Skip to main content
This guide shows how to build a tweet dataset with Xquik for a study. It covers cost, collection and the run record your methods section needs.

​
What a tweet dataset costs

X’s pricing docs, read on 2026-11-04, list USD 0.005 per post read. Xquik bills 1 credit per delivered tweet. A top-up credit costs USD 0.00015. Empty pages, duplicates and filtered rows cost nothing on Xquik. Check the price of a job before you run it with the cost estimator.

​
Pull a small sample

Each search call returns 1 page of tweets. Put the date window in the query with since: and until:. Days are UTC.
Continue with next_cursor until it is empty. Store tweet IDs so a rerun skips rows you already have.

​
Collect a large dataset

1

Estimate the job

Send the query to the cost estimator first. It returns estimatedResults and creditsRequired.
2

Start an extraction job

Create an extraction with toolType set to tweet_search_extractor. Pass your query as searchQuery and a cap as resultsLimit.
3

Wait for the job

Poll Get Extraction until the job finishes. Its totalResults field gives the row count.
4

Export the rows

Download the rows as CSV, JSON or XLSX with Export Extraction. Use sinceDate and untilDate to keep one study window.

​
Cite the run in your methods section

Each extraction job keeps its own record. Copy these fields from Get Extraction into your notes. Add the sinceDate and untilDate you exported with. A methods sentence can follow this template.
Keep the job ID and the tweet IDs with your data. Another team can then repeat the same query and window.

​
Plan the study

  • Fix the query and window before you collect. A changed query is a new run with a new job ID.
  • Keep raw exports unchanged. Clean a copy, and record each cleaning step.
  • Follow your ethics board’s rules for storing and sharing user data.

Search tweets by date

Set exact date windows with since: and until:.

Create an extraction

Start a bulk job for tweets, replies, followers or lists.

Export an extraction

Download rows as CSV, JSON or XLSX.

Python SDK

Install the SDK and run your first search.