What is SEO Testing: A Guide to Effective SEO Testing
Table of content
There’s no denying how important a good website is for businesses. These digital pages provide users a way to find businesses they might not normally discover while also allowing many business models to exist that couldn’t before the digital age.
However, as people build more websites, the need to stand out matters. While great content and page organization helps, testing how optimized your website is for search engines matters, too. Let’s look at what goes into search engine optimization (SEO) testing for websites and find out how you can start testing website SEO for your business.
What is SEO Testing?
Search engine optimisation testing (SEO testing) refers to the practice of systematically testing various elements and/or strategies on a website with the goal of improving its SEO performance.
For example, if your website had an article that was ranking around Position 15 (approx. the middle of Page 2) for your main query, and you hypothesised that a meta title change could give it the boost it needed to get onto Page 1 of the SERPs, you would be conducting an SEO test.
SEO refers to the steps website authors, and administrators can take to improve their website’s ability to show up on search result pages across certain keywords. These keywords come from the words entered by users into the search bar of a search engine. Websites taking the steps needed to rank high on search pages receive more traffic due to their likelihood of being seen by users.
The Benefits of SEO Testing on Your Site
Almost everyone working in digital marketing understood a long time ago that SEO helps a website’s ability to generate traffic. Regardless of what that website wants traffic for, utilizing SEO brings more viewers to your website.
The reason SEO does this comes down to organic search traffic. Many websites utilize practices that encourage users to become repeat visitors. Newsletters, subscription models, habit-forming design features, and so on all help to convert new users into returning ones.
However, you cannot have a returning visitor without first bringing them to your site as a new visitor. Given how relevant search engines are to the average user’s web experience nowadays, with many browsers offering a search engine as a default home page, it is clear that building your website for SEO best practices is the right choice.
It’s also worth noting that having a well thought out and executed SEO strategy is key to the success of almost all businesses who operate online, no matter the size of the company or the size of the team.
One of the main reasons for this is its innate ability to bring new users though to your website, as we discussed above. Other areas of digital marketing (for example a CRO team or a newsletter marketing team) would then be tasked with taking these brand new users and converting them into, firstly, paying customers, but also returning visitors.
SEO can also form a crucial part of a website’s sales funnel in its own right. It is commonplace now for a user to search a query in Google (or another search engine such as Bing), visit a website once and immediately become a paying customer. This is especially true for ecommerce sites. This is just another reason you need to ensure your website’s SEO is kept in check at all times.
Who Should Do SEO Testing For Their Website?
Truth be told, almost every single website on the internet (providing it is active, regularly publishes new content and gets organic traffic) can benefit from SEO testing in some way, shape or form.
The extent of your SEO testing, and the different SEO testing methods you use, will depend on your website size, the tools you have available and the amount of organic traffic your websites receive.
Smaller websites, who do not get much organic traffic, can benefit from SEO testing. This is because, essentially, we have all been SEO testing since “SEO” became part of mainstream digital marketing work. We change something on a website (whether it is writing new content, changing internal links, acquiring new backlinks, etc) and we wait to see what impact this has on our traffic and rankings. You can still do this, just ensure you are noting down (or using a tool that does this for you) what you are changing and the impact this has over a more-determined period of time.
Larger websites, on the other hand, can arguably have a bigger use case for SEO testing because they have more resources available, and have more organic traffic available to test. This makes it more likely that statistical significance can be achieved with most SEO tests a larger site does. Larger sites, especially large ecommerce sites, can also receive the benefit of experimenting with A/B split testing, which we will discuss in more detail a little later on.
So whilst all websites can make good use of SEO testing, if you have a larger website then you are (probably) able to utilize it a little better, especially since they have a better chance of getting statistically significant results.
If you don’t have a large website, don’t worry. Smaller scale SEO testing can still be incredibly useful! You will just need to be wary that correlation might not equal causation, so you will need to be a little more reserved in making large-scale website changes based on the results of SEO tests.
The Basics of SEO Tests
Before diving into the best practices for SEO testing or tools like Google Analytics, Google Search Console and other tools, let’s look at some of the foundational concepts of SEO testing:
The Features You Should Test
Not at parts of a website will contribute equally to improving your website. Rather than testing every facet of your website, there are several features that we know can improve your SEO ranking.
Some of those features include:
- Meta titles: Also called the title tag or page title, this HTML element helps web pages like landing pages get recognized by search engines.
- Meta descriptions: These short descriptions display underneath links in search result pages and help further distinguish a web page or blog post.
- Structured data: The way your website organizes its data can help or hinder your website’s readability to search engines
- Copy length: The length of the content on the web page can affect how readily search engines parse through it
- Titles and headings: Relevant titles and headings with the proper HTML tags make a blog post or similar website content easier to read and more organized for search engines
- Images: Pictures with descriptions further improve a search engine’s ability to read a website
- Videos: Much like with images, videos and relevant descriptions help web pages rank higher
- Internal links: Links to other pages on the website show that you want to relate everything together instead of creating random pages
- External links: Links from other websites to your websites. This can give your website more authority in the eyes of Google & its crawlers, so you may rank better
Here are 28 SEO tests/experiments you can test on your own website.
The Metrics You Should Watch
Without measuring your SEO performance results, you won’t know if the changes you make to the features listed earlier are effective.
- Clicks: This metric can include clicks from organic traffic or through hyperlinks on a target web page
- Impressions: This value indicates how often your website shows up in search results
- Result page position: Also called your SEO ranking, this position shows which slot you show up in on average for a target keyword or search query
- Click-through rate: This metric measures how often people click on your link from another page or the results page of a search engine
- Number of ranking keywords: Websites can rank for more than one keyword, meaning that this value will show you how many keywords your website relates to
You won’t measure the same thing for every change you make or test you run. Identifying what you want to happen with the changes you make will indicate what metrics you’ll need to look at to qualify your results.
It’s also worth noting that not all SEO tests will impact “SEO metrics” like clicks, impressions and CTR. Some tests can be performed on-page with the view of impacting “on-page metrics” like time on page and engagement rate in a positive way. You may find this has an indirect benefit to your SEO metrics, anyway.
Best Practices for Website SEO Testing
Not every practice in SEO testing will give the best results. If you want to make the most out of your SEO testing, you’ll want to follow these steps:
Pick Pages or Keywords Search Engines Receive
Ideally, you want to rank your website along high-traffic, low-competition keywords. Given how many websites are out there now, finding these keywords will be almost impossible. So, the next best thing you can do will be to pick keywords that have a good amount of traffic and relate to what your website is about.
The reason for choosing these keywords comes down to sample size. You want to run your tests on keywords that will provide enough visible difference between your two methods without running into the risk of your margin being accounted for solely by statistical significance.
In other words, you don’t want keywords that get a small amount of traffic because you won’t be able to tell if your changes or random chance made the difference.
This all comes down to ‘statistical significance’. If you use high-traffic pages to test, you are more likely to get a statistically significant result, meaning you can be more confident in pushing this live to the rest of your website, if this is what you were testing for.
For example, if I was testing to see if blog posts could be improved with the addition of an embedded YouTube video at the beginning of the article, it would be wise to test this on an article (or group of articles) that have a lot of traffic, leading to a more statistically significant result. That way, I can be more confident in replicating this on the rest of my blog.
You can run tests without statistical significance, but ensure you take the result with a pinch of salt. Correlation does not always mean causation.
Form A Clear Testing Hypothesis
Before you can set out with your testing, you need to know what you want to achieve with your new strategy first. Going into SEO testing without a clear direction of what you want to do means you won’t get results related to your preferred outcome.
Now, that doesn’t mean you’ll get the results you want with testing. But, constructing your test parameters with the goal you have in mind will make it clear what your results say when you get them.
For example, you might want to change the verbiage of your titles to improve their emotional appeal and thus generate more clicks. Already with that sentiment, you know what variable you want to change and what you want the outcome to be. From here, you’ll create the titles you want to use and the content to go with it, and wait to see how the new pages do over the next month.
Actionable hypotheses give tests a clear direction to go and outcomes that relate to the overall goal.
Test One Variable at a Time
Part of the scientific method is that you should study one variable at a time. Doing so allows a test to clearly show what effect that variable has on outcomes for a test. SEO testing can use the same principle for the same reason as a scientist - to see if your hypothesis holds.
This concept exists in most fields looking for objective results. Even economists and lawyers refer to this concept with the phrase ceteris paribus, Latin for other things equal. Keeping things consistent save for one variable ensures you receive results with no complicating factors.
Let’s say, for example, you made some changes to one of your product listing pages. You added 2 internal links to the page from other relevant pages on your site, and you also re-worded the copy on your page. The impressions and clicks went up, following the test. Great! But how do you know if it was the internal links, or the copy change? Or both? Testing one variable at a time is crucial to understand what works for your website.
Know What You Want to Optimize For
Knowing which outcomes will give you the best results for your end goal will give you better results than just applying a generic SEO strategy to your website. Knowing how a user’s query affects your website and the intent behind the query matters more than you may know.
For example, optimizing for low-volume keywords related to your industry could be the difference between ranking low and ranking high on a results page. If you know that users searching those keywords tend to purchase a good or service not long after searching, these low-volume keywords might be the best thing to optimize for.
High-volume keywords are another story. SEO testing on these keywords might allow a website to focus on improving its click-through rate or impressions rather than its search engine rankings.
The point here is that you need to know along which parameters you want to improve your website’s SEO. Understanding how keywords and query motivation go far in developing an SEO test.
Utilize Split Tests Where Possible
Also called A/B testing, split tests refer to SEO tests in which you take two groups of similar pages and test out a hypothesis on one group to see how this change impacts the entire group.
Ecommerce websites are the perfect use case for A/B testing as they have a lot of similar pages, with similar traffic levels. Think product listing pages and product display pages! If I wanted to test a change to the layout of my website’s product listing pages, an A/B test would be the best SEO test to use.
Folks with a marketing background will notice that this is different from standard AB testing, where two new concepts run at the same time. This difference comes down to how Google indexes websites.
Google doesn’t like it when websites have multiple groups of webpages on the same site indexed for almost the same thing. Spam sites did this in the past to try and trick Google into ranking their pages better, forcing Google to knock websites that try this strategy.
So, instead of running multiple new groups, you compare new pages or blog posts against old ones to see how they perform. It isn’t a perfect reproduction of traditional AB testing, but it helps show how your new strategy affects your metrics all the same.
Test For The Right Duration
Finally, the last practice you’ll want to consider is how long the test duration will be. Testing a new idea for too long can hurt your efficiency or discovery rate, while a short test won’t generate enough results to be worth the time and resources invested.
In general, four to six weeks is the industry average for SEO testing. Ryan Jones, Marketing Manager at SEOTesting, says something similar. He says… “We recommend 8 weeks for any SEO test you are running. This gives both your control group & your test group (if you are using this type of SEO test) enough data to really compare whether the change you have made to your site has had an impact, and whether the result you have seen is statistically significant. If you do not have 8 weeks available for an SEO test, the absolute minimum we’d suggest for an SEO test is 2 weeks.”
SEO Testing Methods
With the best practices explained, you should also think about what type of testing method you want to use for your SEO testing. These methods measure different metrics, meaning they won’t fit in all testing situations.
Still, here are some of the common methods used in SEO tests:
SEO Serial Test / Single Page Test
SEO serial tests (also called single page tests) are among the first tests created for optimizing websites. These tests involve making one change to a single page type across a website and monitoring the changes in organic traffic to those pages over time. Testers also tend to review changes to page ranking in these tests and include that data along with the changes to organic traffic.
In general, these tests take a long time to conduct. Commonly, months of testing time and data collection are important to ensure that the changes work the way they were intended and aren’t the result of a fluke or Google algorithm updates.
Still, these tests are useful to those with the time and size to commit to them. Something as simple as a new title format or the addition of meta descriptions in these extended tests can lead to massive changes in organic search traffic to the page group.
Time-Based SEO Testing
For a finer tuning of your website’s SEO, you can look to time-based tests. Rather than changing all the pages in a page group, you change one page instead and track its progress over time.
This testing method might seem like a major timesaver over serial testing since it only involves the use of one web page. However, outside factors can come into play when performing a time-based test.
For example, updates to the algorithm, seasonal trends in search traffic, and other changes made to the website during that period can affect the results of this test. You may decide to make one change to a web page that ends up receiving a lot of internal links back to it, resulting in inflated numbers as the web page receives more internal visitors rather than organic ones.
Time-based tests are useful for small-scale changes or initial ideas. Scaling up from a time-based test generally involves going to a serial test or moving on to split testing.
A/B Testing or Split Test
Finally, when running SEO tests, your final popular option is A/B testing or split testing. This testing method tends to be the go-to method for many website optimizers since it offers comparative results without requiring changes to entire page groups.
A split test involves serving up two web pages, or groups of pages if desired, to the search engine. One group represents a control, which reflects the SEO practices of your website currently. The other group contains the change you think will improve your SEO metrics and rankings. You then track the traffic of these groups over time to see how they perform.
This type of testing doesn’t work on all websites, but it functions well for many. eCommerce sites and other service websites tend to receive clear results when performing AB testing, making it a worthy investment for those businesses.
How to Choose Which SEO Checker is Best for You
SEO tools come in many forms. However, an SEO checker can run SEO tests on your website and provide feedback about what to do to improve them. These tools tend to provide an SEO score for content you submit. This SEO score is a numerical summary of how well your content performs in organic search queries.
However, not every SEO tester is equal to others. Some focus on specific industries or search engines to improve your website’s SEO. Google Analytics provides useful information for ranking in Google but not for other search engines like Bing.
Choosing the best SEO tool for your website depends on several factors:
- Testing options: Not all SEO testing tools provide the same level of SEO testing, meaning you have to match an SEO checker with your testing needs
- Budget: Whether you use free tools or premium services, an SEO checker won’t help you if it falls outside your business’s budget
- Tech integrations: Many SEO checkers have integrations with other analytics platforms like Google Analytics and Google Search Console, making aligning these integrations with your other services matter
So, the best tool to use for your website will depend on how it fits into your business's unique circumstances.
Measuring Your SEO Test ROI
With your SEO test in place, it’s time to see how the SEO test performed. However, there’s more to improving SEO than just watching numbers go up. After all, businesses wouldn’t care about SEO if it did not provide some tangible benefit to the company or website as a whole. This is why calculating your SEO ROI is important.
SEO ROI just refers to your return on investment for engaging in SEO testing. While the term comes from economics, the phrase here just means how much money the SEO testing provides as a result of suggesting changes or discovering better ways to rank in results pages.
At its core, calculating your SEO ROI requires knowing how much you spent engaging in the test and how much extra revenue came in, thanks to the changes from the test. Whether you calculate this every month or throughout some other period will depend on your business and how long the SEO test took.
After these calculations, you’ll have a number presenting how much more you made thanks to the SEO testing results. You and your business can then decide if the money spent on SEO testing was worth it or not. Generally, large websites engaging in a few optimization tweaks can see double-digit percentage differences in their traffic and revenue, though.
Frequently Asked Questions
Here are the answers to some of the common questions out there about SEO testing and related topics:
How do I test my website for SEO?
The easiest way to SEO test your website is with an SEO testing tool. Most of these SEO tester tools use your Google Search Console account to track search results and rankings for you. There are dozens of SEO checkers out there, meaning the best choice for you will depend on you and your website’s needs.
How do I know if SEO is working?
Without looking at metrics, you won’t know if the changes you made to your website affected your SEO or not. Looking at data such as impressions, click-through rate, page rankings, and more will show you if your SEO strategies worked or if there is still more work to do to improve your SEO.
How do I audit my website for SEO?
An SEO audit is different than SEO testing. SEO audits are a process where a person or service reviews your website’s SEO strategy and determines what your SEO issues are. SEO testing involves taking two different strategies and comparing their results to one another.
Both automated tools and paid services can provide an SEO audit, but some research from a website author or admin is needed to select the best one for your business.
SEO testing offers website owners a way to improve their traffic through increasing organic search visitors. The testing methods used in SEO testing show ways for website owners to change how they structure or craft their websites, improving the chances search engines place them higher on search result pages.
While not all SEO tests will generate positive results, the process of making changes tends to work out in the long run. When combined with online tools and services dedicated to SEO, many websites can find some way to improve their rankings with a little experimentation.