How Qubit can help provide insight on your current site's UI/UX/design and help you make improvements through A/B testing
Updated July 26, 2016
How Qubit can help provide insight on your current site's UI/UX/design and help you make improvements through A/B testing
Score 9 out of 10
Vetted Review
Verified User
Overall Satisfaction with Qubit
We use Qubit to implement strategic A/B testing (also known as CRO: Conversion Rate Optimization) for various mid-to-large tier clients on various e-commerce platforms. We seek to improve overall site user experience, conversions (e.g., revenue per visitor, average order value, and other key performance indicators) and conversion rate, enhance/extend functionality, improve design and aesthetics, and in general make the site more efficient and easier to use for customers.
- Qubit is a great tool for building experiments for A/B testing, replete with advanced JavaScript capabilities, numerous built-in modules, a robust API, and a "basic mode" editor for non-developers to use and build experiences.
- Qubit also supports segmentation (e.g., segment users by device type, browser, location, etc.) and advanced URL targeting to really hone in a select user groups and personas.
- For analyzing your tests' performance/data, Qubit offers advanced reporting within the dashboard and a great support team who will also pull raw data upon request.
- Because Qubit is a newer tool, some small bugs are to be expected and have been found, most notably within using the actual UI of the tool.
- The Qubit team is always enhancing their product which means it is undergoing constant development. This is a pro for sure in the sense that Qubit will continue to offer new features to users while also squashing prior release bugs, but it can also inadvertently introduce new bugs, which we've experienced mid-test development and/or worse--post test launch.
- Qubit does not currently offer the ability to filter out reporting data in the dashboard by device type, browser, or other segmentation criteria. This means we have to make special requests to the Qubit team for this data, where other tools offer this easily directly in the dashboard.
Qubit is a great choice for A/B testing where advanced JavaScript capabilities are required. This is especially true for large "legacy" sites that have a ton of on-site JavaScript that has culminated over years, often overriding changes made in a WYSIWYG or "basic mode" editor found in other platforms. In these cases, a developer must use JavaScript to override and/or extend methods on-site in order to ensure that changes made for a variation will remain so and not be overridden/destroyed by the on-site JS. QA'ing Qubit tests is also nice given their slick preview tool and the ability to leverage some Qubit JavaScript objects with experiment metadata.