The past 19 and 20 May, under the Practitioner Web Analytics 2009 and in one of his presentations, Avinash Kaushik We talked about the importance of A/B Testing when increasing the performance of a website.
He A/B Testing It is to test different versions of a web page or an area of the website, and decide which of these versions is the one that gets better results. Although this service is offered from GeaIPC, Few companies decide to use it even when the results are truly spectacular. Indeed, my presentation at the PWA09 It was to explain a case in which the booking engine one of the hotels AC Hotels spent to double the number of bookings due to changes implemented in said engine, testing it with techniques A / B Testing.
And it was also under the PWA09 that Avinash said the 5 things we should all test our site:
- All "call to action" on our website. That is, all the buttons and texts that incite to action. For example: the typical "Call us now ….", or the "Click here to contact", O "Subscribe now and get... ". We should try colors, fonts, different phrases, etc ...
- The steps of the home to inland areas that interest us. That is, is supposed to have determined conceptualize our navigation different routes ... so we must test if the page is designed so that these routes are followed by our visits.
- Las “pain pages” (Avinash and called them and I think it aptly named). Are pages that have a lot of traffic, but also a high bounce rate (people coming directly to them and leave the site without having sailed). Must review and test different versions of the same to get lower bounce rate.
- Try if it improves performance or scroll page without scrolling. This point is interesting. In the study Usability Eyetracking we published in January and indicated that very few of those under 25 we tested, did scroll in the Google results pages (which were those testeábamos in that study). If our website has a user such as target audience, this is a good thing to test.
- "Dramatically different" designs (English is better than in Castilian, but good). Avinash what I meant is that we must try to completely different designs we are using. We may find one that also has "dramatically positive" results.
The good thing about A / B Testing is that it is easy to make and if we use Google Website Optimizer, may not even cost us money. Google offers free Website Optimizer that allows the real-time test as many versions as you want, and then indicates which of the tested versions have better results. The image we can see along these lines shows the control panel of Google Website Optimizer.
- More information about Google Website Optimizer here: Google Website Optimizer
- Information about our service A / B Testing: A/B Testing en GeaIPC
- Blog de Avinash Kausik: Avinash Kausik
- Página web del Practitioner Web Analytics 09: PWA09
Especialista en estrategia en Marketing Digital. Impartiendo conferencias, formación y consultoría desde hace más de 20 años.