Datastream data capture
Companies running large numbers of surveys can find it difficult to combine or blend survey data from one survey to the next, or to combine survey data with other data sources for deeper analysis. Cxoice uses a 'datastream' approach which means that data can be easily pooled across surveys, making it easier to aggregate data for a bigger picture.
Traditionally, market research surveys collect data in 'column and code' format where, say 1 is Male and 2 is Female. The coding itself is specified by the questionnaire definition, which means each participant data record is unique to the survey in question. To do something across surveys, such as to pool all open ended answers by individuals under 24, requires matching and extracting data survey-by-survey, then aligning and collating that data into a final file.
Cxoice collects data using a 'datastream' approach. that means that individual variables are collected and stored one-by-one with both the coding, the question and the text of the coding are held together. For individual surveys, data is collated and extracted by survey reference. However, because the data is in a stream - not a record per participant - the data can also be collated and extracted across surveys by response, by question type, by question text or by question name.
Where this is useful is that many companies standardise certain variables with every survey - for instance demographics, the use of a net-promoter type question, and final wrap-up or survey experience measures.
With Cxoice, these variables can be pulled out across the surveys conducted allowing, for instance, a fast and easy way to keep an eye on survey experience measures by subgroups across surveys.
The use of a datastream also allows external data and variables to be blended into the survey data (subject to GDPR and consent requirements). For instance, in a survey taken from a database of customers, the database information is easily blended with the survey responses to add in factors like purchasing events, service queries and so on from real behavioural data.
The datastream also ensures data integrity and audit trails, as data is only added to the data stream and not overwritten. For quality control, it means participants trying different answers in order to get through screening questions can be tracked. It also allows post-research monitoring and adjustment to the dataset without overwriting the original data, making Cxoice suitable as a platform where high data reliability is required, such as in clinical research studies.
The datastream approach, also allows for repeat and time-based data collection - not necessarily directly from a survey but, for instance, collecting event-based data from monitoring applications or computer systems - such as weather forecasting, or location-based tracking.
With companies collecting increasing volumes of data from multiple sources, Cxoice's datastream design allows that information to be centralised and managed in a single pot, as well as providing all the expected data and analysis from individual survey type designs.
