Add More Flavor to Your Data with New Recipe Ingredients

Dataset recipes include a smorgasbord of new features to help you get even more out of your data. Bucket date fields to analyze data by whatever periods you choose. Extract components from dates to get a yearly view, or zoom in for a second-by-second look. Free numbers from their dimension fields with the dimension-to-measure transformation, and then do math on them. And when you’ve done all that, edit field names and labels so that your explorers really know what their data’s telling them.

Bucket Date Fields

Bucketing lets you create values based on existing values in another field. Previously, you could only do this for measure and dimension fields. But now it’s available for date fields too.

In a lens or dashboard, you can group the data by regular periods, such as quarter and month. But what if you wanted to view data by some other period, like season or trimester? That’s where date bucketing comes in.

Let’s say your company wants to analyze support cases by season. Now you can open up your cases dataset in a recipe and add a bucket field to show the season in which each case was opened.

Bucketed date field with absolute ranges

To bucket a date field, click the field column header, and then click the Add Bucket (Add Bucket button) button to open the bucket field dialog. The default mode is Absolute, which lets you specify a date range and name for each bucket. In this example, each bucket is a season.

Bucket date field with absolute ranges dialog

When you’re done, the new field appears alongside the date field you’re bucketing.

In Relative mode, each bucket is a period of time relative to the date that the recipe runs. You can create bucket periods based on days, weeks, months, quarters, or even years. For example, create a bucket field from the case created date to categorize cases by when they were opened. This takes a lot of work in the dataflow, but not in a recipe!

Relative date bucket field

When you create a relative date bucket field, enter the start and end of each period, or use the sliders.

Bucket date field with relative ranges dialog

Again, when you’re done, the new field appears alongside the date field you’re bucketing.

Find a Perfect Match with Composite Key Augments

When you add data in a recipe, you can now match rows using up to five pairs of lookup keys. For example, imagine you are adding contact information to lead data in a recipe. Using names alone to match can result in duplicate matches when leads have the same name. Now, you can match on name and company to ensure you’re matching the right data.

To use composite keys, click the Add Data button (Add data button) and select the lookup dataset. Wave still suggests the first pair of keys if it can, but now you also have the + Add Another option. Click this to select additional keys.

Add data dialog showing composite keys

New Transformations and Changes

The Truncate transformation, which lets you create a field by taking characters from another field, is now called Substring to better reflect what it does. We also added the Extract Date Component and Convert Dimension to Measure transformations. To use these transformations, click the column header of the field you want to transform, and then click the Transform Suggestions (Transform Suggestions button) button.

Extract Date Component creates a field using a date or time component from a date field in the recipe.

Extract Date Component transform

For example, you can extract the hour component from the case created date to analyze case creation by hour of the day.

Extracted date component field

The Convert Dimension to Measure transformation takes numerical data in a dimension field to create a measure field.

Convert dimension to measure transform

Convert numerical data so that it can be used in calculations in recipes, dataflows, lenses, and dashboards. For example, you extracted the hour from the case created date to analyze metrics such as the average or earliest times of day cases are opened. But the extracted hour is sitting in a dimension field! Use the Convert Dimension to Measure transformation to free it up for statistical analysis in lenses and dashboards.

Dim to measure column

Change Field API Names and Labels

You can now edit the API name and label of new fields you create in a recipe, and the labels of existing fields. When you create fields, for example, when using bucketing or transformations, Wave creates the field name for you from the source field and the action. But CreatedDate_grain_Hour_dimensionToMeasure“CloseDate_delimitedSplit_1 is not all that helpful. You can edit these generated names to make them more meaningful to the people exploring the dataset.

To change a field name or label, click the field’s column heading, and then click the Navigate Fields (Navigate Fields button) button. In the field navigator, click the Attributes tab to change the API name and field label.

Field attributes
Note

Note

You can edit the API name only for new fields added in the recipe. API names for existing fields aren’t editable.