One of the more challenging aspects of SEO is proving that the juice is worth the squeeze, especially with eCommerce websites. With so many moving parts involved, it can be difficult to directly attribute improvements in organic performance to SEO initiatives. In turn, convincing your company or client to invest in those initiatives can take a lot of time, persuasion, and frustration on both ends.
Proper SEO testing can help. There are now more resources than ever to quickly and efficiently pinpoint what does and doesn’t work in SEO marketing — and how a business should invest in SEO strategy moving forward. If you’re not using an SEO testing tool yet, you’re missing out.
In this post, we’ll give you all the info you need to start your SEO testing initiative successfully, including:
- What SEO testing is and why it’s important (especially for eCommerce sites)
- How SEO testing works
- How you build and run tests as well as track results with SEO testing tools
- And more
What is SEO Testing?
SEO testing is just like any other technical performance test: You make a change to your site or web page and evaluate how it affects your organic traffic.
In theory, it sounds simple. But, unlike PPC testing or CRO testing, it’s not always easy to trace one SEO-related change to increases in organic traffic. A lot of factors will influence your site’s organic performance at any given time; concluding that a single change made a specific difference is near impossible if you don’t do it right.
For this reason, SEO testing must be conducted precisely. Like any good experiment, you can’t adjust a bunch of variables and expect clean results. You’ll need to choose one change to make, set it up with a control group, and then see what happens.
If you conduct testing on your product pages incorrectly, you may find yourself doing damage control a few weeks later — watching organic traffic fall for your product pages and sales numbers take a hit. Working with an SEO testing tool (or an experienced SEO strategist) makes this far less likely.
The Basics of SEO Testing
Search engine optimization isn’t an exact science. There seems to be a new theory every week on “the next big ranking factor,” and SEO strategies go in and out of fashion all the time.
So, when you sit down to do website SEO testing for your product and category pages, put those out of your mind. The goal is to find out what works well for your site’s performance, no one else’s.
How It Works
We’ve found most eCommerce businesses go about SEO testing in three ways — but only one is the clear winner in terms of efficacy, timeliness, and results.
SEO Serial Test
This is the OG SEO test: You make a change to an entire page type across you site, and then you monitor the organic traffic and rankings for that page type over the weeks and months to come.
We used this exact tactic to test category page descriptions for an eCommerce client. We added short 75–100 word descriptions to category pages (following specific guidelines) and, over the course of 18 months, saw a 46% increase in organic sessions to those pages. While we couldn’t confidently say the increase was solely due to the category description optimizations, the results did give us reason to move forward with this optimization on any remaining category pages.
Time-Based SEO Testing
A time-based SEO test uses the same concepts as a serial SEO test. However, in this test, you’ll only make a change to one page and watch how it performs.
Here’s an example of and the results from a traditional time-based test we ran for one of our clients through SEOTesting.com:
This may seem like the simplest way to test SEO changes, but proceed with caution. Many outside factors can impact your page during the testing period, including algorithm updates, seasonality trends, and other seemingly unrelated website changes. It’s risky to assume positive results came straight from your page edit, especially if you plan to roll out that change across your entire site. Be proactive in eliminating as many variables as possible before implementing an SEO test like this.
A/B Testing or Split Testing
If you want the best results for your eCommerce site, SEO split testing is the way to go.
SEO split testing uses the same concept as its more popular cousin, CRO A/B testing. But, because we’re only serving an audience of one (a search engine), we have to change up the process a bit.
Split testing begins by creating two groups of web pages — the control group, which will remain the same, and the variable group, where we’ll apply the test we want to run. After enough time to generate statistical significance, we’ll evaluate the results. If the variable group’s organic performance improved, the test is a success.
Here’s an example: Say you want to test whether adding “free shipping” to your meta titles will increase clicks from searchers. You’ll create one group of products for your control, keeping their meta titles as is. Then, you’ll create another group (the variable group), and you’ll apply “free shipping” in the same place in the meta title. Analyzing the results will tell you whether or not your hypothesis was right — and whether you should roll out the change across all of your product pages.
Most of the SEO testing tools available today use split testing. Keep in mind: A/B SEO testing is most successful in eCommerce sites with substantial (at least 100k) monthly organic sessions.
A/B testing has often been the most successful testing method for our clients. In one of our most recent tests, we took two groups of blog posts: a variant group of 159 and a control group of 159. The client’s site had H2s and H3s all over the site, including on unimportant text. To test the effects of changing those subheads, we experimented with the variant group by keeping only one H1 and one H2 on the most important headers of the page. The control group we left untouched.
A few weeks later, we took a look at the results:
Overall, we saw an increase in organic traffic to the variant blog pages. We concluded it came from increased rankings from our experiment changes — and that we could reasonably assume the same effect would be felt on other blog pages, as well.
The Features You Should Test
For best results, these accepted ranking factors should be a part of any SEO testing strategy. But, remember: Don’t start changing them all at once, or your test results will be worthless.
As you’re testing changes on your page types, remember to follow SEO copywriting guidelines for the best chance of success.
The Metrics You Should Watch — And How to Measure Them
It’s no use testing SEO changes if you’re not going to measure the results. But exactly which metrics to measure will often depend on the changes you make. For example, if you edit your meta titles or descriptions, changes in clicks to that product or category page may indicate efficiency.
Before you implement an SEO testing strategy, identify exactly which metrics you will monitor. It will make your job easier in the long run.
Proper SEO testing will reveal changes in:
- Clicks
- Impressions
- Position
- Click-through rate
- Number of ranking keywords
- And more
The bottom line: The more data you collect while SEO-testing your website, the better.
There’s another reason we recommend an SEO testing tool: They make tracking and reporting on your results simple. You’ll easily be able to see whether your changes made a difference and, if they did, whether they’re significant enough to roll out across your site. Bonus: Many tools will also report statistical significance for you, indicating whether you can expect similar effects on other pages.
How to Choose Which Technology is Best for You
Before you choose an SEO testing tool, you’ll need to evaluate a few things:
- What kind of tests you want to run
- Your budget
- Your tech stack
Certain tools will require integration with apps, like Google Search Console or Google Analytics. Others will collect the data and analyze the results within one platform, but you may need to standardize results with your other reporting tools for accuracy.
Learn more about your options and find the tool that’s right for you below. Or, if you want our help in selecting the best tool for your eCommerce site, request a proposal here.
The 5 Best SEO Testing Tools
1. SplitSignal
Semrush recently introduced a SEO split-testing tool, which they (fittingly) named SplitSignal. The tool is client-side rendered, focused on the number of organic sessions, and based on a statistical model. Inflow is one of the few agencies invited to test SplitSignal in its current early stages.
How It Works: SplitSignal automates the process of splitting test groups and generating Javascript snippets for the pages to be tested. It also establishes a statistical model based on a site’s existing traffic trends, so you can easily see the estimated impact of changing certain on-page elements.
Tech You Need:
- Semrush account
- Google Search Console
- Google Analytics
- Javascript Snippet
What It Tests:
SplitSignal tests on-page elements, including:
- Titles
- Meta tags
- Structured data
- Ad-block placements
- Alt tags
- And more
Signs It’s Right for You:
- You want to get your test set up quickly and efficiently. SplitSignal takes care of all the work; your developers will simply need to add the Javascript snippet to applicable pages.
- Your site has more than 100,000 visits per month and more than 300 pages on a similar template (for example, product pages).
Signs It’s Not Right for You:
- You have a small eCommerce site, selling only a few products.
- You want a tool that will suggest test conditions for you.
Price: SplitSignal pricing is not yet available in the private beta phase. If you’re interested in this testing program, contact our SEO team for more information.
2. RankSense
RankSense is an automated tool that generates optimized, accurate titles and descriptions for web pages, based on the text on the page. It tracks the performance of the updated pages to confirm the changes are working before applying them elsewhere on your site.
How It Works: RankSense is integrated with Cloudflare to track all the visits to your site from organic searchers and search engine bots. Whenever a visit occurs, the tool scans your page for SEO issues and makes any necessary changes to the page’s HTML, including changes to meta data. Because RankSense works within Cloudflare, no additional Google integrations are required; results are analyzed and reported within the tool itself.
Tech You Need:
- A content delivery network (CDN). Currently, RankSense uses Cloudflare, but integrations with Akamai are coming soon.
What It Tests:
- Meta descriptions
- Meta titles (coming soon)
Signs It’s Right for You:
- You have a large eCommerce site with hundreds or thousands of products, making meta changes difficult to complete manually.
- You don’t have the manpower or expertise to choose which optimizations to make, and you want a program to do it for you.
Signs It’s Not Right for You:
- You have a small eCommerce site, selling only a few products.
- Your site runs on Shopify.
Price: Free to $500 a month, depending on how many URLs you want to optimize
3. ClickFlow
ClickFlow is content intelligence software that performs SEO split testing and offers other automated optimization tools, including a keyword-suggesting content editor and a content-decay discovery system.
How It Works: ClickFlow’s SEO testing tool centers on changes to page titles and meta data. The tool allows you to test single URLs or groups of URLs against each other, depending on your subscription type.
Tech You Need:
- Google Search Console
What It Tests:
- Meta descriptions
- Meta titles
Signs It’s Right for You:
- You’re comfortable with writing testable title tags and meta descriptions on your own.
- You want a tool that does more than just test; it also suggests content and meta data optimizations for you to implement.
Signs It’s Not Right for You:
- You want a program that’s as hands-off as possible and implements site changes for you.
Price: Starts at $109 a month for sites with up to 10,000 Google Search Console clicks a month
4. SEOTesting
SEOTesting allows for traditional time-based and split SEO testing efforts. It also offers optimization guidance through search intent highlighting, long-tail keyword reports, and more.
How It Works: SEOTesting integrates with Google Search Console to record the changes you’ve made to pages on your site and track the results. You can also back-date your tests to compare historical data for better analysis. The tool also annotates Google algorithm updates, so you can quickly see how those updates may be impacting your site and SEO tests.
Tech You Need:
- Google Search Console
What It Tests:
While SEOTesting offers optimization guidance, it doesn’t make those changes for you. In theory, you could track tests to all kinds of meta and structured data through SEOTesting, as long as you make changes to the control and variant groups yourself.
Signs It’s Right for You:
- You are willing and able to make changes to the pages you’re testing.
- You want to test only a few pages at a time.
Signs It’s Not Right for You:
- You wish to test a large number of pages at the same time. Because SEOTesting does not integrate the actual changes to your site, your developers will spend a lot of time implementing those changes across your pages — and, if the test fails, reversing the change.
- You don’t have the manpower or ability to make changes to your site on your own.
Price: Free 14-day trial, after which prices start at $19 a month
5. SearchPilot
SearchPilot performs A/B SEO testing through a streamlined deployment. After you choose a group of pages and the edit you wish to test, SearchPilot splits them into a control and test group for you. The SearchPilot Optimization Delivery Network (ODN) analyzes the organic traffic (taking into account external influences) and allows you to immediately make that change across your entire site, should you want.
How It Works: SearchPilot operates as an ODN, meaning the platform makes changes to HTML files as requests pass through to users and search engine bots from your server (think of it as a more advanced CDN). Because it operates directly on HTML and a cloud platform, SearchPilot is compatible with and operates independently with any CMS.
Tech You Need:
- A content delivery network (CDN). SearchPilot is deployed within your site’s CDN.
- An analytics platform, such as Google Analytics or Adobe Analytics
What It Tests:
SearchPilot can test any aspect of a page’s HTML or http response, including:
- Meta data
- Structured data
- HTML structure
- Page content
- Internal links
- And more
The tool can also combine results across multiple similarly structured sites (“compound testing”) to test the changes across multiple different country websites (cctlds). SearchPilot can also conduct “full-funnel” tests, combining analysis of SEO impact with CRO testing impact.
Signs It’s Right for You:
- Your eCommerce site has at least one section of hundreds of pages or more within the same template and more than 30,000 organic sessions per month in that section.
- While you want to choose the modification to test, you want a tool to implement the changes for you.
Signs It’s Not Right for You:
- You have a small eCommerce site without large numbers of pages on the same template.
Price: Contact SearchPilot for a quote.
The Benefits of SEO Testing on Your Site
Most digital marketers are familiar with SEO best practices and apply them across their site. But SEO testing is the only way you can prove that those tactics are actually working. It’s the closest you can get to beating search engines’ algorithms at their own game.
Rather than applying sitewide changes and hoping for the best (and frantically doing damage control when they don’t), you can use SEO tests to evaluate and validate your efforts ahead of time. If a strategy doesn’t work in your test, you avoid implementing failed efforts across your entire site. But, if they do, you can roll out your optimizations with full confidence in the positive results to follow.
In the long term, using test-driven SEO efforts will save you time and money — and, of course, provide the SEO “proof” that your clients always ask for.
Here’s another example: Inflow and SplitSignal ran an additional test (for the same client mentioned above) on meta description updates (specifically, replacing a custom description with the page’s H1 and subheadings).
While we didn’t see significant changes, the test wasn’t in vain. Because we knew the change didn’t have any effect, we could avoid wasted time and move onto another test.
Measuring Your SEO Test ROI
But how exactly can you measure the money made or saved with SEO testing, and how can you prove to your team or clients that SEO testing is worth the investment?
With a little bit of data, it’s easy to calculate the estimated ROI for if you roll out your tested change across the entire tested page type. You’ll need:
- Organic traffic, Average Order Value, conversion rate, and revenue generated over the trailing 12 months for the tested page type.
- The results from your SEO test — specifically the expected lift in organic traffic to the tested page type.
- A calculator or spreadsheet!
An example: Say our SEO test involved adding “New Styles” to a group of Collections pages’ meta data. Here, we’ve calculated the predicted revenue lift that that client should anticipate after rolling out this winning test.
The test was run on Collections (category) pages and showed a 21% lift in organic traffic. After calculating the predicted traffic numbers and considering the average conversion rate, we’d assume about 10,924.85 customers would convert (give or take a few decimal points). Multiply that by average order value, and you’ll get the new total predicted revenue: $1,431,155.
In this case, we would expect the client to see a 12-month increase in organic revenue ($248,382) after rolling out this change across all Collections pages.
Remember to always take into consideration if you hadn’t run any SEO tests. Think about the costs of the worse results if you hadn’t performed any tests, as well as any unavoidable negative changes that are part of the SEO beast.
Because you may not see discernible changes in your analytics quickly, this is the best way to evaluate your test results. From there, you can decide whether the revenue increase will pay for — and validate — the costs of testing.
Start Your SEO Testing Program Now
As with any SEO experiment, SEO-testing your eCommerce website comes with some risk. You can hypothesize all day about how a change will affect your performance, but you won’t really know until the test is underway. For this reason, many eCommerce owners and marketers shy away from this kind of testing, fearing potential losses.
But, in our opinion, SEO testing is well worth the effort. Yes, you may experience some small losses (or, even better, gains) from running the tests, but the knowledge you gain will shape your SEO strategy moving forward. And, with so many tools readily available, getting started is easier than ever.
Want an experienced hand to guide you through your testing strategy? Our SEO strategists frequently run tests for clients and can do the same for your eCommerce site. Request a free proposal now to learn more.
