Rethink Commerce Blog

3 factors that have impact on Conversion Rate

Posted on November 18th, 2009 by

For the last couple of months there has been a frenzy here in our web marketing department about A/B testing and Conversion Rates. Why the frenzy? Basically because all elements came into place: we developed a high performance A/B testing module within the eCommerce platform, we finished rolling out Omniture Site Catalyst on the shopping carts and we improved the template editing areas for all the accounts so it’s easier than ever to start testing templates.

We have finished quite some tests so far, some with better results than others, so we are pretty confident on making some bold statements on the 3 factors that impact conversion rate in our experiences.

Do you have quality traffic?

Among our clients we have vendors who have chosen to use the default and still have very high funnel conversion rates (more than 60% of the visitors entering the cart finish a transaction) and some with highly customized shopping carts that don’t even come close to 1% funnel conversion rate. Yes, the main factor that impacts the conversion rate is the quality of traffic.

When sending the visitor from your website or software product to the shopping cart make sure this is what he expects. Even better, let him know that he is going to be taken to a third party shopping cart on a secure environment. Maybe they would buy your product, but trying to force them to do it will most likely get them away from your website. So quality traffic means visitors that when getting to the shopping cart this is where they expect to be (they know the price, selected products, and additional options are not pre-selected).

Forcing users to the shopping cart just ain’t the right strategy. Most of them don’t return to your website but rather exit for good and you lose any chance in persuading them to engage with your website or your products.

Customized beats default

We’ve been working hard on making improvements to the default shopping cart provided by Avangate and we are continuously getting stuff rolled out in order to increase the conversion rate. Whenever we did an A/B testing between the default shopping and a customized one, the latter always won by around 2 or 3 percentage points at least. Sometimes all we did was to change the header of the cart and place the client logo and even that did much better than the default shopping cart.

Going deeper into the process got us even better results.  The fact is that customizing the cart in a way that the user feels he is still in the same website always rocked the conversion rate. We are dealing with tests on 3 clients right now on which we worked on making the cart as similar to the website as possible and we have great results on all of them. Checkout the before and after for the above cart. Which one do you think is winning?

Well done Cross Selling

Well done cross selling will most likely decrease the conversion rate of a shopping cart. Yeah, you heard me right. However, well done cross selling will increase the average order value, therefore the revenue. Checkout the following charts and see how cross selling has impacted the sales of one of our clients.


What do I mean when I say well done cross selling? Check out the following implementations for 2 other clients of ours.


We will keep you posted with future developments, but till then I would appreciate if you would let us know what your testing ideas are.

0.00 avg. rating (0% score) - 0 votes
0.00 avg. rating (0% score) - 0 votes

Claudiu Murariu

Digital Analyst Consultant

Please note: Comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.

Comments (13)

  • Will says: November 18th, 2009 at 10:38 pm

    You mention percentage improvements!!? Were the differences statistically significant? What was the sample size??

  • Tweets that mention 3 factors that have impact on Conversion Rate | Avangate Blog - Software Sales Tips | Software Business Blog -- says: November 19th, 2009 at 5:48 am

    […] This post was mentioned on Twitter by Adriana Iordan, Avangate. Avangate said: New post: 3 factors that have impact on Conversion Rate […]

  • Claudiu says: November 19th, 2009 at 11:17 am

    Hi Will. Yes, we always try to go for statistically significant data. If before we were just going for big data samples (above 10 000) now we also double check with the chi-square method. Thanks for your comment.

  • Our first Software Selling Conference in China | Avangate Blog - Software Sales Tips | Software Business Blog says: November 26th, 2009 at 6:44 pm

    […] Our first Software Selling Conference in China Black Friday & Cyber Monday 3 factors that have impact on Conversion Rate […]

  • Claire says: November 27th, 2009 at 5:54 pm

    Sorry I do not see in the graph ( the difference in Checkout conversion rate between A and B in order to back up the statement “Well done cross selling will most likely decrease the conversion rate of a shopping cart.” The blue columns look the same height to me

  • Claudiu says: November 30th, 2009 at 11:06 am

    Hi Claire. The difference in checkout conversion rate between A and B in that case was about 0.5% for over 15 000 tests. Basically we tried to optimize the cross sell display in a way in which it would have a very small impact on conversion rate. Clients don’t like to see big drops in there. :) The winning scenario for this client was by using only one product as cross sell directly in the checkout form.

  • Claire says: November 30th, 2009 at 12:46 pm

    ok thanks very much

  • Carl says: December 23rd, 2009 at 4:48 am

    If you want your data to be believable, include error bars. SEM is a good error bar. Also some stats would be nice. Don’t assume your audience doesn’t know a thing or two about statistics.

  • Claudiu says: December 30th, 2009 at 3:18 pm

    Hi Carl,
    Thanks for your message and suggestion. Yep, you are right and for the following articles I will ask our clients permission to add some technical details or stats about the measurements. For the tests in the presented article I confirm the significance level being at least 95%.

  • Avangate in 2009 - Happy New Year! | Avangate Blog - Software Sales Tips | Software Business Blog says: December 30th, 2009 at 4:01 pm

    […] We are particularly happy with the progress made with the Avangate shopping cart’s conversion rates. Using advanced analytics applications such as Omniture and extensive A/B testing, we obtained control over every tiny detail of the shopping cart – and this flexibility translated into higher conversion rates. […]

  • Rex Dixon says: February 17th, 2010 at 8:15 pm

    It appears like you have some great examples. Would you mind sharing a few that don’t violate any privacy issues with our site A/B Tests? ( ) – we’d love to see them, as well as our testing community.

  • kit says: September 20th, 2013 at 5:46 pm

    Tremendous issues here. I am very happy to see your article. Thanks a lot and I am having a look forward to contact you. Will you please drop me a e-mail?

  • Irina Mihaila says: September 24th, 2013 at 4:16 pm


    You can drop us an email at should you have questions or inquiries.

    Have a nice day!