Split testing.

What Type of Landing Page Performs Best for Product Related Keywords?

Choosing the right landing page to direct the traffic from your PPC campaigns to is extremely important, particularly for online retailers. The wrong landing page could cause users to leave an online store without converting. Additionally, AdWords will calculate the quality scores of your keywords partly based on the relevance of the content on your landing page, which directly influences click costs. Consequently, optimizing the landing pages you direct your visitors to will have a direct impact on key efficiency metrics such as ROAS or ROI.

For some keywords there might be more than one landing page that fits and therefore you need to decide which one to use. This is very common for retailers, as product specific keywords could land on a site search page, showing a limited number of highly relevant products, or on a higher level category page with showing more products. The benefit of landing the user on the category page is that they then have more variety of choice and may see other products they like the look of. Landing on a site search page could shorten the users’ journey to the product they were looking for, which can often improve conversion rate.

So the question is: should retailers restrict the results on the landing page to show only a limited number of products which are highly relevant to the users search query, or is it better to display a greater variety of products to give the user a broader selection?

To answer this question we set up a split test for a fashion retail shop.

We focused on specific products and used popular sports shoe models like “Nike Roshe Run” and “Nike Internationalist” for the test. Two identical AdWords campaigns were created. In the first campaign we landed the keywords on a site search results page, referring to the specific product. Consequently, users entering one of the keywords were shown a landing page containing only the specific products they searched for (e.g. Nike Roshe Run products). In the second campaign, users were directed to the category of the search term for the entered brand (i.e. a search for “Nike Roshe Run” would land on a page displaying all Nike shoes).

Test: What Type of Landing Page Performs Best for Product Related Keywords?
Figure 1: Landing Page Test for Specific Products

The two campaigns were scheduled to run in turn, i.e. campaign one from 0am to 2am, campaign two from 2am to 4am, campaign one from 4am to 6am and so on. After 3 weeks, the scheduling was swapped (campaign two ran from 0am to 2am, campaign one from 2am to 4am and so on).
This way, we made sure that the results of the test don’t get blurred due to external effects (e.g. hour of day, daily fluctuations in performance…).


Test: What Type of Landing Page Performs Best for Product Related Keywords?
Table 1: Results for the Specific Product Test

Which landing page performed better?

Looking at the conversion rate (CR), we saw that there was only a small difference (1.73% for campaign 1 showing only specific products in comparison to 1.82% for campaign 2 which directed to the category page).

Other key metrics, however, indicated that linking to categories instead of showing specific products provides better results. Average basket value (ABV) was higher (£ 116 vs. £ 109.) as well as the rate of new customers (NC Rate, 44% vs. 21%, Account average 30%).

Additionally, the quality score (QS) was significantly higher for keywords linking to category pages (7.74 vs. 7, +11%), which lead to a better average position (3.4 for campaign one showing specific products vs. 3.1 for campaign two showing the superior brand category).


Although no solid conclusions with regard to conversion rate could be drawn, other key measures like bounce rate, average basket value and new customer rate indicate that performance is stronger when traffic lands on a category page with a wider selection of products, than a site search page with fewer products.

The new customer rate result is particularly interesting, suggesting that customers who have not shopped at the site before are more likely to buy for the first time when they see a wider selection of the products that the site offers. Also interesting is the bounce rate, suggesting that when users are shown a greater number and variety of products, they are less likely to leave the page immediately. While this test was not extensive enough to categorically prove it, the results indicate that using category pages could be an effective method of increasing new customer recruitment without sacrificing overall conversion rate.

Additionally, the quality score results could indicate that AdWords puts a high value on the number of products displayed on the landing page.

Do you agree with our findings? Have you experienced similar or different results? Let us know in the comments.


Manuela Lang

Manuela Lang is a PPC Manager for crealytics and has 2+ years of experience in PPC and 4+ years of experience in Consulting and Data Analytics.
She has been working across different industries with a focus on markets in UK and DACH.