> For the complete documentation index, see [llms.txt](https://catchdocs.gitbook.io/catch-customer-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://catchdocs.gitbook.io/catch-customer-documentation/flow/flow-service-overview.md).

# Flow - Service Overview

Flow is a real-time audience measurement service for the Out-of-Home (OOH) advertising industry, i.e. real-world advertising assets in public spaces such as Billboards, In-Mall signage, Bus Shelter ads, etc.

The OOH industry is under increasing commercial pressure from advertisers to report the effectiveness of (and trade) its legacy medium with the same metrics as the online advertising industry.

The most important of these metrics is **Ad Impressions**, specifically **Opportunity to see (OTS) Impressions.** Essentially, how many people pass by the ad. This metric is used for forecasting campaigns, reporting performance, and increaasingly to set the pricing of OOH advertising campaigns.

The OOH industry already employs several different approaches to achieve this, as outlined in the [measurement landscape](/catch-customer-documentation/flow/appendices-and-resources/measurement-landscape.md) appendix, each approach with advantages and disadvantages.&#x20;

Despite multiple solutions being available, real-time audience measurement on OOH is very rare, and most media owners trade on statistical data. Statistical data is typically compiled using large spreadsheets that merge multiple data sources and sample points with market assumptions operational rules and estimations. Generating and maintaining these spreadsheets is an extremely arduous process.

Flow was designed to be a balanced, scalable, and fully automated solution that has a strong offfering for the most important benefits and requirements, while accepting most palatable of disadvantages and omissions.

Flow is an OOH measurement solution that offers:

* Real-time, ad impressions, measured per ad play
* Highly accurate OTS impression measurement
* Low Cost installation and operation
* Ease of installation, activation, maintenance and operation
* OTS ad impressions as a native unit of measurement
* Instantly accessible live reporting data, never retrospective
* Native integrations with the most common automated / programmatic ad trading platforms
* Native integration with the most common 3rd party ad media players
* Long term reliability and stability of methodologies and techniques
* Full transparency in methodologies and technologies used, as well as the raw data used to arrive at OTS impression values.

These benefits were prioritised over the following disadvantages:

* No demographic data
* A requirement for on site hardware and software installation

## System Workflow Diagram

<figure><img src="https://3452267159-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FXpeOmkPYTxRQxOVln3Jz%2Fuploads%2FcKeQEBfLDtrcEpSMr7s9%2Fimage.png?alt=media&amp;token=44d37ce7-e981-4e93-beb7-20bd3555bfc8" alt=""><figcaption></figcaption></figure>

1. A Catch Media Gateway is installed on/inside/adjacent to the advertising face, connecting to the Player PC over USB ([Go to Device Installation](/catch-customer-documentation/flow/installation/device-installation.md))
2. Software (Catch Control) is added to the Media Player PC to operate the Catch Media Gateway, Flow service, and communicate with the existing media player software. ([Go to Software Installation](/catch-customer-documentation/flow/installation/device-installation.md))
3. Every time an ad plays, a trigger is sent to the device to start a new measurement session, generating a report when each ad ends ([Go to Playback Measurement Report](/catch-customer-documentation/flow/technical-workflow/playback-measurement-report.md))
4. The report is processed by Catch Control to categorise detections, exclude ambient devices, filter out poor quality detections, and update a rolling average of dwell time of devices. (Go to [Quality Categorisation](/catch-customer-documentation/flow/technical-workflow/quality-categorisation.md))
5. The high quality data is then passed through a model, to convert wireless detection data into a highly accurate count of the number of people (OTS Impressions) (Go to[ Impression Tranformation](/catch-customer-documentation/flow/technical-workflow/impression-transformation.md))
6. The categorised aggregates, human model, dwell time, and the OTS impressions are then sent to the Catch Cloud Infrastructure (Go to [Privacy & Data Transport](/catch-customer-documentation/flow/technical-workflow/privacy-and-data-transport.md))
7. Data is then available for analysis on a Visual Dashboard (Go to [Raw Data Charts](broken://pages/LdF7ZGBTt1ij1TTqiBqM))
8. The raw data is then irreversably deleted from both the Catch Media Gateway, and Catch Control.&#x20;
