Book a demo

Book a demo

Marketing Data Glossary

Struggling to understand the difference between CPL and CPA? Don't know your data warehouse from your data lake? We all know how jargon-filled and buzzword-heavy marketing analytics can be. But, fear not! We've put together our comprehensive marketing analytics to help you understand all these terms.

    A

    A

    Ad Exchange

    Ad Exchanges, such as Google Display Network, are digital markets where ad space is traded. Ad exchanges have arisen to replace in-person negotiations with streamlined and often programmatic transactions, enabling media buyers and sellers to instantly make deals for select inventory using unique data and choosing their preferred type of auction, such as open or closed.
    See also:

    Buying Model, Programmatic Advertising

    Ad Server

    Ad Servers are the logistical hubs for online advertising, responsible for managing and displaying ads. The easiest way to think of them is as processing factories where ad creative can be uploaded, sorted, and converted into the right format so it can be quickly and efficiently served when publishers issue calls for certain types of ads after trading inventory in ad exchanges and auctions. By controlling ad delivery, ad servers can see how ads are interacted with once they land using tags that measure user views and clicks.
    See also:

    Ad Exchange, Programmatic Advertising

    Affiliate Marketing

    Affiliate Marketing is a collaborative approach towards brand promotion. This frequently involves retailers allowing business affiliates to place product links for products or services on their sites and earn a commission for each click, with user activity often measured by dropping third-party cookies and using them to track individual behavior. Conversion, Conversion Pixel, Cost Per Click (CPC)
    See also:

    Conversion, Conversion Pixel, Cost Per Click (CPC)

    Anomaly Detection

    Anomaly Detection is a process of identifying any patterns that deviate in any way from the standard behavior in any given dataset. In marketing, it is often undertaken AI in the form of machine learning algorithms to identify any outliers in marketing or ad performance so that marketers can take action to avoid a reduction in ROAS or spot new opportunities.
    See also:

    ROAS

    Further reading:

    What is Anomaly Detection (and How Does It Help Marketing Campaigns)?

    API connector

    An API Connector is the mechanism through which information can be passed between two applications with APIs. While APIs create the infrastructure that allows different platforms to talk to each other to actually send data to and from a platform, you need an API connector.
    See also:

    Application Programming Interfaces (APIs)

    Further reading:

    Connector Vs. API: What’s The Difference?

    API Documentation

    API Documentation is a set of instructions for plugging into APIs. Because each API is different, connection requirements will also vary, with documentation covering how connectors should be built, authenticated, used, and maintained.
    See also:

    Application Programming Interfaces (APIs)

    Further reading:

    API Updates: How To Deal with Connector Maintenance

    Append

    Append is a function that adds new rows of data at the end of an existing dataset. This 'add-on' function is not the same as overwriting or inserting data, as existing data remains unchanged and in its present format. Appending actions can be performed using a data integration platform as a type of data transformation.
    See also:

    Data Integration Platform, Data Transformation

    Application Programming Interfaces (APIs)

    Application Programming Interfaces (APIs) allow different tools to talk to each other. They are used extensively to connect data sources with destinations so that data can be easily and automatically sent from one to another.
    See also:

    API Documentation

    Attribution Model

    In marketing, Attribution refers to how conversions (whether that is a click, a sale, or something else) are attributed to different activities. Different attribution models use different approaches to calculate this. For example, a First Interaction model would attribute 100% of the conversion to the customer's first activity in relation to your company (say, clicked on an ad). Conversely, a Last Interaction model attributes a conversion to the last action they took before converting. In between that, various models exist (such as Linear, U-Shaped, W-shaped) that look at all of a customer's actions between the first and the last, placing different values on each of them depending on the model.
    See also:

    First interaction, Last interaction, Linear, Conversion

    Average Order Value (AOV)

    Average Order Value (AOV) is a metric that measures how much each customer spends on average. Calculation involves dividing total revenue by either transactions, sales, orders, or purchases.
    See also:

    Cost Per Order (CPO), Cost Per Acquisition (CPA), Cost Per Lead (CPL)

    Average Position

    Average Position is a KPI typically used in search engine marketing and advertising (most notably Google) to indicate the average ranking position of a business' ads or webpages in the search results for specific keywords. It helps businesses understand how visible their ads or webpages are in search engine results, which can significantly impact click-through rates and overall campaign effectiveness.
    See also:

    Search advertising, Click-through rate (CTR)

    B

    B

    Bounce Rate

    Bounce Rate is a KPI that measures the percentage of visitors to a site who navigate away quickly, often after viewing just one webpage or landing page for a short period. It's worth noting that different platforms measure this differently. Hubspot, for example, measures a bounce rate as the percentage of people who land on a page on your website then leave without clicking anything else. GA4, however, measures the bounce rate as the percentage of unengaged sessions with engaged sessions defined as sessions that last longer than 10 seconds, have a key event, or have at least two pageviews or screenviews. Unsurprisingly, this means the bounce rate from these two sources will yield very different results despite sharing a name.
    See also:

    Session, Engaged Sessions

    Buying Model

    A Buying Model, or Buying Type, refers to what marketers and advertisers pay for in each transaction. Although most of the time, this is on the basis of cost per thousand impressions (CPMs), approaches differ depending on unique brand and campaign goals. For instance, marketers aiming to bolster website traffic may opt to align purchases with clicks rather than just served impressions or views, meaning their preferred buying model is cost per click (CPC).
    See also:

    Impression, Cost Per Thousand (CPM), Cost Per Click (CPC)

    C

    C

    Calculated KPI

    A Calculated KPI is simply a KPI created by adding, dividing, subtracting, or multiplying existing KPIs to create a new KPI. For example, bounce rate is the number of bounces divided by the number of sessions, Cost Per Click (CPC) is the overall cost divided by the total number of clicks, and so forth.
    See also:

    KPI

    Call to Action (CTA)

    A Call to Action, or CTA, refers to any design or phrase specifically crafted to prompt an immediate response, sale, or conversion. It usually takes the form of a button, link, or some type of graphic that prompts the viewer to take some steps, such as "Subscribe Now," "Call Today," or "Buy Now".
    See also:

    Online Advertising, Goal.

    Campaign Objective

    A Campaign Objective is the primary goal of a marketing or advertising campaign. This is generally set at the start of any initiative to provide a benchmark against which activity can be optimized and measured to drive the best possible performance. Broad goals might include driving greater general awareness of brands, products, or services, while specific goals may refer to achieving a certain number of defined conversions or sales.
    See also:

    Goal, Conversion, Conversion Rate, Online Advertising.

    Clean Room

    A Clean Room is a technology solution that enables different parties to share their user-level, first-party data, in a secure environment, and without displaying any personal identifiable information (PII) to one another, therefore respecting and complying with users’ privacy regulations, such as GDPR.
    See also:

    General Data Protection Regulation (GDPR)

    Further reading:

    What is a Clean Room for Data and Why Should Marketers Care?

    Click-Through Conversion (CTC)

    Click-Through Conversion is a metric that measures how many clicks on an advertisement or link result in an actual conversion, such as a sale, a sign-up, or another predefined goal. It essentially measures the effectiveness of the click-through rate (CTR) in leading to actual business outcomes or user actions. While CTR indicates the attractiveness or engagement level of an ad, click-through conversions measure the effectiveness of the ad in achieving its final goal.
    See also:

    Click-Through Rate (CTR), Conversion

    Click-Through-Rate (CTR)

    Click-Through-Rate (CTR) is a metric that tracks the total number of clicks on any given link or button divided by the total number of impressions. Used extensively to measure how many times ads or CTAs have successfully encouraged users to click compared to the overall number of people that have seen the ad or CTA.
    See also:

    Click-Through Conversion (CTC)

    Cohort Analysis

    Cohort Analysis is a type of data analysis that groups customers or users into segments based on a common characteristic shared over time. Cohort analysis lets marketers analyze and compare a huge range of factors depending on the data, such as buying patterns, conversion trends, or product popularity.
    See also:

    Marketing Mix Modeling

    Further reading:

    What is Cohort Analysis, and Why Should Marketers Care?

    Comma Separated Value (CSV) File

    A Comma Separated Value (CSV) File is a type of flat file that uses commas to separate values (data) in a clear and simple visual format. They are often used as a way of exchanging tabular data — numbers and text — between tools, programs, or systems.
    See also:

    Flat File, Tab Separated Value (TSV) File

    Continuous Data

    Continuous Data is data that includes values that can be measured on a continuous scale and can take any value within a specified range. In other words, continuous data can assume any numeric value, including fractions and decimals (a floating point number). The nature of continuous data means that it can be meaningfully split into smaller parts, making it ideal for analysis. For example, if you're measuring time spent on a website, the data can be very specific, such as 2 mins 33 seconds instead of being rounded to the nearest whole number. Compare with discrete data.
    See also:

    Discrete Data, Floating Point Number

    Further reading:

    Discrete Vs. Continuous Data: What's The Difference?

    Conversion

    A Conversion refers to any specific action that marketers want people to take. This might be ad clicks, making a purchase, signing up for a newsletter, or any other number of actions a business determines are relevant to its goals.
    See also:

    Click-Through Conversion (CTC), Conversion Rate, Conversion Pixel

    Conversion Pixel

    A Conversion Pixel is a small piece of code (often an invisible 1x1 pixel image) embedded on a webpage that triggers when specific actions (conversions) are completed. Marketers can then use this information to track the effectiveness of a campaign. Conversion pixels differ from cookies in that the tracking is linked directly to specific events rather than the broader browsing behavior of the user. However, they can still be used to track personal data and their usage is regulated by legislation such as GDPR.
    See also:

    Conversion, Conversion Rate

    Conversion Rate

    Conversion Rate is a metric that measures the total percentage of users reached by a particular marketing campaign or activity that went on to convert.
    See also:

    Conversion, Click-Through Conversion Rate (CTC).

    Cost Per Acquisition (CPA)

    Cost Per Acquisition (CPA) is a metric that measures the average cost of acquiring a single customer. Often used interchangeably with cost per conversion; however, in reality, cost per acquisition should be considered an example of cost per conversion since acquisitions themselves are a type of conversion.
    See also:

    Cost per Conversion, Click-through Conversion

    Cost Per Click (CPC)

    Cost Per Click (CPC) is a metric that measures the average amount an advertiser pays each time a user clicks on one of their online ads. It's a common pricing model used in online ad campaigns, particularly with search engines and social media platforms.
    See also:

    Buying Model, Search Engine Advertising (SEA), Click-Through Conversion (CTC)

    Cost Per Conversion (CPC)

    Cost Per Conversion is a metric that measures the average cost it takes for any individual given conversion and calculated as total cost divided by total conversions. Since there are multiple types of conversion, there are multiple metrics that could be said to be a type of cost per conversion, such as cost per acquisition, cost per lead, or cost per order. Frustratingly, "CPC" is also the acronym for the more commonly known metric, Cost Per Click. As a result, the term Cost Per Conversion is increasingly used less often with marketers instead favoring more specific types of conversion metrics.
    See also:

    Conversion, Click-Though Conversion

    Cost Per Lead (CPL)

    Cost Per Lead is a metric used to measure the average cost per customer lead. More commonly used by B2B companies where a lead is generally followed up by a sales team. Can be considered a type of Cost per conversion where lead generation is a specified conversion activity.
    See also:

    Conversion, Cost Per Conversion

    Cost Per Mille/Thousand Impressions (CPM)

    Cost Per Mille, or Cost Per Thousand Impressions (CPM), is a metric used to measure the average cost per thousand impressions of an ad campaign. Frequently used as a buying model for digital advertising transactions, with media buyers making purchases based on the maximum price they are willing to pay per 1,000 impressions
    See also:

    Conversion, Impression, Buying Model

    Cost Per Order (CPO)

    Cost Per Order is a metric that measures the average amount it costs a business to gain one order or purchase. Generally a measure of how much businesses have invested in driving each order or purchase from their customers. Can be considered a type of cost per conversion where a purchase or order is considered a specific conversion.
    See also:

    Cost Per Conversion (CPA)

    Cost Per View (CPV)

    Cost Per View is a metric that measures the average cost needed to gain one view of a video. Often used as a buying model where payment is made on the basis of video views. But what counts as a 'view' can vary between platforms. For example, YouTube will issue charges once ads have been either watched for 30 seconds or sparked an interaction, while Facebook lets buyers select from a range of viewing durations.
    See also:

    Buying model

    Customer Lifetime Value (CLV)

    Customer Lifetime Value (CLV) is a metric that estimates the total revenue a business can reasonably expect from a single customer account throughout the business relationship. The length of this relationship can span from the customer's first purchase to their final interaction with the business. CLV is particularly important in understanding how much a company should invest in acquiring new customers and retaining existing ones based on the profit each customer is expected to generate over their lifetime. Calculating CLV for each customer requires three pieces of data: the average order value, the customer's purchase frequency, and the customer value.
    See also:

    Average Order Value (AOV)

    Customer Relationship Management (CRM)

    Customer Relationship Management Systems (CRMs) are tools that manage all interactions with customers. They bring together every contact point, decision, and purchase in association with unique profiles, helping companies map interactions and maximize their impact. Often combined with other data so companies can measure the impact of campaigns further down the funnel.
    See also:

    Attribution Model

    D

    D

    Dashboard

    A Dashboard is a collection of visualizations and other widgets that display a variety of related data in an easy-to-digest format. For example, marketers might have individual dashboards for specific campaigns depicting all the cross-channel data related to that campaign in a single place. Dashboards tend to be updated automatically to ensure that, at any given time, they are displaying the most up-to-date data.
    See also:

    Visualization, Widget, Dashboard Templates

    Further reading:

    How To Create An Effective Marketing Dashboard

    Dashboard Template

    A Dashboard Template is a generic version of a marketing dashboard that can be cloned and re-used for multiple campaigns. Often used by advertising or marketing agencies to simplify onboarding new clients, it is also used by in-house teams to ensure consistency when tracking the impact of different campaigns. Dashboard, Visualization
    See also:

    Dashboard, Visualization

    Data Cleaning

    Data Cleaning, sometimes called Data Cleansing, is the removal or correcting of errors, inconsistencies, duplicates, or inaccuracies in a dataset to improve data quality and reliability.
    See also:

    Data Transformation, Data Enrichment

    Data Democratization

    Data democratization is the concept of ensuring that all employees, including non-specialists with lower data literacy levels, have the ability to access and gather accurate data about the business without any outside help. Data democratization is considered a fundamental part of fostering a data-driven business.
    See also:

    Data Maturity

    Further reading:

    What is Data Democratization?

    Watch the video:

    Data Enrichment

    Data Enrichment, sometimes known as data enhancement, is the process of adding additional data to a dataset. For example, adding country data to a dataset that includes cities, adding an additional field to the dataset with a calculated KPI based on existing fields, or separating the values in one field into two separate fields. Often used interchangeably with Data Transformation. Data enrichment is sometimes a necessary process when integrating two or more datasets to ensure compatibility between the data.
    See also:

    Data Integration, Metadata, Calculated KPI, Data Transformation

    Further reading:

    What is Data Enrichment and Why is it Important?

    Data Flushing

    Data Flushing, also known as Data Purging or Data Erasure, refers to the systematic and permanent removal of data from a company's databases or storage systems. This process ensures that outdated, inaccurate, or sensitive information is expunged, reducing the risk of unauthorized access, data breaches, and potential misuse. This practice is especially important when businesses collect and store vast amounts of customer data, as it directly relates to privacy concerns and legal obligations regarding data protection.
    See also:

    See also Data Governance, General Data Protection Regulation (GDPR)

    Further reading:

    What is a Data Flush (and why should marketing care)?

    Data Integration

    Data Integration, sometimes known as Data Blending, is the process of combining data from different sources into a single structured data set. The process involves data mapping, transforming, or enriching data from individual datasets so that they can be effectively combined together. It can either be done manually, though this is very time-consuming and will potentially lead to errors, or automated using a data integration platform.
    See also:

    Data Integration Platform, Manual Data Integration, Data Mapping, Data Enrichment, Data Transformation

    Further reading:

    What is Data Ingestion Vs Data Integration?

    Data Integration Platform

    A Data Integration Platform is a platform or tool designed to automate the process of integrating data from multiple data sources into a single dataset to be used for analysis.
    See also:

    Data Integration, Manual Data Integration

    Further reading:

    Top 17 Data Integration Tools

    Data Lake

    A Data Lake is a data storage facility generally used to hold raw, unstructured data in its native format until needed. Compare with data warehouses, which are generally used to store structured, harmonized data, as well as data lakehouses, a combination of the two.
    See also:

    Data Warehouse, Data Lakehouse

    Further reading:

    What is a data lake vs a data warehouse (and should marketers care)?

    Data Lakehouse

    A Data Lakehouse is a data storage facility that combines the features of a data lake and a data warehouse into a single integrated platform. It integrates the scalability and flexibility of a data lake with the structured query capabilities and performance optimizations of a data warehouse, offering the best of both worlds for managing and analyzing large volumes of data.
    See also:

    Data Warehouse, Data Lake

    Further reading:

    What is a Data Lakehouse and Why Do You Need One?

    Data Mapping

    Data Mapping is the process of connecting and standardizing data from one data source with another. It involves identifying specific fields in both data sources and mapping the values to a shared target field. It is part of the overall process of data integration. This may require some prior data transformation or data enrichment to ensure that all the integrated data in the new target field is compatible with each other.
    See also:

    Data Transformation, Data Enrichment, Data Integration

    Further reading:

    What is Data Mapping and Why Should it Matter to Marketers?

    Watch the video:

    What is data mapping?

    Data Maturity

    Data Maturity refers to how developed a business is when it comes to their data and their data-driven approach. Businesses that have implemented measures such as automated data integration, a single source of truth, or AI tools, are considered to have a high data maturity. Businesses that rely on spreadsheets and manual data integration are generally considered to have a low data maturity.
    See also:

    Data Integration Platform, Single Source of Truth, Manual Data Integration

    Further reading:

    What Is A Data Maturity Model And How Can It Drive Business Efficiency

    Data Mesh

    Data Mesh is a methodology that aims to design data architecture in a modular manner. It aims to provide a common set of tools that allow the provisioning, access control, data catalog, and metrics layer across an organization to be leveraged by different stakeholders. In essence, Data Mesh is a move away from centralized data warehouses or data lakes to data as Lego pieces that each stakeholder and/or department can combine as they see fit to enable their particular use cases.
    See also:

    Data Warehouse, Data Lake

    Further reading:

    What is a Data Mesh (and why should marketers care)?

    Data Model

    A Data Model is a blueprint for how a system or organization organizes and manages data, providing a standardized framework for understanding and interacting with data elements.
    See also:

    Data Warehouse

    Further reading:

    What are Marketing Data Analytics Models?

    Data Operations

    Data Operations, or DataOps, is the function of either a person or team that’s responsible for finding the most efficient way to collect and clean data within the company. A large part of this role is to both create and maintain a single source of truth for the business.They collaborate closely with the reporting function to understand what data needs the business has, and then build out the data architecture around those needs.
    See also:

    Single Source of Truth

    Further reading:

    What is DataOps?

    Data Silo

    Data Silos are isolated data sets that live within a single source or, most commonly, within a particular team, and which are not accessible by other data systems or teams. Data silos are, generally speaking, considered problematic in marketing as they reduce visbility over marketing campaigns as well as the ability of marketers to make informed decisions.
    See also:

    Data Democratization, Single Source of Truth

    Further reading:

    What are Data Silos (and How Do They Impact Marketers)?

    Data Source

    A Data Source is any platform or application that generates data. Commonly, data is fetched from a data source via an API and sent to a destination, whether that is data integration platform, a data warehouse, or a data lake.
    See also:

    Datastream, Data Integration Platform, Data Warehouse, Data Lake

    Data Stack

    A Data Stack refers to a collection of technologies and tools employed for gathering, analyzing, and preserving data. In the context of modern data handling, the term "modern data stack" describes a cloud-centric approach that emphasizes flexibility and prioritizes the delivery of processed data. It comprises various software tools that work together to provide a comprehensive solution for handling data.
    See also:

    Data Integration Platform

    Further reading:

    What is The Ideal Modern Data Stack for Marketing?

    Data Transformation

    Data Transformation is the process of converting or modifying data from one format, structure, or representation to another. It involves manipulating data to meet specific requirements, standards, or objectives, such as improving data quality, compatibility, or usability for analysis, reporting, or integration purposes.
    See also:

    Data Integration, Data Enrichment, Data Cleaning

    Further reading:

    What is Data Transformation (and Why Should Marketers Care)?

    Data Validation

    Data Validation is a set of checks that compares your data against the preset rules with the aim of identifying if the received data is complete and correct. Data validation plays a vital role in maintaining the reliability and quality of data within the ELT process. By validating the completeness and accuracy of data at each stage of transformation and loading, organizations can minimize the risk of making decisions based on inaccurate or faulty information.
    See also:

    Data Governance, Extract, Load, Transform (ELT)

    Further reading:

    What is a Completeness Check In Data Validation?

    Data Warehouse

    A Data Warehouse a data storage facility generally designed to hold processed data in a structured format according to a specific data model. They can be hosted on-premises or in the cloud, with common examples of the latter including Amazon Redshift and Snowflake. Compare with data lakes which are generally used to hold raw unstructured data in it's native format, as well as data lakehouses which is a combination of the two.
    See also:

    Data Lake, Data Lakehouse

    Further reading:

    What is a data lake vs a data warehouse (and should marketers care)?

    Database

    A Database is an organized collection of data. They can be either relational or non-relational. It is the most basic building block of any data architecture, where data can live and is usually isolated. A database can become part of a data warehouse where combining multiple databases is possible, as well as running analysis across all sources.
    See also:

    Relational Database, Non-relational Database, Data Warehouse

    Database Schema

    Database Schema provide a formal definition of how data is organized and stored within a database, specifying the names, data types, and constraints for each table and column.
    See also:

    Database, Data Model

    Datastream

    A Datastream is the flow of data from a data source to a destination. Typically datastreams are set up via an API connector to automatically fetch data from individual data sources and send it to a destination such as a data lake, a data warehouse, or, in the case of Adverity, a data integration platform.
    See also:

    Connector, API, Data Source, Destination

    Destination

    In data management and integration, a Destination refers to the endpoint or repository where data is delivered and stored after being collected, processed, or transformed from its original sources. Typical destinations are data warehouses or BI tools.
    See also:

    Connector, ETL, Data Source

    Discrete Data

    Discrete Data is data that is countable, finite, and usually comes in the form of whole numbers (integers). Common examples in marketing include number of clicks and website visits - since these numbers are neccessarily whole numbers (you cannot have half a click). Compare with continuous data.
    See also:

    Continuous Data, Integer

    Further reading:

    Discrete Vs. Continuous Data: What's The Difference?

    Display Advertising

    Display Advertising is a common type of online advertising that involves using visually engaging media that usually link to brand sites or product pages. These ads can appear on websites, apps, and social media platforms and in various formats, such as banner ads, interstitial ads, or rich media ads. An integral pillar of the digital media space, display advertising is a common mechanism for many web publishers to generate income and continues to feature heavily in ad campaigns, particularly those managed using programmatic advertising.
    See also:

    Online Advertising, Ad Exchange, Buying Model, Programmatic Advertising

    E

    E

    Email Service Provider (ESP)

    An Email Service Provider (ESP) is the software used by companies to manage their email marketing. Common examples include Mailchimp, ActiveCampaign, and Klaviyo.
    See also:

    Organic

    Engaged Sessions

    Engaged Sessions refer to website or app sessions during which users actively interact with the content, features, or functionalities, indicating a high level of engagement. Unlike passive sessions where users may simply browse or view content without taking any action, engaged sessions involve meaningful interactions such as clicking on links, filling out forms, watching videos, making purchases, or spending significant time on the site or app.
    See also:

    Bounce rate

    Extract, Load, Transform (ELT)

    Extract, Load, Transform (ELT) is a variation of ETL. Instead of transforming the data first and then loading it into a target database or data warehouse, an ELT process first loads the data and then performs any necessary cleaning up processes. While it requires less from the original data source(s) – only the raw data generated by them - it demands more from the target data warehouse as it has to handle the transformation of large amounts of messy, unprepared data.
    See also:

    Extract, Transform, Load (ETL)

    Further reading:

    ETL vs. ELT: Why They’re Complementary Data Integration Solutions

    Extract, Transform, Load (ETL)

    Extract, Transform, Load (ETL) is the process of extracting data from various sources, transforming it into a suitable format or structure, and loading it into a target destination such as a data warehouse, database, or data lake. ETL is commonly used in data integration and data warehousing to consolidate, clean, and organize data from disparate data sources for analysis, reporting, or business intelligence purposes.
    See also:

    Data Integration, Data Warehouse

    Further reading:

    ETL – What is it and How Do I Do it?

    F

    F

    First Interaction

    First Interaction is a type of attribution model whereby conversions are entirley attributed to a customer or user's first interaction. For example, if a user clicked an ad, went to your website, viewed several pages, then bought a product, then 100% of that revenue would be attributed to the ad they clicked.
    See also:

    Attribution Model, Last Interaction, Linear

    Flat File

    A Flat File is a simple type of database that stores data in a plain text format typically consisting of a single table of data, with data fields separated by delimiters such as commas (in CSV, or comma-separated values, file) or tabs (TSV, or tab-separated values, file). This format is often used as a common medium to transfer data between systems that may not share a common model or standard or for simple data storage.
    See also:

    Comma Separated Value (CSV) File, Tab Separated Value (TSV) File

    Floating Point Number (Float)

    A Floating Point Number, more commonly known as a Float, is any number that has a decimal place. Floats can have zero, positive, or negative values (for example, 0.13, -3.5, 10.00). When classifying numerical data, it is important to specify whether a metric is a floating number point (has a decimal place) or an integer (is a whole number).
    See also:

    Integer

    Frequency

    Frequency is a metric that measures the number of times a target audience is exposed to a particular advertisement or marketing message within a specific period. This metric is crucial for understanding the impact of an advertising campaign's reach and effectiveness and can also help to avoid sparking user irritation through excessive retargeting. Frequency is calculated by dividing the total number of ad impressions by the reach (the number of unique viewers). For example, if an ad has 1,000 impressions and reaches 250 unique viewers, the frequency would be 1,000/250 = 4. This means, on average, each viewer saw the advertisement four times.
    See also:

    Impression, Reach.

    G

    G

    General Data Protection Regulation (GDPR)

    General Data Protection Regulation (GDPR) is a digital privacy law that has established blanket security legislation across the European Union so that all countries operate to the same standards. Organizations within the EU must comply with GDPR requirements by incorporating online privacy settings and keeping them switched on at all times.
    Further reading:

    What is GDPR and Why is it Important? — A Guide for Marketers

    Gross Ratings Point (GRP)

    Gross Ratings Point (GRP) is a metric that measures the total level of exposure an ad or ad campaign can achieve in a specific media market. A long-time staple of TV advertising, GRP helps advertisers understand the impact of their advertising efforts in terms of exposure among their target audience. GRP is calculated as Reach (%) x Frequency such that if, for example, an ad campaign reaches 50% of the target audience and the average viewer sees the ad four times, the GRP would be 200.
    See also:

    Reach, Frequency

    I

    I

    Impression

    Impressions is a metric that measures the number of times an ad (or any form of digital content) is displayed on a user's screen. Impressions are counted regardless of whether the ad is clicked or not, simply measuring how often it appears. Impressions are generally used to understand the reach of digital campaigns across various digital marketing platforms, including social media, search engines, and display advertising networks.
    See also:

    Cost per Thousand (CPM), Reach

    Impression Share

    Impression Share is a metric that measures the percentage of times an ad is actually shown in relation to the total number of times it was eligible to be shown. Often used in Pay-per-click (PPC) campaigns to provide insights into the extent to which an advertiser's ads are being displayed compared to the total opportunities available within a specific campaign or ad group.
    See also:

    Impression, Cost per Thousand (CPM), Pay-per-click (PPC)

    In-Stream Ad

    In-Stream Ads are ads delivered within video content, usually as a pre-roll message that appears before the video. Especially relevant for YouTube, but can apply to other video platforms. They are distinct from in-display ads, which are served beside content on those platforms, such as in search results, in the same manner as regular display ads.
    See also:

    Display Advertising

    Integer

    An Integer is any number that is a whole number i.e., it has no decimal places. Integers can have zero, positive, or negative values (for example, 10, -7, or 0). When classifying numerical data, it is important to specify whether a metric is a floating number point (has a decimal place) or an integer (is a whole number).
    See also:

    Floating Number Point

    J

    J

    Javascript Object Notation (JSON)

    JavaScript Object Notation (JSON) is a data format that is easy for humans to read and write and easy for machines to parse and generate. It is widely used as a data format in various programming languages, including JavaScript, due to its simplicity and flexibility.
    See also:

    Flat Files, SQL

    Join

    A Join, sometimes referred to specifically as an SQL Join, is a method used to combine rows from two or more tables based on a related column between them. Joins are often used to enrich and blend data from different sources into a single standardized dataset.
    See also:

    Structured Query Language (SQL), Data Integration, Data Enrichment

    K

    K

    Key Performance Indicator (KPI)

    Key Performance Indicators (KPIs) are specific metrics that are deemed critical to the success of an organization, department, project, or campaign. However, while all KPIs are metrics, not all metrics are KPIs. KPIs are strategic indicators that focus on high-level goals and objectives, whereas metrics are more granular measures that provide detailed insights into specific aspects of performance.
    See also:

    Metric

    Further reading:

    What is Data vs a Metric vs a KPI vs a Report?

    Keyword

    Keywords, also known as Keyword Phrases, are specific words or phrases that describe the main topics or themes of a piece of content, webpage, or search query. In the context of Search Engine Optimization (SEO), keywords play a crucial role in helping search engines (such as Google) understand how relevant a piece of content is to a user's search query. The more relevant a piece of content is to a user's search, the higher it will rank on the search engine results page (SERP) and, therefore, more likely to be clicked. In search advertising, keywords play a fundamental role in determining when and where ads appear on search engine results pages. In PPC (Pay-Per-Click) advertising campaigns, advertisers bid on keywords relevant to their products or services. When users search for those keywords, the ads appear at the top of the search results, and advertisers pay a fee each time their ad is clicked.
    See also:

    Search advertising, Search Engine Optimization (SEO), Pay-per-click (PPC)

    L

    L

    Last Interaction

    Last Interaction is a type of attribution model whereby conversions are entirley attributed to a customer or user's last interaction. For example, if a user clicked an ad, went to your website, viewed several pages, then bought a product, then 100% of that revenue would be attributed to the last page they were on before making the purchase.
    See also:

    Attribution Model, First Interaction, Linear

    Linear

    Linear is an attribution model that attributes whereby conversions are attributed equally to all the interactions a user takes before converting. For example, if a user clicked an ad, went to your website, viewed several pages, then bought a product, then the total revenue for that purchase would be shared among all those activities.
    See also:

    Attribution Model, Last Interaction, First Interaction

    Location Data

    Location Data is information that can be used to tie an ad or consumer’s behavior to a specific geographical region. By tying marketing data to a specific location, marketers can analyze a campaign’s performance across regions and countries.
    Further reading:

    What is Location Data and Why is it Important for Marketers?

    Watch the video:

    M

    M

    Manual Data Integration

    Manual Data Integration is the process of manually combining the data from multiple data sources into a single dataset, often using a spreadsheet and manually copying and pasting data. Manual data integration is a highly time-consuming and resource-intensive task and can cause multiple mistakes and add inconsistencies to the data due to human error. Compare with data integration platforms that automate this process.
    See also:

    Data Integration Platform, Data Integration

    Further reading:

    7 Best Practices To Move Away From Manual Data Integration

    Marketing Mix Modeling

    Marketing Mix Modeling (sometimes called Market Mix Modeling or Media Mix Modeling) is a method for measuring the performance of different marketing channels using aggregated data. While methods such as attribution modeling take a bottom-up approach, analyzing how different channels and campaigns contribute to sales at an individual user level, MMM uses statistical modeling to find correlations between marketing investments and returns at a macro level.
    See also:

    Attribution Model, Cohort Analysis

    Further reading:

    What is Marketing Mix Modeling (and how can it help with attribution)?

    Marketing Tag

    A Marketing Tag or Website Tag is a segment of code, written in Javascript, that is placed on a website in order to track user actions and collect data. When a visitor undertakes a specific action that is configured to trigger the tag, that tag is said to have ‘fired’ - in other words, it collects data about that activity and sends it to, most commonly, an analytics or marketing tool. There are different types of tags depending on the function they serve. For instance, Pageview Tags monitor visits to different pages on a website and Conversion Tags track every time a sale is made or a lead form is filled out.
    See also:

    Conversions

    Further reading:

    What is a Marketing Tag (and what are they used for)?

    Watch the video:

    Media Plan

    A Media Plan is a strategic document that outlines the details of how an advertiser will use various media channels to achieve its marketing objectives. It is a comprehensive roadmap that guides the execution of advertising campaigns, detailing the allocation of resources, selection of media channels, scheduling of ad placements, and measurement of campaign effectiveness.

    Metadata

    Metadata refers to data that provides information about other data, such as the content, structure, format, or context. Metadata is effectively additional data about your primary data (metrics, KPIs) that helps to organize, manage, and understand that data. For example, this might include campaign names, campaign start/end dates, campaign ad creative copy, or target audience information.
    See also:

    Metric, Key Performance Indicator (KPI)

    Metric

    A Metric is a quantifiable measurement used to assess the performance, progress, or effectiveness of a specific process, activity, or system. Metrics are the first step in making sense of the raw data and applying it to real-life situations. Where raw data is generally a number without context, metrics are specific measurements such as Revenue, Clicks, or Impressions.
    See also:

    Key Performance Indicators (KPIs)

    Further reading:

    What is Data vs a Metric vs a KPI vs a Report?

    Metrics Layer

    A Metrics Layer is a method for standardizing metrics and KPIs across an entire company or organization in order to prevent different definitions or calaculations for the same metric being used by different teams. The idea of the metrics layer is to standardize and centralize the definition of metrics so that every time a user retrieves data from a system, they are working with metrics that use the same calculation. As a data process, the metrics layer sits along the data retrieval pipeline and performs calculations on metrics from different sources based on standardized definitions.
    See also:

    Data Governance, Metric

    Further reading:

    What is a Metrics Layer?

    N

    N

    Naming Convention

    In marketing, a Naming Convention refers to a standardized system or format used to name and organize various elements, assets, or components associated with a campaign. For instance, a classic example might be "campaign name-country-brand-objective". Implementing and adhering to naming conventions is essential to ensure consistency and enable efficient data evaluation across data sources and teams.
    See also:

    Metadata

    Further reading:

    Marketing Campaign Naming Conventions Best Practices

    Non-Aggregatable Metric

    A Non-Aggregatable Metric, or a Nonag, is a quantitative value that can’t be summed or averaged. This is because this value depends on a more granular level of data that is not given. As a result, such metrics can only be displayed at their given granularity level, otherwise, discrepancies will be generated. Common examples include reach, followers, or caculated KPIs such as CPC.
    See also:

    Reach, Calcuated KPIs, Cost Per Click (CPC)

    Further reading:

    What is a Non-Aggregatable Metric?

    O

    O

    Offline Advertising

    Offline Advertising refers to traditional marketing efforts that are conducted through non-digital channels or media. This covers traditional mediums such as print (newspapers, magazines, flyers, and brochures), linear TV, radio, and offline outdoor billboards. The performance of such activity is often less trackable than online advertising, with limited access to audience data.
    See also:

    Online Advertising, Out of Home (OOH).

    Online Advertising

    Online Advertising, also known as Internet Advertising or Digital Advertising, refers to advertising on digital channels or platforms on the Internet. This extends across display, search, social media, direct email marketing, and e-commerce platforms or websites. As opposed to offline advertising, online advertising generally has far more data for marketers.
    See also:

    Offline Advertising, Display Advertising, Search Advertising

    Organic

    Organic refers to any marketing activity that uses non-paid methods such as Search Engine Optimization (SEO), social media engagement, content marketing, email marketing, or PR. While organic activities are not necessarily free (there is always a cost in terms of resources), they are not paid for directly as with other forms of marketing, most notably paid advertising. The term Organic is often used to describe the source of a given customer interaction, conversion , or sale such as organic traffic, organic leads, organic revenue, etc.
    See also:

    Paid, Search Engine Optimization (SEO)

    Out-Of-Home (OOH) Advertising

    Out-of-home (OOH) Advertising, also known as Outdoor Advertising, refers to advertising in public areas such as traditional paper billboards and connected digital screens. Ad locations include city centers, retail parks, and sports stadiums, with ad impact generally estimated and reported on a weekly or monthly basis.
    See also:

    Offline Advertising, Online Advertising

    P

    P

    Paid

    Paid refers to any marketing activity that uses paid methods such as paid advertising, sponsorship, or paid media partnerships. Generally, paid is distinguished from organic, which uses non-paid methods to market or otherwise promote a business or product. The term Paid is often used to describe the source of a given customer interaction, conversion, or sale, such as paid traffic, paid leads, paid revenue, etc.
    See also:

    Organic, Display Advertising

    Pay-Per-Click (PPC)

    Pay-Per-Click (PPC) is an advertising model where advertisers pay a fee each time their ad is clicked by a user. It is a type of online advertising commonly used in search advertising, display advertising, social media advertising, and other digital marketing channels. In a PPC campaign, advertisers bid on keywords or target audience segments, and their ads are displayed to users based on relevant search queries, demographics, or interests. Advertisers are only charged when a user clicks on their ad, regardless of how many times the ad is displayed (impressions).
    See also:

    Search Advertising, Real-time Bidding

    Placement

    Ad Placements refer to where ads are served. The level of specificity with which this term is used can be highly varied, ranging from just one ad unit to a group of ads or even an entire website. By far, the easiest way to think of placements is as ad spots, which can be split across different channels and aimed at different target audiences.
    See also:

    Online Advertising, Offline Advertising, Ad Server

    Primary Key

    A Primary Key is a field or set of fields that are designated within a database table to hold a unique identifier for each row. For example a campaign ID, a product ID, or even a date. Primary keys are important to the structure and functionality of relational databases. They are also crucial when joining data from two tables. When two tables have a common primary key relationship, they can be efficiently linked to retrieve related information.
    See also:

    Relational Database, Metadata

    Programmatic Advertising

    Programmatic Advertising is the automated buying and selling of digital advertising space in real-time through algorithms and technology platforms. Instead of traditional manual processes involving negotiations and human intervention, programmatic advertising uses data, machine learning, and artificial intelligence to optimize ad placements and target specific audiences with greater precision and efficiency. A key component of programmatic advertising is Real-Time Bidding (RTB), whereby advertisers automatically bid on ad impressions in real-time auctions according to preset budgetary limits, competing with other advertisers to have their ads displayed to a given target audience. The auction often takes place through an ad exchange, such as Google Display & Video 360, coordinating bidding on key placements or sites meeting certain targeting criteria.
    See also:

    Online Advertising, Ad Exchange, Buying Model

    Python

    Python is a high-level, general-purpose programming language commonly used for data analysis, machine learning, and AI.
    See also:

    Structured Query Language (SQL)

    R

    R

    Reach

    Reach is a metric that measures the number of recipients of a particular message or campaign. Generally, reach is measured in terms of number of people, though could be measured in a number of alternative ways, such as households or number of cars. Reach differs from impressions, which measure the number of times an ad was displayed but not necessarily seen. In this sense, the reach is the count of unique impressions received over a certain period of time.
    See also:

    Impression, Impression Share, Frequency, Gross Rating Point (GRP)

    Real-Time Bidding (RTB)

    Real-Time Bidding (RTB), sometimes called Real-Time Advertising (RTA), is a key component of programmatic advertising whereby advertisers automatically bid on ad impressions in real-time auctions according to preset budgetary limits, competing with other advertisers to have their ads displayed to a given target audience. It is sometimes considered a synonym for programmatic advertising.
    See also:

    Programmatic Advertising

    Relational Database

    A Relational Database is a type of database that organizes data into tables with rows and columns, where each row represents a record and each column represents a specific attribute or field. Importantly, a relational database allows relations between tables using keys, such as primary keys and foreign keys, allowing data to be connected across multiple tables. This enables efficient storage, retrieval, and manipulation of data, making it easier to manage and analyze large datasets. Users can interact with relational databases using SQL to perform various operations on the data stored in the database, such as retrieving specific records, filtering data based on criteria, sorting results, and aggregating values. Relational databases generally store structured data, and for this reason, most data warehouses are based on relational databases. Relational databases can be compared to non-relational databases, which do not have the same uniform structured nature as relational databases and are, therefore, more adept at storing large amounts of raw, unstructured data.
    See also:

    Structured Query Language (SQL), Database Table, Non-Relational Database

    Return On Ad Spend (ROAS)

    Return on Ad Spend (ROAS) is a key performance indicator (KPI) that measures the revenue returned for every dollar of ad spend committed for a specific campaign. In marketing, it is often used interchangeably with Return on Investment (ROI), though ROAS is a more specific term relating solely to advertising.
    See also:

    Revenue

    Revenue

    Revenue is the money generated by the sale of products and a calculation of quantity price. Revenue is one of the most important metrics for any business and is used to calculate profit, as well as KPIs such as ROAS or ROI.
    See also:

    Return on Ad Spend (ROAS)

    Run Of Network (RON)

    Run of Network (RON) is a placement option that allows advertisers to target ad placements over a number of different sites that all sit within the same network. Less specific than run of site (ROS), this approach allows for ads to appear anywhere in a specified network, not necessarily just one specific website, and prioritizes reach.
    See also:

    Placement, Run of Site (ROS), Reach

    Run Of Site (ROS)

    Run of site (ROS) is a placement option that allows advertisers to target ad placements across an entire website rather than targeting specific sections or pages. It is more specific than RON.
    See also:

    Placement, Run of Network (RON)

    S

    S

    Search Advertising

    Search Advertising is the placement of advertising within search engine results pages (SERPs) - most commonly Google. Advertisers bid on keywords or keyword phrases relevant to their products or services so that their advert will show up on the SERP if a user searches for those words or phrases.
    See also:

    Paid Advertising, Keyword, Pay-per-click (PPC)

    Search Engine Optimization (SEO)

    Search Engine Optimization (SEO) is the process of improving a website through multiple methods to organically rank high on a search engine results page (SERP) for specific keywords or keyword phrases - most commonly Google. Typically, this is achieved by writing high-quality content that utilizes the keywords a business is targeting. Search engines such as Google also favor webpages that include rich content as well as websites that meet certain performance ranking factors.
    See also:

    Organic, Keyword

    Session

    A Session is a group of user interactions with a website that take place within a given time frame. For example, a single session can contain multiple page views, events, social interactions, and ecommerce transactions, including desired conversion actions. Usually tracked by analytics tools such as Google Analytics (GA4) and can be an important metric when evaluating website performance and user behavior.
    See also:

    Conversion, Bounce Rate, Attribution Model

    Single Sign-On (SSO)

    Single Sign-On (SSO) allows users to enter login credentials once to access a range of applications and systems. This is made possible by SSO solutions passing an authentication token onto configured applications. Compare with Two-factor Authentication (2FA).
    See also:

    Two-factor Authentication (2FA)

    Single Source Of Truth

    A Single Source Of Truth is the concept of creating a single coherent database from all your data in one single location accessible by multiple teams or individuals. The concept is important to marketing since a single source of truth gives marketers a more holistic view of how their activites are performing. From a marketing perspective, a single source of truth could contain and combine performance data, advertising data, audience data, location data, or indeed any sort of data that marketers might deem useful to know in order to improve their overall marketing performance.
    See also:

    Data Silo

    Further reading:

    What is a single source of truth (and why should marketers care)?

    Structured Data

    Structured Data refers to any set of organized and formatted data that follows a logical structure and is commonly stored in relational databases or data warehouses. Generally, structured data consists of a table with rows and columns with each row containing a record, and each column containing values that correspond to it. Structured data is extremely useful for data anlysis because it is quantitative, measurable, and easily comparable.
    See also:

    Unstructured Data, Data Warehouse, Relational Database

    Further reading:

    Structured Vs. Unstructured Data: What's The Difference?

    Structured Query Language (SQL)

    Structured Query Language (SQL) is a programming language used to communicate with relational databases. It allows users to perform various operations, such as querying data, inserting, updating, and deleting records, defining database schemas, and managing permissions. SQL provides a standardized syntax and set of commands for interacting with databases, making it a powerful tool for data management and analysis.
    See also:

    Relational Database

    T

    T

    Tab Separated Value (TSV) File

    A Tab Separated Value (TSV) File is a type of flat file that uses tabs to separate values (data) in a clear and simple visual format. They are often used as a way of exchanging tabular data — numbers and text — between tools, programs, or systems.
    See also:

    Flat File, Comma Separated Value (CSV) File

    Table

    Tables are the basic building block of a relational database, where data is stored in columns and rows. A database can contain multiple tables. Tables are essentially sets of data that can later be queried or joined together.
    See also:

    Database, Relational Database

    Two-Factor Authentication (2FA)

    Two-Factor Authentication puts users through multiple security steps before allowing them to access systems or services. Typically, that includes a basic password check at level one before asking for additional information when users reach level two. These are often unique codes generated by tools such as Google Authenticator. Double security gateways can be applied to entire stacks or specific areas.
    See also:

    Single sign-on (SSO)

    U

    U

    Unstructured Data

    Unstructured Data is data that does not have a predefined data model or consistent structure. Examples include books, journals, emails, videos, tweets, and other forms of data that are not easily categorized into structured databases. Unstructured data is typically more complex to analyze and process because it does not fit neatly into tables or relational databases which is why it is mostly stored in data lakes.
    See also:

    Structured Data, Data Lake

    Further reading:

    Structured Vs. Unstructured Data: What's The Difference?

    V

    V

    Video View

    Video View is a metric that measures the number of times a video has been watched by users. A view is classed differently on different platforms depending on how much of the video was watched.
    See also:

    Cost Per View (CPV)

    View-Through Conversion

    View-Through Conversions track users that did not interact with an ad impression but later converted. This is attributed if they return within a 30-day window and convert, whether directly to a website, via organic search, or referral.
    See also:

    Impression, Conversion

    Viewability

    Viewability, or Ad Viewability, is a metric that measures whether or not an ad has the chance to be seen by a real human. It is calculated as a percentage (%) by dividing the number of viewable ads by the number of measured ads and multiplying by 100. For an ad to be viewable, it must meet the minimum standards set by the Media Rating Council, and should not include bot traffic or other forms of ad fraud. For display advertising, the ad must be at least 50% in view for at least one continuous second. For video, the ad must be at least 50% in view for at least 2 continuous seconds.
    See also:

    Metric, Video View, Display Advertising

    Further reading:

    What is Ad Viewability?

    Watch the video:

    Visualization

    A Visualization is a graphical representation of data such as a bar chart, pie chart, heat map, or any number of other formats that visualize data sets to allow users to immediately spot trends and insights. Can be compared to tables, which simply display data in numbers. Visualizations are a type of widget that are used to build dashboards with different dashboards often featuring multiple visualizations.
    See also:

    Dashboards, Widgets

    Further reading:

    What Is Data Visualization: Common Mistakes and Pitfalls

    W

    W

    Walled Garden

    Walled Gardens refer to platforms that are closed systems able to maintain a large degree of control across their eco-system, particularly in regards to data and keeping data within that system. Prominent examples of walled gardens in advertising include Google and Amazon, both of which have created a very convenient 'one-stop-shop' environment for marketers and advertisers that offer access to vast collections of data, inventory, and centralized audience management.
    Further reading:

    What is a Walled Garden in Advertising?

    Widget

    Widgets are elements that make up a dashboard. They can be visualizations, data tables, quickfilters, images, or simple text.
    See also:

    Dashboard, Visualization