English | Nederlands
Home eStrategy Web Analytics Website Optimization eMarketing About ClickValue
Contact
Strategy Development
Strategy Report
KPI Review
Dashboards
Audit
Vendor Selection
Tool Implementation
Training
Analyst Outsourcing
Report Service
Google Analytics
Expert Review
Usability Testing
A/B & Multivariate Tests
Landing Page
Conversion Funnels
On Site Search Function
Website Optimizer
Strategy Development
Campaign Optimization
Website Findability
Google AdWords
Banner Optimization
Contact
Vision
Team
Client List
A/B Tests and Multivariate Tests

What are A/B tests and multivariate tests?
Decisions on website design frequently involve choices between dozens of alternatives. To add a simple button, choices have to be made about color, borders, shadows, text and its position on the page. In the case of complex matters like the message communicated by the page, the relation between the elements, the visuals used, the typeface and the call-to actions will have to be considered.

AB Testen


A/B and multivariate tests enable you to determine which solution will be the most effective. They provide you with clear evaluations of the design and pave the way to a more successful web channel.

How does A/B testing work?
An A/B test presents different versions of the page in question to two or more groups of users. The web statistics for each version, such as the click-to-next-page and conversion percentage, reveal the most successful version. By conducting new A/B tests, that version can be optimized further

How does multivariate testing work?
Multivariate tests offer fully computerized presentations of different alternatives for your website and evaluate which alternative will lead to the best result. If every variable on the website, such as background color, text, heading and image, had only four different options, this would still mean a total of 256 alternatives. Computerized testing makes it possible to optimize Landing Pages, homepages and Check Out funnels using thousands of alternatives.

What to test?
All forms of visitor interaction can be optimized through A/B testing and/or multivariate testing. Depending on your goals, you might consider the following tests:

If you want to achieve:

 

Run tests for:

More leads for registrations for newsletters or alerts, downloads of white papers or contact requests.

 

• More concise registration pages
• More extensive descriptions of the services
• More prominent privacy statements
• Attractiveness of typeface

More turnover by attracting more customers or generating more turnover per order.

 

• Image type
• Wording of the proposal
• Discounts, warranties
• Purchasing suggestions

More savings because customers increasingly use the online self-service rather than placing a telephone call.

 

• FAQ descriptions
• FAQ descriptions
• Feedback after input
• Handling of forgotten passwords

What to expect?
Based on your online targets, we will review the website’s most important points of interaction. We will subsequently set up the tests together with your web team and optimize all points of interaction. In doing so, we will use software which guarantees that one user will always see the same design version. Furthermore, we will ensure that the results of all tests are relevant from a statistical point of view.

What is the final result?
The tests can drastically increase your online return. The following results can be achieved:

  • a 50% increase in conversion from landing pages
  • a 20% increase in return from AdWord campaigns
  • a 15% decrease in cancellations and break-offs in check-out funnels
  • a 20% increase in pages per visit
  • a 20% increase in average order value

Draft your own business case or prepare one with us and decide what the results above can mean for your return on your online channel. If these possibilities appeal to your organization, please contact us for an appointment, without obligation.

Interested in this service?
Name:
Email:
Comments:
Recent Clients
- Transavia
- ANWB
- Ilse Media
- Heineken
- SNS bank
More
Copyright 2008 | Disclaimer | Privacy Policy | Terms and Conditions