The factors that a search engine uses to rank a web page are relevant keywords, meta titles, meta descriptions, anchor text, schema markup and more. You can test each element without waiting for the search engine to give back results.
SEO split-testing allows you to do A/B testing of each of the elements. The three steps to follow are:
- Divide your templated pages like - blogs, product pages or category pages into two segments. A Constant segment that does not change and a Variant segment on which the change is implemented.
- Make changes to the Variant segment.
- Check the difference in the organic traffic between the two segments.
You can make one change to your website and monitor the response to it. You get the search engine’s response to your changes, not the users. You can use A/B testing to test how the changes can affect your website before rolling them out site-wide.
SEO split-testing experiments
This helps to tell the search engine what your page is about with H1 as the heading and H2 and H3 as subheadings. The tags help to make your content readable. You can test these to improve the clarity of your website’s content.
You can make changes like the placement of your title tag, length of sentences, add the H2 and H3 tags and use different keywords.
This is the first thing that the user sees when you get listed in the SERPs. You can test this by adding or removing words like ‘Free shipping’, ‘Buy’, ‘Download’ and more. You can change the character length and add or remove the brand name. You can check how the changes affect your CTR.
This appears below the meta title in the SERPs and is used to convenience the searches to click your result. The variety of SEO experiments that you can run are secondary keywords, star ratings and incentives like ‘Free shipping’. Check how the changes affect your potential customers and the CTR.
Now rich results can be shown in the search results like recipes, reviews, headlines from news stories and more. This can be done by adding schema markup to your content. You can check if adding a schema markup makes a difference and use different schema types and JSON-LD.
This helps to direct the search engine and the users to relevant sections of your website. It helps to improve metrics like bounce rate, pages per session and time on the website. You can experiment with the page being linked to, shallow vs deep links and the position of a link on the page. Test and check the results of the quality of the content and the user experience.
Link anchor text
This is used to describe to the search engines the link to which you are pointing. It helps the crawlers to determine what a page is about. You can experiment with the character length, exact match vs broad match text and format (link over an image).
This is used by search engines to interpret the images and find the right results. You can make changes to the Alt-text and test. Some of the experiments are adding and removing characters, using different keywords and using hyphens vs spacing.
Experiment by adding images, adding internal links to another page and changing the content block above the fold. Check if it makes a difference.
Check if deleting a certain element causes a positive or negative impact. You can try deleting images, breadcrumbs, pop-up forms and social media buttons.
Test if moving certain elements makes a difference to your SEO result. You can try out moving images below or above the headings and moving the content blocks to other places on a page.