Datasets Overview

Before you can create segments in Simon you need data with which to segment. Likewise, before you can create any content-rich campaigns, you'll want data to populate that content. Fortunately, you can get started immediately bringing data into Simon; no engineering necessary.


Dataset dashboard

The dataset dashboard looks like this:


Dataset overview

a. Create Dataset: click to navigate to the new dataset screen
b. Tags: use tags to organize your datasets

c. Search: search your datasets by their name or tags you've created. Use the first drop down to filter by category or search by a specific field. Use the seconf drop-down to search all datasets or narrow it down by one of multiple options. Leave the search bar blank to see all existing datasets.
d. List: you can sort of the resulting list of datasets by creation date


Dataset creation process

There are multiple parts to creating a complete dataset. See:

  • Create a new dataset
  • Dataset validation
  • Field configuration

Dataset lifecycle

Saved datasets exist in one stage of the dataset lifecycle, indicated by the dataset dashboard status field.


Status column


Un-used Datasets

Datasets that have not been used, viewed and updated in the past 30 days are subject to automatic archival. Archiving these datasets might cause your data pipe to complete faster. If you have any questions, please contact your account manager.


Newly created datasets start in the develop status. While in develop, datasets are not executed against your database, nor do they save fields to Simon. This stage is useful for iterating on datasets before they go live.


After you click the commit button, datasets move from develop to live (see Create a New Dataset). In order to go live, a dataset must pass validation (see Dataset Validation) and have fields configured (see Field Configuration). A live dataset will be run and ingested by the Simon pipe. A live dataset and its fields are editable, but they must always pass validation before saving.

Paused (queries only)

In special circumstances it may be desirable to pause extraction of a query. For example, the underlying table schema is changing and you need time to refactor the query. It is possible to pause a query, and Simon will instead rely on the data from the previous extract going forward.

While paused the query data goes stale, and so we don't recommend you leave queries paused for extended periods of time.

To pause a query, contact your account manager.

Related Documents