Google’s John Mueller and Martin Splitt discussed making changes to a web page, observing the SEO effect, and the importance of tracking those changes. There has been long-standing hesitation around making too many SEO changes because of a patent filed years ago about monitoring frequent SEO updates to catch attempts to manipulate search results, so Mueller’s answer to this question is meaningful in the context of what’s considered safe.
Does this mean it’s okay now to keep making changes until the site ranks well? Yes, no, and probably. The issue was discussed on a recent Search Off the Record podcast.
Is It Okay To Make Content Changes For SEO Testing?
The context of the discussion was a hypothetical small business owner who has a website and doesn’t really know much about SEO. The situation is that they want to try something out to see if it will bring more customers.
Martin Splitt set up the discussion as the business owner asking different people for their opinions on how to update a web page but receiving different answers. Splitt then asks whether going ahead and changing the page is safe to do.
Martin asked:
“And I want to try something out. Can I just do that or do I hurt my website when I just try things out?”
Mueller affirmed that it’s okay to get ahead and try things out, commenting that most content management systems (CMS) enable a user to easily make changes to the content.
He responded:
“…for the most part you can just try things out. One of the nice parts about websites is, often, if you’re using a CMS, you can just edit the page and it’s live, and it’s done. It’s not that you have to do some big, elaborate …work to put it live.”
In the old days, Google used to update its index once a month. So SEOs would make their web page changes and then wait for the monthly update to see if those changes had an impact. Nowadays, Google’s index is essentially on a rolling update, responding to new content as it gets indexed and processed, with SERPs being re-ranked in reaction to changes, including user trends where something becomes newsworthy or seasonal (that’s where the freshness algorithm kicks in).
Making changes to a small site that doesn’t have much traffic is an easy thing. Making changes to a website responsible for the livelihood of dozens, scores, or even hundreds of people is a scary thing. So when it comes to testing, you really need to balance the benefits against the possibility that a change might set off a catastrophic chain of events.
Monitoring The SEO Effect
Mueller and Splitt next talked about being prepared to monitor the changes.
Mueller continued his answer:
“It’s very easy to try things out, let it sit for a couple of weeks, see what happens and kind of monitor to see is it doing what you want it to be doing. I guess, at that point, when we talk about monitoring, you probably need to make sure that you have the various things installed so that you actually see what is happening.
Perhaps set up Search Console for your website so that you see the searches that people are doing. And, of course, some way to measure the goal that you want, which could be something perhaps in Analytics or perhaps there’s, I don’t know, some other way that you track in person if you have a physical store, like are people actually coming to my business after seeing my website, because it’s all well and good to do SEO, but if you have no way of understanding has it even changed anything, you don’t even know if you’re on the right track or recognize if something is going wrong.”
Something that Mueller didn’t mention is the impact on user behavior on a web page. Does the updated content make people scroll less? Does it make them click on the wrong thing? Do people bounce out at a specific part of the web page?
That’s the kind of data Google Analytics does not provide because that’s not what it’s for. But you can get that data with a free Microsoft Clarity account. Clarity is a user behavior analytics SaaS app. It shows you where (anonymized) users are on a page and what they do. It’s an incredible window on web page effectiveness.
Martin Splitt responded:
“Yeah, that’s true. Okay, so I need a way of measuring the impact of my changes. I don’t know, if I make a new website version and I have different texts and different images and everything is different, will I immediately see things change in Search Console or will that take some time?”
Mueller responded that the amount of time it takes for changes to show up in Search Console depends on how big the site is and the scale of the changes.
Mueller shared:
“…if you’re talking about something like a homepage, maybe one or two other pages, then probably within a week or two, you should see that reflected in Search. You can search for yourself initially.
That’s not forbidden to search for yourself. It’s not that something will go wrong or anything. Searching for your site and seeing, whatever change that you made, has that been reflected. Things like, if you change the title to include some more information, you can see fairly quickly if that got picked up or not.”
When Website Changes Go Wrong
Martin next talks about what I mentioned earlier: when a change goes wrong. He makes the distinction between a technical change and changes for users. A technical change can be tested on a staging site, which is a sandboxed version of the website that search engines or users don’t see. This is actually a pretty good thing to do before updating WordPress plugins or doing something big like swapping out the template. A staging site enables you to test technical changes to make sure there’s nothing wrong. Giving the staged site a crawl with Screaming Frog to check for broken links or other misconfigurations is a good idea.
Mueller said that changes for SEO can’t be tested on a staged site, which means that whatever changes are made, you have to be prepared for the consequences.
Listen to The Search Off The Record from about the 24 minute mark:
Featured Image by Shutterstock/Luis Molinero
