# An analytics firm moved about 17,000 listening tool queries onto raw provider data

Queries of up to 18,000 characters, running in full and delivered to its own cloud storage.

About 17,000 jobs · Queries up to 18,000 characters · Fix shipped within 1 week

## At a glance

- **Customer:** Analytics firm
- **Sources:** Raw provider data, replacing a listening tool
- **Volume:** About 17,000 jobs
- **Delivery:** Its own cloud storage
- **In production:** On its own stack

## About the customer

An analytics firm that delivers insight work to its own clients.

**Industry:** Analytics and Insights

## The Story

Challenge, solution, results.

### The Challenge: Paying tool prices for data

The firm was using a social listening tool as its data provider, and paying listening tool contract prices to do it. It also wanted to control its own tools and add to them, which it couldn't do inside a listening tool. So it moved to raw data from a provider.

The hard part was the queries. There were about 17,000 of them, written as long Boolean expressions in the listening tool's own syntax. Some ran past 18,000 characters, far longer than the new provider would accept in a standard request.

### The Solution: Queries that run in full

Datastreamer worked through the syntax differences between the two. When the longest queries started failing, the team traced the cause to a request-size limit on the provider's side, at about 8 KB.

Within 1 week, Datastreamer changed how long queries are sent so they run in full, and rolled the fix out to every customer using that provider. Results land in the firm's own cloud storage, ready for its analysts.

### The Results: Control over its own stack

About 17,000 jobs now run on the firm's own stack, at provider data prices, with the freedom to build what its clients need.

## How the pipeline works

1. **Queries:** About 17,000, carried over from the listening tool.
2. **Translation:** Syntax differences worked through for the new provider.
3. **Collection:** Long queries sent in full, past the provider's standard size limit.
4. **Delivery:** Into the firm's own cloud storage.

## Using a listening tool as a data provider?

See what your queries look like running on raw provider data.

[Talk to Sales](https://datastreamer.io/book-demo/) · [Read the Documentation](https://docs.datastreamer.io)

## Related

- [Trend Prediction Case Study](https://datastreamer.io/trend-prediction-case-study/)
- [Media Monitoring Case Study](https://datastreamer.io/media-monitoring-case-study/)
- [Social listening API limits: what you can export in 2026](https://datastreamer.io/social-listening-api-limits/)