Live TV Events

Better predict surges in demand caused by televised events such as sports. Improve forecast accuracy by incorporating this unique data set into your models.

Customers are leveraging Live TV Events and have reduced forecasting error rates by double digits after incorporating historical and predicted TV viewership into their models.

  • “Dominos is a pioneer in their industry and they’ve been at the forefront of cutting edge technology and sophisticated forecasting. For many years, they’ve viewed themselves as a technology-first, ecommerce company that sells pizza. PredictHQ has been a valuable input into their forecasting models for years. TV viewership is an additional data set that provides incremental benefit on top of existing powerful models. PredictHQ’s predicted viewership data has had a positive impact on their models, enabling continued data-driven decisions.”

  • Gain access to reliable future TV viewership

    Gain access to reliable future TV viewership

    PredictHQ is the only source of predicted TV viewership. Companies use our three years of historical data to pinpoint the impact of televised sports and then improve forecast accuracy with our predictions for future games.
  • Build strategies with forecast grade data

    Build strategies with forecast grade data

    There are vendors of historical-only viewership data that designed for media buyers. Our TV viewership data is designed for data scientists to improve forecasting at the county and store level.
  • Tap into the stay-at-home<br> economy

    Tap into the stay-at-home

    The pandemic has created a new focus for business: the stay-at-home economy as people purchase more items from home. Unique data pinpointing the impact sports on these purchases enables companies to optimize for it.
Forecast ready data

Get up and running quickly with our TV viewership data

We’ve created Jupyter notebooks around how to integrate our TV Viewership data into your models to help your Data Science teams get up and running quickly. Use our Data Engineering notebook to learn how to get the data and how to use it. Our Data Exploration notebook explores the data and how you can get value from it and our Feature Engineering notebook provides examples of how to build data models and features.
Jupyter Notebook Screenshots

Build business strategies with forecast grade TV viewership data

Filter by city or county to access relevant data rather than by broad designated market areas (DMAs) used by other providers. Access to county-level viewership data enables you to confidently forecast at a more granular level, such as for individual stores if you’re a quick-service restaurant or grocery store.
Expansive Coverage

Track the televised sports games that matter most to your business

PredictHQ’s Live TV Events data includes the seven top US leagues: NFL, NBA, NHL, MLB, D1 NCAA Basketball, D1 NCAA Football, and MLS. The data also includes the top 100 sports games based on viewership that includes golf tournaments and boxing matches. This expansive coverage gives insights into any potential drivers of demand. Pizza deliveries may skyrocket during football games, while demand for peanuts and beer may increase during baseball games.
    import requests
    response = requests.get(
          "Accept": "application/json",
          "Authorization": "Bearer $ACCESS_TOKEN"
          "event.label": "nfl,mlb",
          "location.place_id": "5368381",
          "start.gte": "2019-09-01",
          "start.lte": "2019-09-14",
          "sort": "start",
          "limit": "20"

Easily Aggregate

Aggregate TV viewership data to improve demand prediction accuracy

Aggregating Live TV events data is often the most impactful way to understand peak demand for your specific business. Depending on your business, you may need to aggregate TV viewership by day, by time of day, by county or by sports type.

All Live TV Events features

Live TV Events was built from scratch and is based on a variety of inputs, including sport team entities, county populations, broadcast information, league interest, external ratings and more. We featurize all relevant information into a probabilistic inference framework to predict viewership.
  • Seasonality
    Preseason vs playoffs drive different viewership numbers. Our proprietary Predicted Viewership can surge ~15% from an early to an end of regular season game for a team that is performing well.
  • Quality of match up
    Quality of match up
    We track the strength of match up based on the two teams competing using team rank, standing points, whether its home vs away and more.
  • Predicted viewership
    Predicted viewership
    We predict up to 6 months forward that you can incorporate into your models. Plus, our machine learning models are also constantly being updated up until the game as shifts happen in rankings, broadcast schedules and more.
  • Home vs. Away games
    Home vs. Away games
    We track the strength of matchup based on the two teams competing using team rank, standing points, and more.
  • Uncertainty of results
    Uncertainty of results
    We track differences in performances to measure the quality of the game, which influences viewership heavily.
  • National vs. local coverage
    National vs. local coverage
    One sports game could have high viewership and impact in many different cities or even nationally.

You can’t prepare for what you don’t see coming

Harness the power of demand intelligence

Knowing the impact of demand causal factors like events will transform your business. The American Society of Hematology has a $45M estimated economic impact — and that's only one event in one city.

  • 0
    data points enriching
  • 0
    events across
  • 0
    cities, accessed via
    1 API

Get Started

Contact our data science experts to find out the best solutions for your business. We'll get back to you within 1 business day.

Talk to an Expert