, really have to.
(00:34:09): I would just add and I want Andrew’s opinion on this too specifically about Divi themes and other ones like that that use page builders. That’s the issue, is that most of these themes come with, grouped with page builders.
Page builders add a lot of code bloat, they immediately kind of put you behind the eight ball with regards to excessive resources that load on the page. We can optimize them. It’s just harder to do.
So we would try to have you not use a page builder. Gutenberg is a page builder. You don’t need two of them. Use Gutenberg. Use the blogs. That’s literally what WordPress is trying to do is move everyone to Gutenberg, which is the page builder. Honestly, that’s what blogs are.
(00:34:52): Before Andrew jumps in, we just recently analyzed a site and this goes with Divi, a site that was moved from one theme to Divi and then from Divi to a different theme, and now they came to us because things are just not functioning right.
And then when we started, when I actually crawled through the site and looked at some of the historical content, some of the historical posts, they still had the Divi page builder code, the HTML that wraps around each div being rendered as text on page on these posts.
It came back as having wrapped around each sentence. It was like an h6 or something. Unless you’re like an E… Not even eCommerce. Lead gen and you’re constantly producing different types of landing pages and you don’t have the money to hire a developer or somebody to build these out for you, then a page builder like Divi or UDesign or any of those guys will work for you. But for recipe bloggers, it’s unnecessary.
(00:36:09): Don’t use Divi no matter who you are. It’s awful. There are some page builders that are okay. If you really need a page builder, Elementor is a decent one. There’s philosophical problems with these page builders.
Like Arsen was saying, you get locked into them and especially if you’re using them for your content, that’s really hard because then you have to go back and edit your content.
(00:36:32): Philosophically, I’m opposed to these page builder tools. I don’t think it’s a great idea to go to ThemeForest, for example, and pick a theme that you think is pretty and then run with it. Because the themes that have all these features, they’re adding all this extra code to make those features work. If you’re just building a little portfolio site and that’s not your core business and you don’t need to get lots of traffic, then it’s probably fine.
But if you’re building a food blog or a travel blog and this is your business and you need to be competitive, those just very likely will not cut it.
I think Skylar’s themes are definitely a great way to go. They’re very reasonably priced. He’s also supporting them through the Feast plugin.
As all of us are figuring out the CLS stuff or whatever it is that comes next, Skylar is busy updating the plugin to make sure that the best practices are implemented. It’s so much easier to update the plugin than to update themes.
And so he’s able to keep actually improving the features there. I think it’s worth considering Trellis. They just released their beta. I personally don’t think it’s ready for primetime.
I hope nobody from Mediavine is on the call yet. But I think it has a lot of potential. At this point, especially if you’re a food blogger, do not switch your theme in Q4 unless you really have to. We’re at the week before Thanksgiving. This is peak traffic. If it ain’t broke, don’t fix it.
But I think Trellis is growing. I think Trellis is going to be a very compelling replacement for Genesis which is kind of being phased out. Genesis is the parent theme that the Feast themes are actually built on. Right now Trellis only has three child themes and they’re not all that pretty.
Are there any Feast updates coming soon with Trellis?
(00:38:25): I know he’s working on it. I don’t know what the state is right now. But I think Trellis is complicated. Building a theme is complicated. It’s taking them a lot longer than they expected. I know everybody’s been like, “When can I get in on the beta?” I’ve worked on a few sites. I think it’s going to be great. It may not be there yet, but I think it’s going to be. I’m excited about what they’re building with it.
(00:38:49): Well, I’m excited they corrected the issues. I audited two last spring, early times, tons of issues, would not have recommended it to my worst enemy.
It is substantially improved now. They’re not wrapping everything and heading tags, they’re using headings correctly, they’ve provided some functionality that comes standard and things like the feasts themes or the Genesis as a default framework, ability to optimize category pages, having featured images, some built in schema when it makes sense.
But yeah. Again, the only concern I have with Trellis is that they tend to optimize for their own plugins first. The goal, of course, is to make sure that everything works with what you have.
You should not have to change everything about your site to make Trellis work for you. It should be the other way around. Trellis should work as much as possible with widely accepted best practices. That’s always been something that we struggled with Mediavine specifically is that creates growth.
These other plugins that they have tend to have more issues than their competitors, because they literally design them to work well with each other but not with their neighbors. And hopefully that’s something that we can see an improvement on with trellis and as we enter 2021.
Should I hire a designer to create a custom theme?
(00:40:06): Also, just to continue the spectrum of theming real quick. If you’re at the point where you’re ready to hire a designer for a custom theme, then speed and accessibility should be a big part of that conversation. You should be looking at samples of other themes that they built and running them through GPSI testing and seeing how fast they are.
They may not have total control, because the user might have added another plugin that slows things down. But you can put that stuff in your contract. You can say there needs to be… The CLS needs to be 0.0 when tested on webpage test and GPSI.
A good developer will actually lead with that stuff. They’re going to say, “Oh, yeah, we’re on top of this stuff. So don’t worry about it.” So if you’re shopping for a custom theme, I think it’s something to pay attention to.
(00:40:51): Yeah. Custom themes are a tough thing. We get it all the time. Like, “Oh, my gosh Casey! Should I invest $20,000 on a custom theme?” That’s a lot of money.
We think of how many goods I can buy with that. I tend to tell bloggers, “If you’re going to make that investment, it better be going towards their long term view. Because the worst thing you can do is tie yourself to a designer, who makes a theme and then have the designer disappear or be very hard to reach.”
Let me tell you, I won’t mention any names. That is an absolute issue in the food blogging niche.
You pay for issues on a custom theme and then it’s very hard to get this designer to come back and make small edits or to make changes to update that site for accessibility or best practices. That’s why if you’re going to get a custom theme, you want to work with someone who’s building on a fully supported framework like Genesis.
We tend to recommend Genesis as a framework and then you build on Genesis with custom features. That’s what Feast does, that’s what Bill Erickson does. I believe that Madison Weatherwell, is that how you pronounce her last name? I always screw it up. I believe that she works with Genesis and she builds on Genesis. That’s what we’re looking to do.
The days of having these full custom themes hurts you, absolutely hurts you. I’m against that. You really do not need because then you have to just go back and have them add these features that they’ve forgotten because you don’t have the experience to ask the designer what you want in the theme until you’ve got an audit with the professional like myself or Arsen or someone else and we’re like, “How much did you spend on this theme? They didn’t do this, this and this. Can you go back and have them add this?” And usually it’s not possible.
(00:42:31): I’d say also, you want to make sure your developer’s, if they’re building a theme for WordPress, follows WordPress coding standards. If they do that, then any good developer should be able to pick it up and help with things.
There are some companies that turn things around and make them [inaudible 00:42:46]. I don’t want to say anything bad about any companies-
Out of everything that needs to be fixed on a site, what tasks would you recommend a publisher potentially hire out instead of trying to do themselves?
(00:45:05): Yeah. If you know what you’re doing, if you have an engineering dev background and you’re comfortable modifying files and CSS and working with the databases and all of that, there’s stuff that you can do.
Stuff that you do from just WordPress’s interface would be… And again, if you’re working with somebody like Andrew, with NerdPress, a lot of this stuff is already being done for you like minimizing your JavaScript and caching and browser caching and all of that.
What’s the first thing you should modify on your site if you’re making changes yourself?
(00:47:01): It is get rid of all that extra crap, is that the right word? All the extra stuff, all those things that… Those plugins that you installed and you’re not actually using the feature or you have three plugins that do kind of the same thing.
All of those can add JavaScript and stylesheets at the front end of the site and slow things down. It’s always a great idea to review your active plugins, get rid of your inactive plugins and really pare that down.
That’s, I think the easiest thing you guys can do yourselves right now especially because you know what features you need on your site the best. Just doing a quick plugin audit I think is a good-
(00:47:37): Absolutely. Still 117, that’s the record. I haven’t found any sites with more than 117 yet. It’s been a while. Good times.
(00:47:46): You should be reviewing your plugins at least once a month. You should be looking at what’s happening, does that need to be updated? Anything weird, like, “Oh, my God! What is this plugin? Why is it even here?” By using this, you should not have anything under that you’re not using.
You can always download it later again. Sorry, if you install the plugin to help with one specific task to move a post to a page, get rid of it. You can always download it and install it again. You don’t need to have it live there.
What can you do to prevent cumulative layout shifts?
(00:48:27): It’s similar stuff Arsen was just talking about, where if you’ve got ads, make sure your ad network is handling that. I know Mediavine is working on it. I assume AdThrive is working on it. They should handle that stuff. You want to have your ads loading below the fold. Actually, I saw a question from Sean.
A few people actually are asking about CLS issues on desktop instead of mobile. Desktops probably read because you have above the fold ads, whereas you don’t have above the fold ads on mobile. I know with Mediavine, also everybody’s deferring ads on mobile, which is great, but they’re not deferring ads on desktop.
Most desktop users, that’s probably not an issue because they’re going to have a faster connection. But if the ad is above the fold, that’s going to cause a layout shift.
(00:49:06): If you’re willing to take the revenue hit, then you can get rid of those ads above the fold on desktop and that’s most likely going to solve it. Beyond that, it’s going to vary from site to site. If you’ve got Clickstream or if you’ve got a banner ad, that’s a common one.
You’ve got a GDPR or a cookie notice that says like, “Hey, we have cookies.” If that shows up at the top and pushes your content down and that loads late in the process, that’s going to cause a layout shift.
One solution is to put it at the bottom, although that might interfere with sticky ads. Or you could have something on the sidebar. If you have an opt-in form, “Get my new recipes.”
And you have an email form at the top, if that loads late in the process, that’s going to cause a layout shift. You have to have a spacer in there. Some plugins will be good at this and some won’t. Some of that might be changing out which plugin you’re using.
Beyond that, it’s kind of you have to do it on a case by case basis and really look and see, use the tools and see what’s actually moving and then you can address it that way.
Do you think the Core Web Vitals are going to impact Google Discover traffic?
(00:50:20): That’s a good question. Although Google has never said anything about it, specifically, we can kind of glean some insights based upon their information that they publish about Google Discover.
I’m going to paste over a link which has a ton of information about Google Discover well worth your time. It’s under Google’s advanced SEO module that they’ve launched recently. The thing to be aware of is that Google is very clear that Discover traffic is serendipitous.
For those of you who won’t know the big deal about that big word I just used, basically, we can’t plan for it. Google Discover is very unique. It’s tied to content based upon Google’s automated systems. It’s matched to your own individuality and your interest that you set up on your Google Discover feed. I’m going to go ahead and paste that over here a little bit. You can take a look at that.
Now, we can’t really create optimized content for that. Therefore, I’m not really too concerned about the Core Web Vitals impacting Discover traffic too much. Clearly, we want everything to load fast as possible.
But one of the things that you want to be aware of with Google Discover, especially if you’re a food or lifestyle blogger and you’re on the call, is that you want to start looking in web stores. Web stories, there is a new carousel in Google Discover, completely devoted to web stores.
What are Web Stories and why are they such a hot topic right now?
Web stories are currently going through a land rush. It’s literally like the Homestead Act going through right now. People are going out and they’re putting together these web stories.
They went from having 1% Discover traffic to five, seven to 10% within a week. There’s a lot of people putting these web stories together. The problem is, is there’s kind of a correct and an incorrect way to put these web stories together.
I’m going to go ahead and put over some information in the chat in a minute about what you need to know for these web stories. But mostly, these web stories are again and I’m not a fan of amp, but if you’re going to use amp, only use them for the web stories.
These web stories, you download the web stories plugin from WordPress and start playing around with putting these web stories together and you want to use some of your already top performing recipes.
The web stories are already canonicalized to the page. There’s no duplicate content or anything going on here. The goal is to use the web story to drive traffic to the actual recipe. Now on that note, I want to go ahead and talk about JumpRope. I know a lot of you on the call are using JumpRope. I am 100% against jump rope and this is why. Jump rope is a great way for you to drive traffic to JumpRope.
JumpRope is a great way for you to drive advertising income to JumpRope. JumpRope is not a way that you’re going to be using to build your traffic. It’s one of the most interesting things I’ve seen is how fast they came onto the screen but if you’ve ever pulled out your phone and looked at the Discover feed, those Jump Ropes, don’t go to your site, those JumpRopes, go to a JumpRope channel that you set up.
So you’re requiring users to click twice from that, from the JumpRope to the JumpRope channel, from the JumpRope channel to your site. You’re going to have extremely low converting traffic on that. It’s just not worth your time. If you’re going to use web stories, don’t use JumpRope.
Go ahead and build up your own web stories so that we can go ahead and have them sent to your site on every frame of that web story. That’s what we want to do.
There is no benefit for you getting to the carousel using JumpRope, when you can just go ahead and start making your own web stories now and have the same visibility. Be aware of that going forward.
How can you satisfy, especially with this update coming up, both the audience giving them the recipe and Google who wants the content?
(00:54:31): Well, again, so Google… There’s no such thing as content length as a ranking factor. It just doesn’t exist. It can actually have a reverse effect from what we’re seeing. You’ve got to keep that in mind. Google is really heavy on intent and understanding what the intent is behind a query. We’ve covered this in a previous webinar.
If my intent is to find a temperature to cook something like a tomahawk steak or a portabella mushroom, whatever it is, I’m going to go in there, I’m going to find the temperature and I’m going to get out. I’m not going to read anything else. My primary intent is to do that one thing.
(00:55:09): I might have secondary passive intent, which focuses around what else can I serve it with? Maybe there’s a different way of cooking this. But it might be.
Users are going to go in, they’re going to look at the piece of content that they’re interested in and they’re going to leave. If I’m trying to figure out what I’m going to make for dinner and I’m searching for keto friendly chicken dishes without cheese or with bacon, now, I’m presented with a page, whether it’s a category page or it’s a roundup post where I’m presented with a bunch of different recipes and I’m going to make my selection.
I’m going to spend a little bit more time on that page. Because that’s my intent. Google looks at that. We’re assuming that Google understands what or assumes what the intent is behind the query. I wouldn’t say that you have to write long form content in order to rank. Look at the big players. Look at the guys who have been winning through these last few updates.
Content is short to the point addressing the primary intent. The primary intent is to get that recipe. We’ve been preaching this. Prioritizing content based on intent. You don’t need a long story. Give the user what they want and then everything that addresses secondary intent should be moved towards the bottom right.
Focus on giving the user or matching the content to the query syntax and the query syntax are the words in that query. If I’m looking for the temperature to grill a tomahawk steak, that should be the first thing that I see when I land on a page.
How do you improve LCP and SCP?
(00:57:12): LCP is the largest contentful paint, we haven’t really talked about that one much. And FCP is the first contentful paint. First contentful paint means it’s basically the first thing that gets drawn on the page.
Largest is the largest thing and this is within the viewport. So basically, before you scroll, and we’re talking mobile, so if it’s on your mobile phone, whatever the first thing to appear is the first contentful page, whatever the biggest thing to appear is, is the largest.
The Web Vitals are really focusing on the largest contentful paint with the assumption that if it’s bigger on the page, it’s more important for the visitor.
It’s all the standard page speed recommendations to fix these. You need to basically be on a fast host, you need to have page caching enabled. If you’re generating content on the fly, it’s going to take too long. You need to defer stylesheets and JavaScript.
Using critical path CSS is really important. Critical path CSS basically embeds the styling that’s needed to render just above the fold content, actually putting it in the HTML document so that when that document comes back, the browser has that style information already, it doesn’t have to then make a query for another stylesheet and get that back.
It’s still going to do that for the rest of the page. But it’s already drawing stuff at the top in the beginning. This stuff is really important for user experience, particularly because up until this point, the user is staring at a white screen. If that takes two seconds, it feels like a long time.
If you shave off two seconds, you want to do it from the beginning of your load process, not to the end. Once there’s something there for the visitor to look at, if it takes an extra second or two, it doesn’t matter as much, it feels faster because the user sees stuff happening.
If you can get that stuff to happen really fast, using a content distribution network is really important. For example, with our services, we use CloudFlare. We have an enterprise configuration now. We’re caching HTML at CloudFlare.
Once it’s cached, we’ll return that HTML document to the visitor in under 50 milliseconds no matter where they are in the world. And then all of the other static assets can be returned really fast from CloudFlare as well.
When you really start getting into this stuff, there’s lots and lots of layers. Some of the stuff we were talking about earlier like getting rid of extra plugins, you want to get rid of all that bloat first and then optimize the other things.
Our primary tool is WP Rocket. It’s a premium plugin, we include that with our plans as well. That will do most of what I just mentioned. We couple that with the CDN, we make sure somebody is on good hosting like BigScoots or Agathon. We take all of those things together and that should really make a significant improvement on both your first contentful paint and largest.
One other weird thing about largest contentful paint is it can vary from page to page. We’ve seen this a lot where let’s say you have a really short post title and then you have another post with a really long post title which wraps under two lines. That becomes a bigger element.
If that pushes an image down, you could have a different largest content, full paint on each of those pages and you could have a significantly different speed. If you test the two pages, one might be two seconds, one might be six seconds and it’s because it’s loading something different or it’s looking at something different that loads at a different point in the sequence.
We’ve seen that if you have a weird discrepancy where some pages are very different, you have to actually look at that thumbnail image and look at what’s changed and that can make a big difference.
You have to dig into this stuff and really… Unfortunately, there’s a learning curve here. The other options are to hire people. I think it’s important to hire experts once you hit your limit.
But it’s also important to understand what you’re hiring them for so you can make sure they’re doing a good job and you can talk with them smartly about it to make sure they’re actually doing right by your site.
(01:00:54): I do want to add one thing. Sorry, let me jump in. I’m noticing this in the comments. Like, “Who do I hire to fix X, Y, Z?” Don’t approach it that way.
Don’t look for help with fixing a specific thing that you got from a tool that’s telling you this is what you need. This is just a suggestion. A tool is still just a tool. I’ve said this before. You need somebody who really understands what’s happening.
Keep in mind, the tool is looking at it on the page level, not the entire site as a whole level. It’s telling you this piece of JavaScript is not being used here on this page but it could be being used somewhere else.
So going to a developer and saying, “Please remove this JavaScript specifically for that.” You might cause harm. You need to work with someone who really understands what they’re doing. Audit is a scary word for us.
But somebody who can audit the situation of your CLS or any Core Web Vital situation that you’re having and then tell you, “Yeah, this is something that should be fixed,” or, “this is something that should be optimized,” or, “this should be left alone.”
When might be a good time to switch themes?
(01:02:20): Yeah, Apaline is signed up for an audit. In this case, do not listen to what we just said. You have a theme that is… Again, I don’t want to call you out. But you’re riding an Edsel and we need to get you switched to a Ferrari and your theme is so bad that any improvement we make is going to be phenomenal for you.
That’s just for her situation folks. Okay, so just be aware of that. That’s why I want to answer it quickly here. If you have any questions on that, we’ve been trading emails, you’ve already also been in touch with us Skylar, you’re in good hands. The faster we can get you converted the better.
If your traffic is okay and you’re doing okay and we don’t need to make any dramatic changes. I’ll have a call with a client today. Andrew knows who that client is. She’s a current subscriber. She has issues with her theme but she is doing very well.
There is no reason for us to upseat an apple cart or do anything dramatic right now as we go into a busy period just to switch themes. But if your site is throwing up errors on every page and your page speed is non-existent and you have schema issues and literally again your theme looks like it was put together by a four-year-old, we need to make some changes.
Again, don’t take any of this stuff personally but our goal is to get you to provide you the best advice, but that advice is relative to the site we’re looking at.
Why do I have discrepancies between my data when it comes to desktop vs mobile performance?
(01:05:33): First of all, I want Andrew and Arsen to speak on this. I do not believe that Google is using desktop at all for Core Web Vitals at all, at all. I think that they’re doing it because they want to provide us with both the information.
But everything with Google is mobile first. It is my belief that unless someone can show me a quote from Google otherwise, I wouldn’t even lose any sleep over the desktop nor would I even concentrate any resources on that.
I would focus everything on mobile first because I don’t believe that you improving your Core Web Vitals on desktop is going to mean anything because we’re using the mobile first version of your site for all metrics. Unless someone tells me differently, for all of you on the call getting poor Core Web Vitals for desktop, I would ignore them completely. We don’t even worry about them.
(01:06:17): And just to add to that, I get the same thing. I have been running our website TopHatRank through both desktop and mobile. It varies. Why? I don’t know. The content shifts. So you have a responsive layout versus a desktop layout. But I really don’t care what’s happening in the desktop version. I only care… I’ve got the notification that I’m mobile first with Google. I don’t care about the desktop at this point.
(01:06:43): Tammy’s asking specifically on mobile about alternating between green and red on posts too. It’s also like there’s this weird fickleness in the reporting. I think there’s a lot of variation here that’s using real world data. So if you happen to have a different post that goes viral with different users who have slower connections, like that stuff actually can move the needle.
Because we’re talking about very subtle differences sometimes. You do have to still take some of this with a grain of salt. What you want to do to see is improvement in general. I think if you’re doing that, you’re going to be in good shape.
Do we need a sidebar if we were optimizing for mobile first?
Yes, you need a sidebar, please do not remove your sidebar. If you remove your sidebar, you’re going to have ranking drops because you’re removing strong site wide signals to Google that you had a lot of links on that. I know this site, I know your site specifically Laura. Don’t do that.
We don’t see sidebars on mobile. No, it’s not having the impact on your CLS issues most likely. There are other things in play. Just very quickly there, I want to make sure that we got that answered.
