A/B Testing - Genel Bakış V1

A/B Testi Nasıl Uygulanır? 

A/B testini uygularken yapılması gereken tek şey farklı içerik ve tasarım modellerini farklı ziyaretçilere yönlendirmek. Aslında farklı örnekleri aynı kullanıcılar üzerinde test etmek de uygun ancak yeni ziyaretçi trafiğinin yüksek olduğu durumlarda ciddi bir geri dönüş alınması zor olacaktır.

Farklı test örneklerine yönlendireceğiniz kullanıcıları trafik yönlendiren sitelere göre ayırabilirsiniz. Örneğin arama motorundan gelen ziyaretçileri bir örneğe diğer kullanıcıları farklı bir örneğe yönlendirebilirsiniz. 

A/B testleri, bir site üzerindeki farklı değişkenlerin karşılaştırılması ve kullanıcı odaklı performansın arttırılması bakımından hayati önem taşıyabilir.

A/B Testi Hangi Alanlara Yapılabilir? 

• İçerik ve Tasarım

The layout of a site has a huge impact on how likely a shopper is to convert. Shoppers decide whether they want to stay on a site within seconds. An easy to navigate landing page is important to drive conversions. Product and checkout pages are equally important, as more shoppers drop off at every stage of the purchase funnel.

RMC Target lets you create unlimited variations of content directly on a webpage. You can modify images, text, layout, and so on. You can move components around the webpage, hide them, add new ones, or change visual components such as colour, text and design. Any changes or modifications you make on your website can be tested and evaluated enabling you to implement the right changes to increase engagement and maximise overall site conversions.

• Test your personalised experiences (Targeting Actions) you deliver to your audiences.

RMC Target allows you to test the Targeting Actions you deliver to your audiences. Targeting Actions allows you to deliver personalised experiences (offers, messages and content) to different audiences. But sometimes you're not sure what type of experience will make the biggest impact for a particular group of visitors. Use A/B testing to test multiple variations of your targeting actions for a single audience. Visitors who meet the Targeting Rules will be randomly grouped to see the different versions of your Targeted Action. RMC allows you to choose a percentage of your targeted audience (control group), who will not see the changes being tested, giving you a point of reference by which to judge your conversion goals (improvement rate). 

A/B testing process 

The correct way to run an A/B testing is to follow a process. The following is an A/B testing guideline you can use to start running your tests:

1. Study your website data: Your analytics tool will help you find the problem areas in your conversion funnel. Look for pages with high drop-off rates or low conversion rates that can be improved.
   
   
2. Identify conversion goals: Conversion goals are the metrics that you will be using to determine whether or not the variation is more successful than the original version. RMC allows you to define your own conversion goals. Depending on your action, a goal can be anything from purchases to clicking a link and newsletter subscriptions.
   
   
3. Generate hypothesis: Once you have identified areas of improvement and a goal, you can begin building a hypothesis aimed at increasing conversions. 
   
   
4. Create variations: Use targeting actions to make the desired changes to an element of your website or to build an experience. Once you have created a targeting action, create different variations of it to test against the original. This might be changing the colour of a button, hiding page elements, or displaying a popup with a special offer.
   
   
5. Select your target audience: Add Targeting Rules to run the test only on visitors that match the specific segment criteria you are specifically targeting, rather than to all visitors to your site. 
   
   
6. Run your testing campaign: Launch your testing campaign and wait for your visitors to participate. Visitors will be randomly appointed to either the control group or a variation of your targeted action. Each visitor interaction is counted and measured to determine how different variations perform.
   
   
7. Analyse test data: Analyse your A/B test results, and see which variation delivered the highest conversions and whether there is a statistically significant difference.
   
   
8. Implement winner: If your test result indicates that one of the changes significantly improved conversion, declare the variant as the winner and go ahead with its implementation. If the test remains inconclusive, either allocate more time or go back to step 3 and rework your hypothesis.