If creating an online study with multiple conditions, it is common to see people creating a different survey for each condition. This can be quite inefficient due to jumping between conditions, and keeping common aspects of the questionnaire consistent across conditions. Qualtrics has many powerful features for controlling multiple conditions in a survey study and in most cases it can all be easily managed within one survey document.
If it is a quasi-experimental design, for example, then the survey can be adapted based an earlier user response (like age, gender or a preference question). In an experimental design, Qualtrics can create various condition variables, and randomly allocate respondents to various conditions, either using simple random sampling or stratified random sampling. Once the variables are created, various flow options can be used to control the order of questions, when to display a manipulation, or to tweak the wording of instructions. Sampling ratios can be controlled.
Although it is possible to simply randomise the flow of a questionnaire to various Blocks, in most cases it makes sense to use the randomiser to create the conditions using embedded data. An embedded data variable can hold the condition number and be created using the randomiser in a survey flow (or a variety of other means). The embedded data can then be used in various conditional flows to control the presentation of information of questions. The embedded data also appears in the data file so can be interpreted as the condition number in analyses. Nested conditionals and randomisers allow many more complex designs to be expressed.