Build Data Marts

How to create and modify data marts

What is a Data Mart?

Before you can build Reports in Indicee, you must upload your data.  Indicee Data Marts are the location within the application that understand your data and the specific business function it addresses. Once you upload your data to Indicee it is contained in the appropriate Data Mart, and you can build Reports against that Data Mart to answer your business questions.

Why do I need to create a Data Mart?

Data is held in buckets called Data Marts. If no buckets exist, there is no data, and thus Reports cannot be built.

How do I create a Data Mart?

Indicee has a Data Mart Editor that steps you through the process of creating a Data Mart.  You can launch the Data Mart Editor from My Stuff > Data Marts, and then click the Create Data Mart button. To get started you will need a Report Data Set to upload to Indicee. The Data Mart Editor will then guide you though the process of teaching Indicee how to manage your data and how you want to analyze it. Once you have created your Data Mart, you can upload newer versions of your Report Data Set, and Indicee will automatically add the new data to your Data Mart.

To create a new Data Mart, select the Data Marts tab, and click Create Data Mart.

In order to create a new Data Mart, select the Start button from within the blue “Create Your Own Data Mart” box. Choose Browse and select the primary file of this new Data Mart. The title and description of the new Data Mart will be defined later in the Data Mart Editor. Additional files may be uploaded later.

Once Indicee has successfully uploaded the new Report Data Set, it will display the Preview Data tab. It is important to verify the formatting of the data. Follow these four steps:

  • Select the starting row and column for the uploaded file. Crystal Report files will already be properly formatted; however, if working with a CSV, TSV or XLS file make sure that the headers are properly defined.
  • Confirm the “Data starts at row” value. This is the first row in which the Data Mart should collect data from.
  • Data Sets often have additional information that you may not want in your Data Mart. By deselecting the header boxes, certain data columns can be removed from the Report Editor. Note: This does not physically remove the data from the Data Mart, it simply “hides” it from view in case you need to add it back later.
  • If any additional information was contained in the file above the significant data, it can be applied as a new column. For example, instead of each row having a date, the entire page could have a common date (which was placed above the data). This single date can be applied to all rows within that data set.

In order to filter out unwanted data rows, Indicee uses values found in the data set. In order to remove a row, we must define which column we will define the value for, and then define the value. For example, If I want to remove rows with an “Item Number” of “555” I would select “Item Number” from the drop down box, type the value “555”, and click Add. To remove this exclusion, click the “x” beside the column name.

In order to define where your data ends, follow the same procedure as to filter out unwanted data rows. The first value which agrees with the defined value will be used as the end point.

With these items confirmed, advance to preview the Data Mart.

The “Preview Data Mart” tab is used to build the structure of data. For example, “item” is nested within “category”, which is nested within “department”. There are six key functions used to manipulate columns on this page:

  • Show Items - By dragging items onto the vertical boxes, hidden items can be added to the list of categories.
  • Hide Items - By dragging items onto these rectangular boxes, items can be removed from the Report Editor.
  • Nested Items - Items can be nested under broader categories by dragging them to the bottom of their “parent” item. Once the blue line appears, release the item to nest it. Note: Items can only be nested if they have one “parent”. For example, at Indicee Bikes “categories” are nested within “departments” because no category can be part of more than one department. However, departments must not be nested within “Store Name” as all the branches of Indicee Bikes have the same categories (ie. Departments have multiple parents)
  • Item Descriptions - By dragging an item, and placing it on top of another, it applies itself as a description. Items can have multiple descriptions, but a description can only apply to a single item.
  • Calculated Item - Similar to revenue or units sold, calculated items may apply to any level of summation (item, category, store) over a period of time, and with limits. Calculated items are unlike the common calculated columns found in spreadsheet applications as they to not generate a list of values, but are a calculation based on a summation of data.
  • Combine and Split Items - Some values need to be combined or split before they can be added to the Data Mart. For example, if a report contains a customer’s name as separate first and last names, Indicee can join these two into a single name. Alternatively, if a Report Data Set exports a region and country as a single value, Indicee can split them into two columns allowing for the nesting of these values.

Once the Data Mart had been organized, details can be defined in the Data Mart Details tab. The Data Mart and the Report Data Set must be given names. Because a single Data Mart can simultaneously contain multiple reports, a more generalized name is suggested such as “Indicee Bikes Data Mart”. The Report Data Set must also be labeled. The first Report Data Set will be used as a template to import similar reports. Our Report Data Set was called “Indicee Bike Sales Jan 2009”, and we have named it “Data Set like Indicee Bike Sales Jan 2009”.  Click Finish to proceed.

What kind of data works best with Indicee?

Indicee works best with cumulative business data that adds up over time. For example, sales data that adds up daily works well, because at the end of the month you will have an accurate total for the entire month. For instance, 10 sales per day will equal 300 sales at the end of 30 days.

What data file types can I upload into Indicee?

Indicee can upload Crystal Reports, Excel (XLS, XLSX) and comma or tab separated variable (CSV, TSV) files. For Excel, Indicee currently accepts Excel 97 - 2003 workbooks and the new XLSX format. If you have an Excel file saved in an earlier version, open it in Excel and use the Save As command to save it as an Excel 97 - 2003 Workbook or XLS.

How do my CSV, TSV and Excel data files need to be formatted?

For Excel, CSV and TSV files, it’s important that the data be in a columnar format with the following characteristics:

Data of similar types arranged in columns

Titles at the top of each column

Each row represents one complete record

Indicee does not currently understand files that have:

Groupings or sub-totals

Values for multiple time periods across multiple columns (e.g. financial statements are often in this format)

Does my data need to have dates?

No. However, because most users want to see how their data changed over time, it is recommended. There are some situations however where dates are not needed. For example, if the National Hockey League wanted to know which players had more than 500 career goals, they would not need to know in which years the goals were scored, but rather the total number of goals.

What is a hierarchy (drill down path, parent child relationship)?

Items can often be nested under broader categories. For example, at Indicee Bikes “categories” are nested within ‚“departments” and “items” are nested within “categories”. Items can only be nested if they have one “parent” because no category can be part of more than one department. However, departments must not be nested within “Store Name” as all the branches of Indicee Bikes have the same categories (ie. Departments have multiple parents).




What is a description?

By dragging an item, and placing it on top of another, it applies itself as a description. Items can have multiple descriptions, but descriptions can only apply to a single item.

How do I change the name of an item?

When creating or editing a Data Mart, display formats of an item can be modified. From the Data Mart Editor navigate to the Preview Data Mart tab. Double-click the item you would like to rename, type the new name in the text box, and press enter.

How do I change the display format of an item?

When creating or editing a Data Mart, display formats of an item can be modified. From the Data Mart Editor navigate to the Preview Data Mart tab. Hover over the item that you would like to modify, and select the icon. From here the Display Format can be changed and Reset to the default.

How do I get rid of data fields that I don’t want in my Data Mart?

When creating or editing a Data Mart, data fields can be removed. From the Data Mart Editor navigate to the Preview Data tab. Any check-box can be unselected in order to remove the column from the Data Mart.

How do I show items in the Preview Data Mart tab?

From the Preview Data Mart tab, items can be shown by dragging them onto the vertical rectangular outline beside the active items. This adds the item to the Report Editor.

How do I hide items in the Preview Data Mart tab?

From the Preview Data Mart tab, items can be hidden by dragging them onto the rectangular outline below the active items. This removes the item from the Report Interface.

How do I make calculated Items?

Similar to revenue or units sold, calculated items may apply to any level of summation (item, category, store) over a period of time, and with limits. Calculated items are unlike the common calculated columns found in spreadsheet applications as they to not generate a list of values, but are a calculation based on a summation of data. From the Preview Data Mart tab, select Create Calculated Item. Metrics, and Operators can be used to create a new Calculated Item.

How do I make combine or split Items?

Some values need to be combine or split before they can be added to the Data Mart. For example, if a report exports a customer’s name as separate first and last names, Indicee can join these two into a single name. Alternatively, if a report exports a region and country as a single value, Indicee can split them into two columns allowing for the nesting of these values.

  • 30 Day Free Trial
    on all accounts. If you upload less than 10MB of data you can use Indicee for free.
  • Try Indicee For Free

© 2010 Indicee Inc. All Rights Reserved

RSS Twitter