Digital advertising guide

To create a data-driven advertising strategy, focus on audience, context, and measurement. Continue reading to learn the who, where, and how of digital advertising.

The opportunities of a data-driven digital advertising strategy

This may come as a surprise, but many advertisers and marketers struggle with how to make the most of data-driven marketing and advertising. According to a recent Association of National Advertisers survey of 93 top marketers, the most important key performance indicators (KPIs) aren’t the actual ones being used most often.

Audience—Who you should connect with. This could be existing customers or prospects.

Context—Where should you engage with your audience.

Measurement—Understand if and how they responded to your messaging.

In fact, many of the most commonly leveraged KPIs are considered to be "head fakes"—KPIs that may be endorsed broadly by the industry but aren’t that useful in driving campaign effectiveness or business outcomes. The key is to separate the data from the noise.

To gather the right data and then apply it properly as you build out your campaigns, focus your energies on three core data pillars: audience, context, and measurement.

With the right tools to effectively tap into each of these data pillars, not only is it less daunting to build and use a data-driven strategy, your campaigns are more successful. Using data correctly to build highly personalized campaigns does save money and increase ROI, delivering five to eight times the ROI on campaign spend. Let’s see how:


When you are able to target the right audience, you are able to reach your best and most valuable customers, prospects, and visitors with relevant messaging, leading to higher engagement levels, more sales, and less wasted advertising.

target audience
impact with context


Context lets you make the biggest impact with your campaigns as possible by finding the best places where your creative assets and messaging are the most effective.


The right measurement tools give you that all-important 360-degree campaign view, so you know what’s working, why it’s working, and what you need to optimize to make it work better.

optimize with measurement

The 3 data pillars for effective digital advertising


With a strong audience strategy, marketers and advertisers gain a better understanding of who their ideal customers and prospects are. With that information, they can tailor their campaigns—including the message, creative assets, and media plans—to make the largest positive business outcome.

But not all audience strategies are created equal. An effective data-driven audience strategy uses consumer data with the highest potential for impact and influence. While there are many ways to segment, analyze, combine, and implement customer data to plan and build your audience, the ultimate goal is to drive smarter advertising investments using campaigns that increase lead generation, sales, or brand awareness (or a combination of all three).

Effective audience planning relies on three segmentation practices that provide unique audience insights and attributes. In other words, providing value without sacrificing scale.


Purchase-based targeting

Create a 360-degree view of customer and audience segments by matching online shopping to offline purchase behaviors.


Audience modeling

Create new audience segments by identifying prospects with similar traits and characteristics to your ideal customer profile.


Audience augmentation

Enhance first-party data (CRM, ERP, and supply chain) with third-party data to reach new audiences, discover new channels, and increase retention.


Context, or contextual intelligence, is the ability to analyze content to understand how people interact with it. Digital marketers and advertisers use context to:


Avoid inappropriate content, nonviewable inventory, or invalid traffic

The risks of negative exposure are real. Customers construe brand alignment as a deliberate indication of your brand’s values.

Exposure within, or adjacency to, clearly negative content can do a lot of harm, and is certainly why many brands already apply brand safety filters. But marketers must be mindful of the environments that work—or don’t—for their brands’ particular needs.

This is the idea behind brand suitability, which takes brand safety to another level by customizing the strategy to your unique brand. Consider the difference in the types of content that are suitable to align with an alcoholic beverage versus a type of diaper. Content with more mature themes may be appropriate for the alcohol brand whereas that same content would be off limits for the diaper brand.


Actively target environments with relevant context

Context can also be used to find environments that align with the customer mindsets that are most compatible with your brand. This is the premise of positive context, which opens up new opportunities for brands to deliver relevant, timely experiences.

Although there are many ways to think about what the “right” context means, here’s a framework to help you decide what’s right for your campaign. Make sure that the content:

1. Aligns with your identified customers’ needs

For example, a family-oriented, fast-food chain serving children’s meals should actively align its advertising with digital content featuring tasty, easy, and fun kids’ meals, nutrition tips, recipes, and related subject matter.

2. Aligns with an identified persona/lifestyle choices

For example, a luggage company may want to align with people who are interested in adventure, overseas travel, backpacking, foreign culture, and international cuisine.

3. Aligns with equity-building content that reinforces your broader brand objectives

For example, if a major celebrity endorses you, you could align your advertising with content about that person.


of marketers have integrated their digital strategies into their overall brand strategy
of advertisers say it’s challenging to assess cross-channel performance
of agencies say it’s challenging to assess cross-channel performance
of media companies say it’s challenging to assess cross-channel performance

Getting Media Right: Creating Breakthrough Marketing in a Connected World (PDF)

Measuring your digital advertising campaigns is no longer just about measuring impressions, clicks, conversions, or even sales. It’s about understanding cross-device and cross-platform performance to better optimize all parts of the campaign.

The following are some of the most important data points that digital advertisers should be tracking. Keep in mind that the true value of campaign performance isn’t determined by looking at individual metrics in silos, but how they all impact each other.

Standard baseline metrics

  • Impressions
  • Clicks
  • Conversions

Cross-device and platform metrics

  • Desktop
  • Mobile
  • In-app
  • Video
  • Connected TV/over-the-top (OTT)

Advanced metrics

Invalid traffic

Did the ad generate bot traffic or human traffic?


Was the ad viewed, and how long was it viewed?

Brand safety

Can you monitor and track brand safety efforts across platforms and ad formats?

People-based reach and frequency

Did your message reach the most relevant audience (the segment of the population who would care about the message), and how many times did that audience see your message?


Did you reach the right people, and are they engaging with your ads?


Did your creative deliver interactions and capture the audience’s interest?

Return on investment (ROI)

Did your digital campaign impact offline sales lift?

Deploying the 3 data pillars: A checklist

The next step is to learn how to apply audience, context, and measurement to your digital campaign. Here’s a three-part, step-by-step checklist.

Step 1: Define your audience

Here’s an activation checklist to write a winning audience plan:

1. Collect

Gather all available first-party data and analyze for trends. Determine if there are patterns in your data that reveal who is purchasing, what channels they’re using, what types of content resonates, and other metrics.

2. Augment

Supplement your first-party CRM and ERP data with high-quality, third-party data to increase scale. Onboarding solutions with high match rates across specific ID types can help with this.

3. Create

Build behavioral profiles of your audience. Combine your audience’s online and offline behavior to develop a clearer picture of who your ideal customers and prospects are and how best to reach them.

4. Personalize

Tailor your message and creative based on what you discover about your audience and where they can be reached. Balance the positioning of your brand and advertising/marketing campaigns with the needs and wants of your target audience.

5. Measure

Evaluate all relevant aspects of your campaign to determine if the messaging and creative resonates. You may need to optimize the channel, creative, platform, or even the audience/customer segment to get the desired results for your campaign. For more information, see Step 3.

Getting better outcomes: 5 questions to ask your data provider to get high-quality data

  • 1. How do you measure the performance of your audiences?
  • 2. What signals or data points are used to build audiences?
  • 3. How do you validate the accuracy of your data?
  • 4. How do you ensure data privacy?
  • 5. What is your approach to audience modeling?

Learn how to put audience data at the center of your digital strategy.

Step 2: Connect with your audience in the right environment

Here’s an activation checklist to improve reach, maintain relevance, and reach the right audience:

1. Protect

Keep your ads from appearing next to negative, inflammatory, or inappropriate online content with an always-on brand safety solution.

2. Build

Build custom contextual segments using keywords and phrases that align with your brand or campaign for greater control over where your advertising is appearing online.

3. Partner

Use a context partner to help automate the building and updating of custom keyword segments. Doing so will allow you to capitalize on popular trends as they unfold and appear next to new, brand-safe content as it’s published.

4. Optimize

Optimize your campaign and get creative with your context strategy by expanding beyond the obvious environments and content that will serve your campaign. For example, a travel company running a summer campaign may want to consider targeting content about swimsuits and outdoor gear in addition to travel inspiration and tips.

Getting better outcomes: 7 questions to ask your context partner/provider

  • 1. What is the value of using both people-based and contextual audiences, and how do I use them interchangeably?
  • 2. How effective is contextual targeting in finding actual buyers?
  • 3. How quickly can you identify trending content, and at what scale?
  • 4. How quickly can you make custom segments available for use?
  • 5. How do you guarantee that my message will appear in the right environments?
  • 6. Do you offer full-page or page-level analysis of keywords?
  • 7. How do your contextual segments perform?

How you can use contextual intelligence for maximum brand protection, and to expand your reach.

Step #3: Measure, manage, and attribute your campaign

Use this checklist to create a robust measurement plan with your measurement partner:

Align strategies

Align media measurement strategies with business objectives to determine the right metrics to track to determine campaign effectiveness. For example, if your business objective is to increase product awareness by 10% and gain 5% of incremental audiences in six months, then you should track the following metrics:

  1. Sales lift
  2. Online-to-offline sales impact
  3. Viewability and attention

Identify criteria

Determine what criteria will be needed to provide details on how your campaign and ads are performing. Consider these baseline metrics that are commonly used for display, video, and mobile ads.

Metric Video/display Video Mobile
Invalid traffic percentage
Percentage of unfiltered impressions that were delivered to an invalid endpoint. This includes general IVT (spiders, excessive activity, and/or data center traffic categories) and sophisticated IVT (invalid proxy, automated browser, incongruous browser, invalid source, hidden ad, and/or session hijacked traffic categories).
In-view percentage
Percentage of impressions where at least 50% of an ad was in-view for at least one continuous second. If the ad is as large or larger in area than 970 x 250 (for example, 300 x 1050 or 970 x 418), then it only needs to have 30% of its area in-view.
In-view time
Average time in seconds that the ad was visible to users who met the requirement for a two-second, in-view impression.
Audible and visible on complete (AVOC) percentage
Percentage of valid impressions where the ad was visible and audible at the point of completion.
Audible on…
Percentage of measurable impressions where the ad was audible at a given quartile (start, first, second, third, end).
Video score
Benchmark for the sight, sound, and motion aspects of video. The score is based on the average percentage of the video that was audible and/or visible, amplified by the screen’s real estate. Ranging from 0 to 100, it’s intended to assess the quality of different video ad exposures on desktop and mobile devices.
Visible on…
Percentage of measurable impressions where at least 50% of the ad was visible on a given quartile (Start, first, second, third, end).
Universal touch rate
Percentage of impressions where an ad has been touched on a mobile device.

Measure effectiveness

Track the effectiveness of your campaigns by using three key performance indicators (KPIs):

Media quality. Are your ads reaching the right people, and are they paying attention?
  • Valid and view ability rate
  • Valid impressions
  • Percentage of ad in-view
  • Attention quality
  • Percentage of video played in-view
Brand safety. Are your ads being placed in the most relevant, suitable environments?
  • Safe and unsafe percentage
  • Contextual reporting
  • Content categorization analysis
Performance and ROI. Did your ads perform and meet revenue goals?
  • Sales lift
  • Online-to-offline sales impact

Getting better outcomes: 5 questions to ask your measurement partner/provider

  • 1. Did I reach real people with my impressions?
  • 2. Did I reach people relevant to my campaign objectives, across the right channels?
  • 3. Did my campaign drive the outcomes I was expecting?
  • 4. How strong are your verification capabilities across brand safety, content quality, media quality, and fraud protection?
  • 5. What is your methodology?


Addressable TV A method of segmenting TV audiences and targeting different ads or ad pods at the household or zone level (groups of homes) through cable, satellite, and set-top boxes into linear or video on demand (VOD).
Ad fraud Sometimes referred to as invalid traffic. Fraudulently representing online advertisement impressions, clicks, conversions, or data events in order to generate revenue.
Ad verification A technology service offered by vendors to ensure that ads appear on intended sites, reach the targeted audience, and are executed in a manner consistent with the terms of a campaign.
Ads.txt An Interactive Advertising Bureau (IAB)-approved text file that aims to prevent unauthorized inventory sales and allows buyers to check the validity of the inventory they purchase. Publishers drop a text file on their web servers that lists all the companies authorized to sell the publishers’ inventory. Programmatic platforms integrate ads.txt files to confirm which publishers’ inventory they’re authorized to sell.
Advertising technology (AdTech) Enables advertisers and ad agencies to create, run, measure, and manage online advertising campaigns across a number of websites or apps. It allows publishers (websites and apps) to sell their available ad space, also known as inventory, to a large number of advertisers. It represents the use of third-party data/external platforms to reach customers.
Audience profile A list of attributes you know about your target audience, such as demographic, geographic, interests, intent to purchase, and purchase history.
Audience reach The percentage of audience targets that were reached through digital programs during a given time period.
Audience segmentation The process of dividing your target audience by defined attributes, such as demographics, purchase intent, interests, geography, past purchases, and other behaviors that indicate a likelihood for customer conversion.


Behavioral targeting Ads shown based on a consumer’s online browsing behavior.
Bot A nonhuman form of traffic.
Brand safety The set of measures that aim to protect a brand’s image from the negative or harmful influence of inappropriate or questionable content on a publisher’s site where the ad impression is served.
Buy-through rate (BTR) A factual measure of sales activity for a defined audience against a national baseline.


Connected TV A device that can connect to a TV or a smart TV that facilitates the delivery of streaming video content. Examples include Xbox, PlayStation, Roku, Amazon Fire TV, Apple TV, and Chromecast.
Contextual intelligence Information and insights available through analyzing content and how people interact with it. The content that individuals are interacting with at any given moment offers a vast amount of insights about those people, and how a business might engage with them.
Conversion The point at which a recipient of a marketing message performs a desired action.
Cookie A small piece of data sent from a website to a web browser to anonymously capture a user’s browsing behaviors and preferences.
Cost per acquisition (CPA) A method of advertising whereby the advertiser only pays when an ad delivers a conversion.
Cost per click (CPC) The amount of money spent on online media divided by the number of clicks.
Cost per lead (CPL) A pricing structure where an advertiser pays for an explicit sign-up from a consumer interested in the advertiser’s offer.
Cost per mille (CPM) A commonly used measurement for advertising that is purchased on the basis of showing the ad to 1,000 viewers.
Cross-channel marketing The ability to identify and engage a target audience with consistent and relevant message across offline, online, mobile, social, and search interactions.
Cross-device matching The ability to connect multiple identities across disparate marketing channels and devices to one customer, enabling marketers to tie their interactions to an actionable customer profile. This empowers marketers to orchestrate a relevant, personalized experience for each individual while minimizing media waste.
Custom audiences Those built from relevant interests, engagements, and domains to target high-value prospects.
Customer data platform (CDP) Software that aggregates and organizes customer data across a variety of touchpoints and is used by other software systems and marketing efforts.
Customer relationship management (CRM) A model or technology for managing a company’s interactions with current and future customers


Data activation The ability to take data intelligence and push it seamlessly to the execution layer to influence ad targeting, site optimization, lookalike modeling, or creative customization.
Data as a service (DaaS) A cost-efficient, agile model for marketers to leverage data in unlimited quantities. DaaS enables users to consume data across a variety of systems and processes to make customer engagement more relevant and impactful.
Data-driven marketing (DDM) Using data from any source to make better decisions on who, where, when, and how to market.
Data management platform (DMP) A system that allows marketers to centralize, manage, and analyze their first-party data for use in online marketing campaigns. Some DMPs allow for the addition of second- and third-party data, as well as connections with other media platforms to push data out across a variety of channels and devices.
Data onboarding The process of transferring data gathered offline, such as in-store transactions or CRM data into the online world to use in marketing and advertising campaigns.
Declared data A type of first-party data that has been willingly and explicitly shared by individual consumers. It provides context to consumer behavior, often about a prospect’s motivations, intentions, interests, and preferences.
Demand-side platform (DSP) Technology that enables campaign management, allowing buyers to manage multiple ad and data exchange accounts through one interface.
Deterministic A term referring to an “observed” association versus one that is “inferred.” Deterministic data often starts with stable IDs, such as an email address, a user ID, or a phone number.


No entries at this time.  


Frequency The average number of times an individual gets served an online advertising message during a defined period of time.


General invalid traffic (GIVT) Nonhuman traffic generated by known industry crawlers, such as search engine crawlers.


No entries at this time.  


Identity The body of information about an individual, organization, or electronic device that exists online.
Identity resolution The process of gathering and matching identifiers across devices and touchpoints to enable marketers to deliver personalized experiences.
Impression The presentation of an online ad to a consumer.
Inferred (See “Probabilistic.”)
In-flight A term meaning that a campaign is running.
In-market A term meaning that a customer is actively looking to purchase.
Interactive Advertising Bureau (IAB) A business organization that develops industry standards, conducts research, and provides legal support for the online advertising industry.
Invalid traffic (IVT) Any clicks or impressions generated by bots or other nonhuman traffic. While not necessarily fraudulent, invalid traffic may artificially inflate an advertiser’s costs or a publisher’s earnings.


No entries at this time.  


No entries at this time.  


Lead generation The initiation of consumer interest or inquiry into a business’s products or services.
Linear TV A traditional system in which a viewer watches a scheduled TV program at the time it’s broadcast and on its original channel. It can also be recorded via DVR and watched later. (Essentially, this is TV as it was known before the digital age.)
Lookalike audiences A group of prospects who have similar behaviors to an existing seed audience. By expanding their campaigns to find similar audiences, marketers can greater reach and efficiency.


Malware Any software intentionally designed to cause damage to a computer, server, client, or computer network.
Marketing technology (MarTech) The software that allows marketers to create, run, and manage online marketing campaigns and conduct onsite marketing. Examples include email marketing, social media management, A/B testing, personalization, user-feedback surveys, and web analytics. It represents the use of first-party data/consent/personally identifiable information (PII) to inform customer journeys.
Match rate The percentage of users that a demand-side platform (DSP) recognizes within a given audience segment.
Measurement Tracking and understanding the impact of a marketing campaign via data analytics.
Media Rating Council (MRC) An American independent organization whose mission is to ensure valid, reliable, and effective audience measurement services.
Mobile advertising ID (MAID) Also known as device ID. A resettable advertising identification signal generated by a mobile operating system at the device level for advertising usage in mobile apps, for example, Apple IDFA and Android ADID.
Modeling Using third-party data to build statistical models based on a high-performing set of attributes from seed data. This technique increases reach for advertisers and typically uses demographic and behavioral data.
Multi-touch attribution (MTA) A method used to prove the value of a B2B marketing strategy in the form of down-funnel metrics that matter to a business, such as the number of opportunities and revenue.


Native advertising The use of paid ads that match the look, feel, and function of the media format in which they appear. Native ads are often found in social media feeds, or as recommended content on a web page. Unlike display ads or banner ads, native ads don’t really look like ads.


Offline data Data sources derived from consumer behavior in the physical realm, such as in-store purchases.
Online data Data sources derived from online behaviors.


Panel-based data A data validation method wherein a panel consisting of three data scientists validates the accuracy of an online profile.
People-based marketing Gathering customer data from online and offline sources to more accurately recognize and reach customers on any device.
Personally identifiable Information (PII) Personal information, such as first name, last name, mailing address, and email address, which can be used to identify a consumer.
Pixel Also called a marketing or tracking pixel. A small piece of code that is placed on a web page to anonymously capture a person’s internet behavior and interest data.
Postal ID ID created using a customer’s physical mailing address.
Post-bid Occurs after an ad has run. Rather than proactively preventing an ad from running in unsafe environments, it reports on where ads ran the day before. It allows advertisers to determine if all placements were appropriate and whether they want to make any changes to rules for subsequent ad runs.
Pre-bid Ensures that the page on which an ad will appear is appropriate. As an advertiser enters a bid for a placement, a third-party scans the page against a predetermined set of rules.
Predictive analytics The use of data, statistical algorithms, and machine-learning methods to identify the likelihood of potential outcomes based on historic data.
Private data marketplace (PDM) Allows data providers to create custom, private segments that can be made available for individual buyers.
Private marketplace (PMP) An invite-only marketplace where high-caliber publishers offer their ad inventories to a selected group of advertisers.
Probabilistic Data that includes linkages between two IDs that are inferred based on event patterns and ID characteristics.
Programmatic A type of advertising that helps automate and optimize media buying decision-making by targeting particular audiences and demographics.
Purchased-based targeting (PBT) Targeting audiences based on their actual purchase histories under the assumption that past buying behavior is an indicator of future purchases.


No entries at this time.  


Real-time bidding (RTB) Technology that allows online advertising to be purchased and served on the fly. Instead of reserving prepaid advertising space, advertisers bid on each ad impression as it’s served. The impression goes to the highest bidder, and their ads are served on the page.
Recency How recent a profile was tagged with an attribute.
Response rate Responses received as a percentage of promotion pieces mailed or contacts made.
Retargeting A form of online targeting advertising served to people who have already visited a website or who are contacts in a company’s database. For example, leads or customers.
Retention Ability to keep customers satisfied to increase the likelihood that they’ll make another purchase (upsell or renew).
Retention campaign A marketing campaign targeted toward existing customers to keep them engaged with and loyal to a brand, product, or company and avoid attrition.
Return on ad spend (ROAS) Also called gross return on marketing. A metric calculated by dividing the revenue generated from an ad campaign by the cost of that campaign.
Return on investment (ROI) The return on the money invested in a marketing campaign, such as sales lift or a direct sale.
Revenue per mille (RPM) Measures the revenue from 1,000 ad impressions. While CPM is normally measured by advertisers, RPM is monitored by publishers.


Segment A grouping of consumer profiles or attributes. Segment categories range anywhere from very broad with a single attribute, such as “male” or “female,” to very specific, such as “millennial female business traveler,” with multiple attributes.
Site-side optimization (SSO) The ability to customize the look of a website based on the attributes of a visitor’s profile.
Sophisticated invalid traffic (SIVT) Typically a result of fraudulent behavior and more difficult to detect. These fraudsters are making an extra effort to mask their behavior as legitimate so it requires advanced analytics; multipoint corroboration/coordination; and significant human intervention to detect, identify, and analyze.
Supply-side platform (SSP) Software and services that enable audience management for web or mobile publishers. This allows the publishers to manage their advertising space inventory, fill it with ads, and receive revenue.
Syndicated audiences Audiences built using the best available online search and browse signals, plus offline past purchase and demographic data, so a company can feel confident in reaching the right audience.


Taxonomy The process of organizing and classifying data into a hierarchy so that it’s divided in a way that easily understandable. For example: “Past purchase->electronics->television->HDTV.”


Unique visitor (UV) A distinct user who visits a website in a given time period regardless of how many times they visit.


View When a user is exposed to an ad.
Viewability An online advertising metric that aims to track only the impressions that were seen by users on a specific page.


No entries at this time.