What is Qubit

Qubit defines itself as a personalization platform, but its core it is a set of three main components, an experimentation tool, a datastore/visitor profile and a set additional plugins to add additional functionalities.

Experimentation tool

Qubit offers a way to setup A/B and multivariate experiments through code injections. Qubit allows to setup different variation of the experience by injecting javascript and css onto different pages. Qubit offers ways to analyse and track the performance of experiments.

DataStore / Visitor Profile

Qubit also offers a datastore that allows the accumulation of visitor profiles as well as other information within the platform. For aggregation or retrieval, through the use of Qubit datastore, it is possible to offer more personalized experiences. The Qubit platforms also offers the possibility of resolving identities through id matching.

Set of Plugins

Qubit offers different types of plugins to add different sort of functionality such Social Proof, a recommendation engine, a plugin to generate abandoned cart email, …

Qubit as an AB Testing Tool

Setup the data push:

In order for Qubit segmentation, recommendation and reporting capabilities to function properly, it is necessary to provide a certain set of events to qubit.

Qubit provide an extensive documentation with regards to what need to be sent and for which purpose

Setting up a segment:

Qubit allows to setup specific segments using data collected from the customer profile. Using specific segment it is possible to trigger experiments when certain conditions have been met.

Experiment definition

Within an experiment, there are a couple of default settings to define: Goals, Segments, Statistical Threshold and Traffic Allocation. These define how the experiment is setup and data is reported.

Segments

Segments can be used within an experiment for both inclusion and exclusion purpose, meaning that you can setup a targetted experiment based on the data within the vistor profiles.

Goals

The goals setup will at a later stage be included in the experiment dashboard as well as the experimentation metrics dashboard that opens when you click on any of the previously defined metrics

After having setup the goals and launched the experiment, data on how the performance of the experiment is tracking will become available by clicking on each of these metrics, more information is provided in qubit’s documentation.

Stats Threshold

Qubit allows to customize the statistical threshold and effect size to consider an experience to be a winning experiment

Stats Threshold

Qubit allows to customize the statistical threshold and effect size to consider an experience to be a winning experiment

Traffic Allocation

Qubit allows to define the traffic allocation across the different experiment in a variety of ways. The tool offers a simple setup on managing the lifecycle of an experiment through an increase in the experiment allocation.

Segment

Variation JS and css

Variantion JS and css is Where your Experiment code should be placed, HTML code can easily be injected onto the site by using JQuery append and prepend functions for instance.

Trigger Js

The trigger.js script of your experiment contain the logic of when the experiment should fire on the website after they have passed the segmentation conditions. The function cb() is responsible for the experiment firing.

Qubit Packages

The segment store allows us to retain the history of the customers belonging to the different audiences. Segment stores allow for better manage the load on the downstream export flow. Either by providing a means to check the delta of audience memberships/segment attribution or through branching out the evaluation logic for customers belonging to the audience.

Qubit Packages

Qubit offers a structured way to import certain dependencies through Qubit Packages. These are npm style packages

From a data perspective, the core packages maintained by qubit are:

• datasets
• http-api-stash
• http-api-tally

They offer way to interact with the datastore within qubit, to store, retrieve and count data within it.

The datasets api allows you to interact with the different dataset that you might have imported to qubit.

While stash an tally allow the interaction for visitor events.

‍