Free trial limitation: No data exports


Software Evaluation Period

The 3-day free trial comes with one important limit: there are no data exports. There are no exports because the trial is software evaluation period available for testing the software and not about getting the data. We want you to experience our approach to text analytics and to make a decision about whether or not to be a customer.

As part of that experience, we do enable access to the Gnip PowerTrack for Twitter, and also public Twitter via the search API, certain Facebook pages, as well as other data sources. There are real costs to support our free trials, but we also think it makes testing the software more interesting. 

If you decide that data export is important, we ask that you subscribe to our lowest cost service (Professional) for $99/month (discounted to $49/month for academics and $24/month for students) for at least one month. Then you can export CSV files of archives and buckets for use in third party software.

Important Note for Twitter Researchers: The Twitter Terms of Service limit users to "save as" CSV no more than 50,000 Tweets per day. When you pull Twitter data in from the Search API, or license it via a Gnip PowerTrack or Sifter, the data is meant to be manipulated and analyzed in DiscoverText, in line with the spirit and intent of the Twitter Terms of Service. Once raw data (in "Archives") has been transformed through search, filtering, clustering, coding, or machine classification, the results (the "Dataset") can be exported at a rate of 50,000 per day.

Update for Academics: We did launch a new feature to help with academic replication datasets. You can watch the introduction video


Have more questions? Submit a request


  • Avatar
    Magaela C Bethune

    Very helpful! What are the benefits of exporting data via a CSV file vs. analysis within DiscoverText?

  • Avatar
    Stuart Shulman

    There are many benefits to using the advanced text analytics in DiscoverText. The most important are duplicate detection, human coding, coder reliability and validity measurement, and machine-learning. Some people use these or search and filtering to clean the data and then extract samples (100,000 items per day max for Twitter) where they can process the subsets using statistical, network, or other third party software.

Powered by Zendesk