|Call to Action - Increasing Conversions|
What works for you?
| 3:21 am on Mar 18, 2005 (gmt 0)|
I've been looking for ways to increase the visitor to sales conversion ratio. One thing that Ive been looking at is the wording for the "Add to Basket" link. You know Add to Cart, Buy Now!, Add One to Basket and all the rest.
Has anyone done a test to see which converts better for them?
| 4:32 pm on Mar 18, 2005 (gmt 0)|
Test, test, test.
In a recent trial of two different home pages on our site, one home page/landing page led to a 40% higher conversion rate than another.
The pages differed only in what products from our line we chose to show the visitors when they first came.
Our testing methodology was to assign a cookie to each new visitor of the site. One value the cookie contained was an automatically incrementing "user number." Those visitors who were assigned even user numbers saw one home page, and those that were assigned odd user numbers saw the other.
When a sale was completed, the user number from that cookie was read and stored with the sale record. At the end of the test, we pulled all the orders, threw out those for people who did not allow us to set cookies and for visitors who first came to the site prior to the start of the A/B test. What was left was a simple count of the number of purchasers who bought from page A and the number that bought from page B.
The net is, there are lots of "best practices" sorts of things that you'll hear people talk about. But, when it gets right down to it, the only way to figure out what is going to covnert better for you is to test different approaches and see the results.
| 5:52 pm on Mar 18, 2005 (gmt 0)|
I haven't seen much of a difference in the wording on the Buy Now/Add to Cart/etc. but when I changed the color of the button from a navy blue to a bright ugly pink, the sales jumped by 32%.
| 6:18 pm on Mar 18, 2005 (gmt 0)|
We've done lots of A/B testing to improve conversions and have learned much. Best advice: test test test. The sample above of using even/odd cookies is a good one and is basically how we handle things also.
And don't stop with one page. We test our home page, specials pages, product details pages and the check-out page itself for variances.
| 1:37 am on Mar 20, 2005 (gmt 0)|
I suspect your buy button was hard to find, and the ugly pink was so ugly, nobody could miss it, and this increased your sales.
You be surprized how many websites I abandon, because I could not easily find the buy button.
Generally, if you have a well developed site, and don't do anything contrary to good web design techniques, making design changes will only reflect minor changes in conversion.
| 3:08 am on Mar 20, 2005 (gmt 0)|
|because I could not easily find the buy button. |
I think this is the number one mistake I see on any ecomm site. Too many designers want to make the buy button melt into the design when it should be popping like crazy from the design. Above the fold and obvious should be a mantra for any site with a buy button. Drag your customer by the nose through the chaeckout process.
As stated above in other posts, the size, color and the placement of the buy button will prbably have more of an effect than the words. But test the words, too. Everything has an effect.