Qubit, now from Coveo (acquired October 2021) uses visitor history data to understand different user segments and serve personalized messages to segments using JavaScript. It is available as either a managed or self-service model. Data is collected using Qubit's own Universal Variable data model, or by integrating the user's existing model via our API. It combines quantitative data with qualitative visitor feedback to give Qubit users the ability to detect areas for optimization.
Using…
N/A
Google Content Experiments (discontinued)
Score 7.3 out of 10
N/A
Google Content Experiments was a tool that can be used to create A/B test from within Google Analytics. It has been discontinued since 2019, and Google now recommends using its Google Optimize service for A/B testing.
Coveo Qubit is a very helpful platform mainly for organizations that need to provide a solid business model before carrying out any implementation or new functionality. In addition, it is a very good tool to generate changes and show different content to different types of clients, with their personalization and segmentation criteria.It is ideal for simultaneous testing and customization, only one of these activities individually is not recommended.
Do you already have Google Analytics? If so content experiments is a good, free, starting point to dip your toes in A/B testing. Do you need to run Multivariate experiments? If so, Google Content Experiments is not going to fit your needs.
Qubit platform uses solid testing algorithms and delivers reliable reporting and analytics of testing campaigns. The dashboard section is easy to use and provides a good high level overview of the core campaign metric performances.
On-boarding and implementation of Qubit technology was painless and well handed throughout the entire process even with more complicated platforms.
Turn-around time for development and testing of campaigns is extremely fast. This enables the business to have a much higher throughput of tests and allows for quick validation of ideas rather than having to wait for months before a test is ready to go live on production site.
While they do work very closely with us, especially our success manager, more involvement on a technical level or a dedicated engineer, would be a great advantage. We develop a lot of experiences in house and, like all companies that dev in house, we have our own way of working. A dedicated engineer that got to know us better as well and understood how we work could only help. Sometimes, a few errors can get through QA due to this.
Qubit is currently providing resources and support we do not have internally. Our relationship managers are exceptional and I feel well informed and well supported by their team. The tool is nice, but our contacts from the company are the real reason to maintain our relationship. They work hard for clarity and continue to help us push for additional opportunities.
Content Experiments just makes it is simple and easy to implement A|B tests. We will be evaluating other tools in search of a more robust system for multivariate and cross-page testing, such as Optimizely or Visual Website Optimizer. However, for basic testing, you can't really beat it.
Very simple user interface, built on top of advanced functionality, makes the platform easy to use. The team at Qubit are also very open to feedback and introduce new and useful features fairly often. On the reporting side, the inbuilt dashboard reports are good for a top-level view of test results, and Qubit have made a lot of their output data available should you wish to run your own analysis
I would say that the Qubit account managers are always available for any request. We have a lot of different promotions that could always do with last minute optimizing or changes and Qubit can be relied on to get this changes up and running in an impressive amount of time, so that we don't need to patch live or wait for the next IT sprint. Invaluable to our business.
Technology is good for A/B testing and personalisation - allowing any team with a dedicated developer to create test relatively easily and to report/analyse them in a fair amount of details. Some advanced features, especially on the set up of test cells, are dearly missing. Unfortunately, new features are often not free of bugs... Also, support is sub-par, which means new features are realised without proper documentation, example or training (but of our Qubit counterparts and internally).
Qubit are supportive and flexible in providing support. They are happy working out of usual hours, even on weekends and if I have any doubts about the set up of an experience they’re quick to respond and willing to check my work. On particularly big revenue days they monitor our account and they’re quick to identify and problem solve any issues.
Using the free tool, overall "live support" is limited. However, there are plenty of online resources to get started. If you need handheld support, it is best to upgrade the service or hire a developer through one of Google's partner agencies. There could be more support for understanding what makes a test useful or not.
The training was great, but would be great to have a script or PDF with some explanations of the qubit JS layer. Without the script you can just try to learn on your own, so the training is not as powerfull as it could be. On the other hand - would be great to have training related to reading statistics or personalisation.
Implementation couldn't be easier. All we needed to do was insert the tag. (easy) and set up the data layer. (dev required) This was pretty smooth in comparison to some of the other tools we use on our site, and was done in less than a day. Note : Data needs to be collected for a set period of time before you can accurately rely on the data that you are receiving. This is normal though with everyone else that we have used
At the time of taking the product, we found no comparable alternatives. Since then, the product has only grown from strength to strength, so it still does not have any comparable competitors that offer both the technical product and business knowledge that Qubit can. Google appear to offering a competing product which may be one to watch in future, and something for Qubit to keep an eye on
Google Website Optimizer was a better product but has been discontinued. We have also used Test and Target , which has more features but we have been doing fine with Google Content Experiments. Most testing situations can be handled with Google Content Experiments.
Our requirements change throughout the year like most E Commerce retailers. At Christmas and Peak we're dealing with around ten times the usual traffic on the site. Qubit had no problems with this at all, tests continued to fire, and stats were still reported accurately. I don't think that it is the most server intensive .js anyway, but we have seen no issues at all.
