Data Begin With Better Data, End With the Best Results

Understand the varying, complex datasets available and how data can affect their media mix and drive campaign results. The VideoAmp Platform is purpose-built with this in mind, powered by the most trusted, highest quality TV dataset in the industry.

Data-Driven Purpose-Built to Set a New Standard

VideoAmp’s linear TV dataset is one of the largest commingled, deduplicated and enriched Set-Top Box (STB) and Automatic Content Recognition (ACR) television exposure datasets in the industry, across 39M households and 63M devices.





Our digital dataset is built from first- and third-party data sources, log-level ad server exposure files, transactional attributes and other behaviors observed by our partners.



Our predictive methodology, developed within a privacy-protecting framework, converges these disparate datasets to create a unified ID, to discover and connect with more of your audience, at scale.

The VideoAmp ID

ALL ABOUT DATA Not All Data Is Created Equal

Without access to a complete picture of their media performance, those who use a single source of TV viewership data struggle to develop actionable insights and measure with confidence. Each dataset inherently contains its own benefits and drawbacks. While new TV viewership data sources are plentiful and continue to become available, these sources individually are not an accurate reflection of the demographics and viewership habits of the entire U.S. population.

The Value of Commingling

We take multiple raw, disparate sets of viewership, schedule and metadata, some of which may contain conflicting or overlapping information, and combine them to make one accurate, holistic view that answers the question: “Who is watching what and when, and where are they watching it?” While the process appears straightforward, each step involves complex data science algorithms, processes and programs in order to execute these tasks at the scale and accuracy required.

The commingling process takes place in three phases:

Properly receiving and storing ad exposure and TV viewership data is an essential first step of commingling to ensure both data and privacy integrity.

Once the data has been ingested, it’s essential that the data is properly scrubbed and cleansed of additional inaccuracies. This process helps to reduce the number of total viewership sessions, while increasing the overall accuracy of the information.

The real commingling process begins by joining separate, enriched datasets to form a holistic, unified dataset with enhanced accuracy and completeness.


The three distinct data types leveraged within our industry are detailed below, including current challenges and how VideoAmp fine-tunes and enhances these data types to ensure maximum results.

01 Raw TV Viewership

Industry Challenge

Platforms often license and repackage unprocessed, uncleansed TV viewership datasets since few have the data science and engineering capabilities to properly clean and correct these complex datasets. This results in unstable and inaccurate data.

VideoAmp advantage

We source anonymized, raw data across the linear and digital ecosystem, which is ingested, cleansed, processed, stored and published in a familiar format. This ensures the data is available and actionable in market.

02 Ad Exposure

Industry Challenge

The purpose of ad exposure data is to indicate whether a household saw a specific airing of a creative ad.

VideoAmp advantage

We leverage multiple sources of ad exposure data — from STB, Automatic Content Recognition (ACR), Over-The-Top (OTT), commercial airings and TV schedule data — to create a proprietary matching system that identifies ad exposure at the household and device level.

03 Skew-Corrected, Census-Modeled

Industry Challenge

All raw datasets are inherently biased, skewed by age, income, geography, ethnicity, etc. In order to have a balanced and accurate U.S. panel, these skews must be corrected.

VideoAmp advantage

We complete a rigorous process to cleanse raw datasets that includes using demographic weights to correct for demographic skews, modeling the audience to match the size and composition of the U.S. census. Our proprietary method of skew-correction and modeling to census ensures the accuracy of our TV viewership datasets across age, gender, income, education, ethnicity and geography.

The benefits

Advertisers need reliable measurement solutions that are not only trustworthy, but scalable, interoperable and cross-screen to form the foundations of their media investment initiatives. From planning to activation, negotiation and attribution, advertisers who leverage commingled data will exceed the competition in every facet of business.


Supports in-flight optimization for linear and digital.


Improves overall confidence in network recognition and mitigates the effects of phantom viewing sessions where cable boxes are left on, while TVs are off.

Higher Match Rates

Increases match rates of first- and third-party attributes between platforms due to increased footprint and span.

Future Proof

Includes ACR sources to ensure future measurement capabilities of national addressable audiences.

Our commingling methodology is designed to power TV to digital connections — enhancing the entire advertising experience by solving today’s most advanced use cases and paving the way for the future.

Put our data to the test. Let’s talk.