Current page: Data Management

Data Management

Extensive functionality to manage data

Weighting, meta data cleaning, computing variables, net creation, indexes and more – 100s of built-in features eliminate workarounds. Changes populate throughout the system eliminating manual work. Automated data refresh & meta data changes maximize efficiency in continuous study reporting.

Native Functionalities Release
the Power in Your MR Data

Dapresy Pro was built from the start by market researchers to specifically support market research data. There is a wide range of functionality available through an intuitive interface that allows you to easily create objects within your reporting. This greatly accelerates the speed of delivery.

 

Business Intelligence (BI) and many other platforms require scripting and data preparation that are time consuming, increase the number of resources involved in a project and are often the cause of errors. With Dapresy Pro, our user-friendly interface requires no coding or scripting knowledge.

 

A rich set of market research operations can be applied to your data. These are calculated at the respondent level for absolute accuracy, so no data preparation is needed. Some examples follow:

  • Managing multi-response questions
  • Creating grouped answers for calculating top/bottom boxes, net brands and detractors/promoters
  • Setting up calculations for NPS scores
  • Creating indexes based on multiple scale questions
  • Computing new variables for segments or other target groups
  • Excluding “Don’t Knows” from mean values
  • Refactoring of means from, for example, 1-5 to 0-100

Sort Data

Sorting functionality is the core driver of Dapresy Pro and has been implemented within all our market research reporting. It’s easy to sort by data values, sort a chart/table based on the sorting in another chart, or simply align multiple charts/tables in a dashboard. Reporting objects are “data aware” and can be modified to sort variables and values intelligently under your control.

Calculate Data

The list below shows the basic calculation types that can be applied when creating a chart or table. These can then be combined with any other function. For example, sorting by value, showing top lists, applying filters and comparisons between calculated values to view the difference between two time periods.

 

In addition to selecting the calculation type, you can decide if the calculation should be based on weighted or unweighted data. You can also have more than one weight variable. Just simply select the one to use when creating a chart or a table.

 

The base size in the calculation is always based on the unique number of respondents that answered the current question. This means that the system automatically handles the base sizes in, for example, multiple response questions.

  • Percentage shares
  • Mean values
  • Counts
  • Min, Max and Median
  • Significance testing (Z-test)
  • Correlation test (Pearson test)

How HTM Shoes brand easily visualize their tracker metrics

HTM Shoes is a UK-based retailer which conducts quarterly studies to measure their brand across all markets. The Brand Manager for each market tracks 4KPIs that are important to their organization. HTM uses this simple tracker to gauge how strong their brand is compared to their closest competitors, and where they need to put attention in the future..

Easily Handle Time Series Reporting

All data imported into Dapresy Pro requires a time dimension. Because of this, applying time-based filters and data comparisons is easy. The system inherently knows the latest data and supports a range of “floating” time intervals such as ‘last 3 months’, ‘last 2 years’, etc. This makes it simple to both report and update the system automatically.
 

Moving averages calculations are also built-in and easy to use. These can be applied to any interval such as weeks and months.

 

There is support for flexible time logic to allow definition of, for example, a custom finance year.

 

There is also support for variations in the time logic of different reporting objects. This allows you to create dashboards where all data is properly synchronized by time dimension when, for example, tracking data and media spend data have different collection dates.

Only Show Relevant Data

There is often a need to hide a result based on low base sizes. The system allows the ability to freely define the limits and select if the limits are to be based on weighted or unweighted data.
 

Hiding the result might be required due to anonymity requirements for personally identifiable information or to ensure the validity of statistical analysis.

 

Depending on the type of survey, you might not need to hide low base size results. For example, in a brand tracker you can use the “warn for low base size” function to indicate that the base size is low rather than hiding the result.

Create New Variables “On The Fly”

New variables can be created in Dapresy Pro in multiple ways. The advantage of Dapresy Pro (versus BI tools) is that these variables are calculated based on raw survey data. The definitions are then saved into the system and automatically applied to new data imports. The variables can be edited at any time and the changes will be applied globally to all the data in your project.

 

Below are examples of variables that can be created in Dapresy Pro:

Grouped answers
Grouped answers aggregate the results of different answers on a respondent level. For example, in the creation of top/bottom boxes, net brands and NPS detractors/promoters. Grouped answers will not affect base size, therefore, allowing the use of original answers without getting a wrong base size due to the new answer options.

 

Indexes
Indexes are usually used to report KPIs based on multiple scale questions. The scale questions can be weighted. So, a specific question can affect the KPI more than others or they can be equally weighted to have the same impact on the KPI or the impact can be based on the raw data itself. The impact is based on the respondents of the scale questions.

 

Computed variables
Computed variables are most often used to create segment variables that are not part of the survey data. They can also be used to create advanced KPIs, recoding string data to categorical, for merging variables.

The computed variables are created by setting up formulas with Boolean operators, for example, “AND”, “OR”, “NOT”, to filter which respondents to include in each answer option in the computed variable. Using advanced formulas and the ability to base a computed variable on another computed variable, the functionality is extremely powerful and flexible.

 

Recode data
The Recode data function can be used to recode the data within the existing variables instead of creating a new one. As an example, it can be used to merge brands, clean a question for unwanted responses such as “Don’t know” and much more.

 

Weight variables
The weighting of the data can be made in Dapresy Pro if it has not been done prior to the data import. New weight variables can easily be created by simply selecting which questions the weighting shall be based on for the desired distribution. It can be based on a single variable or a combination of variables. For example, to base the weight on both age and gender.

Provide Context with Benchmarks

Benchmarking results can be done in many ways depending on the project requirements. For example, in a customer satisfaction survey the hierarchical filtering is most often used for an automated benchmarking process.

 

By using a hierarchical filter structure, it is easy to add an automated comparison to results between different levels in the organization. For example, each manager on a team level can benchmark their result to the relevant business unit’s result and the overall company result and so on.

 

In a brand tracker, the built in “Benchmark” function can be used to compare the results between different brands, target groups or time periods. The Dapresy Pro “icon & shape” library lets you easily view, for example, the difference in percentage units or in percentage shares with visual appealing icons like a “thumb up” or a “thumb down”.

 

In combination with the benchmark types (above), the sorting and top list functionality can be applied to allow comparisons to, for example, the “best in class”, the mean value of the “three best brands”, etc.

Speak to an expert