Data by Definition
So the API connecting your DMP and DSP is down and you need to ETL a dataset to a CSV ASAP.
The keywords and abbreviations that are often used in data-based marketing can read like a foreign language, and the constant state of evolution brings new terms to the table at an ever-increasing rate.
To help, we’ve collected some of the most-used terms and their definitions to bring clarity to the confusion as you formulate your print and digital advertising campaigns.
Let’s dive into Data by Definition.
1st Party Data
Data points that are directly collected from the source, stored and owned by an organization or company.
2nd Party Data
Another organization or company’s 1st party data that is purchased directly from the owner.
3rd Party Data
Data that is collected or aggregated from various sources by an outside entity that is not the original collector of the data and typically sold in the marketplace to enhance targeting efforts.
Data points that are appended (or “attached”) to an individual’s record, such as consumer data, cell phone number or email address that might help improve match rates.
Due to variance in data quality, it is imperative that data undergoes a series of enhancements and appends so that it is at a high enough quality for marketing. This includes enhancements such as NCOA/CASS (National Change of Address/Coding Accuracy Support System) processing and geocoding.
Matching across data sets allows us to connect an individual to their online presence and devices and then back to either a voter file and/or a company’s 1st party data set.
Not all data is created equal. The quality of data can vary greatly by source and is dependent on how frequently the data is updated, the consistency in the type of data that is being collected and/or duplicative records.
A DMP or Data Management Platform is a central location where various incoming data sources are aggregated and managed, enabling audience generation that leverages multiple data sets, such as consumer data, predictive modeling and 1st party data.
Match rates refer to the total number of records that match between two data sets. Match rates will vary across platforms and can depend greatly on data quality.
Omni-channel refers to an integrated approach to marketing that enables an organization to message consumers seamlessly across online and offline platforms such as mobile, digital, television and print.
PII or Personally Identifiable Information at its core refers to data points that allow you to identify a specific individual.
Predictive modeling in data-based marketing utilizes statistical analysis to predict an outcome. Typically this translates to predicting a consumer’s and/or a voter’s likelihood to either take an action or agree/disagree with a certain issue. This is done by aggregating data set, including survey responses, consumer data and/or a voter file, and then building a statistical model. That model is then used to apply a predictive score from 0 to1 (or 0% to 100%) to each individual.
A voter file is a list of every registered voter in the state as recorded by that state or jurisdiction’s election authority (i.e. Secretary of State, State Board of Election and/or County Clerk). The data included on these files varies greatly by state, but voter files most frequently include name components, registration addresses, political geography, previous vote history, demographics and partisanship.
Ready for more?
#DigitalDelivers: Digital by Definition
#MailMatters: Mail by Definition