Ready, aim…advertise. Okay, maybe that’s not how you usually say it, but you understand.
Before marketers pull the trigger on a new campaign, they need to know which AdTech targeting methods to use:
Behavioural targeting? Demographic targeting? How will we reach our most valuable audience?
The last thing a brand wants to do is fire off an expensive campaign aimed at the wrong people.
So while advertising technology has given companies a multitude of tools to work with, one huge key is knowing how define your target audience and how to get your hands on the data you need to “aim” your campaign correctly.
Of course, there is no “silver bullet” and every campaign will likely use a combination of tactics – even more reason to get a handle on the most popular and most useful AdTech targeting methods.
Targeting users based on context is one of the most basic methods of segmenting audiences for advertising campaigns.
One advantage of contextual targeting is that ads are usually matched more or less to their surroundings providing a more unified, less disruptive user experience.
There are several levels of contextual targeting, each with its specific uses and advantages.
- Site category targeting: At the highest level, ads can be targeted based on the site category of the publisher site or application. The Interactive Advertising Bureau (IAB) has outlined an extensive list of categories and subcategories that most reputable publishers refer to in defining their existence.
- Domain/app-specific targeting: Getting more specific, a campaign may be aimed at one particular domain or application. While the reach of this kind of targeting is relatively limited, it can be useful to target visitors to sites with a close connection to your product.
- Contextual targeting based on page content: Then there is the practice of honing in on specific pages or parts of pages within a site or across several sites. This method may be used for text ads as well as rich-media campaigns.
Most often this contextual content targeting is done through keywords for both text content and/or rich-media content.
- Text (page keywords, title tags or alt text for images)
- Rich-media content (video titles, tags or merely the presence of video or other rich media)
While the most obvious example of keyword contextual targeting is Google’s AdWords, almost all demand-side platforms (DSPs) allow advertisers to insert a list of keywords and target page content or parts of pages based on them.
Time/External condition-based targeting
For products that are meant to be consumed, have specific uses or for promotions, time-based targeting or targeting based on external conditions (weather, big sporting events, etc.) may be most effective.
- A beer company might run a promotional campaign on the weekend to target football fans.
- A car repair service might choose to serve an ad immediately after a bad storm.
Programmatic media-buying and -selling has enabled time- and event-based targeting in ways that simply weren’t possible before. Marketers would be stupid not to fully utilize these possibilities.
The methods mentioned above carry significant advantages, but when it comes to reaching customers on a more personal level, targeting them based on past behavior (and more and more by predicted behaviour) is the real pie in the sky for marketers.
No other method provides better insights into customer interests, likes and dislikes, personality and other things that may or may not make them valuable to your brand.
It also is key for judging customer intent – e.g. data pointing to when a person may be ready to buy a plane ticket, book a hotel, etc.
Of course, behavioural targeting also requires the most data (and most advanced data management tools) to execute properly.
For sites and apps with high traffic volume, behavioural targeting based exclusively on 1st-party data may be possible – and very valuable – especially for content personalization.
However, companies that have less traffic themselves will need to rely more on segments from third-party data vendors in order to achieve the desired reach for their campaigns.
When brands want to use behavioral data that is not their own for targeting, they will access it (and purchase it) through a DSP in bundles based on such categories as:
- Interests → Sports
- Intent → Hotel booking
The vendor who makes this information available will base these groupings on a variety of behavioural data (see points 1-3 below). The one main drawback to this method is, of course, that the advertiser does not always have insights into what data was used to place users into the audience segment, meaning that person may or may not be a valuable target for them.
First-party data targeting
While we’re on the subject of behavioural targeting, let’s explore segment-based targeting. In general, this term can be applied to the practice of targeting users based on a single or group of conditions (often behavioural data, but also a combination of, for example:behavioral and location information.
Targeting based on first-party data can include segments exported directly from your web analytics tool (Google Analytics, Piwik PRO or WebTrends) and activated through a data management platform (DMP) into a demand-side platform (DSP) or using a hybrid DMP/DSP.
First-party segments can be very specific depending on the amount of data that a brand has at its fingertips.
Interaction with site content: One of the main ways to measure behaviour is by recording a site or app visitor’s interaction with content. This can range from the very general:
- product views
- session duration
to the more specific:
- in-site search queries
- scrolling (for written or visual content)
- video view duration
Even more concrete information can be gathered based on on-site or in-app conversions.
These can be both macro conversions (downloading a white paper, signing up for a newsletter, or – in ecommerce – making a purchase or using a coupon) or micro conversions (clicking a “related posts” button on a blog, viewing contact information on a company’s site, etc.)
2) Ecommerce behaviour: interaction with sites or apps that sell products or services online offer an even better gauge of customer interest and intent. In the case of ecommerce sites selling products, data can be collected regarding:
- items viewed (pageviews)
- items added to shopping cart
- average order value
3) Interaction with ad content: clicking through (or not clicking through) display ads, search ads or in-app ads, viewing video ads longer than a certain time.
This information can be sent back to a brand via a conversion pixel and aggregated with other behavioral data in a data management platform.
4) CRM data: In addition to the first-party data that a brand may have collected online, there may also be useful information from CRM platforms that can be used for targeting.
For example, a customer that has consistently made large purchases in-store would be a high-value target for an online promotional campaign.
Because CRM data can potentially be personally-identifiable, it will need to be anonymized and activated – a process that can be accomplished using tools such as LiveRamp.
In many ways demographic targeting (based on location, gender or language) may seem like a less glamorous way of reaching valuable audiences, but it is, in fact, no less valuable than other methods.
Combined with behavioural data, for instance, campaigns aimed at certain demographic groups can be very powerful indeed. And without demographic information, an ad can be totally worthless.
After all how effective is a German-language white paper for someone living in Tennessee?
Gender/Age group targeting
Many products and services are made specifically for either men or women, older people or millennials. So age and gender are pretty important when it comes to targeting.
How is it done?
Well, in the case of desktop browsers, it may be hard to pick out age or gender – although this may be inferred based other information (behavioural data like ad conversions, product views, etc.)
However, for advertisers using specific platforms like Facebook or Pinterest, things are much simpler since users provide this data upon registration.
Some apps/platforms do not ask gender – Snapchat is one example. However, it may be inferred by behavioural data and confirmed by third-party services.
Language is a fundamental element of targeted ads (see the example at the beginning of this section.) For desktop users, this information can be drawn from browser settings, while targeting mobile users can be done through an app’s language setting.
While talking about how much you make might be taboo in social circles, it is definitely a valuable criteria for targeting. It’s no use trying to selling a Porsche to someone who can’t afford it!
Data about income (usually provided in ranges) can be derived from various sources, with greater or lesser accuracy. Income can be estimated based on location (city/country/part of a city) thought this method may not be very precise. More accurate data can be provided by ISP/cable TV companies who know what package of services people have chosen.
Other demographic segments
Targeting can also be carried out based on other demographic conditions – in as much as consumers provide this information. Ethnicity and life status (married/single/married with children) can both be valuable for advertisers, although the former has come under fire recently as being unfair.
You’ve heard the mantra – “location, location, location.”
It might be old, but it’s never been more true than today.
Geolocation is one of the most dynamic targeting conditions (since it can change at random, and often does in today’s world of mobile internet usage) and one of the most valuable for local businesses.
Targeting by geolocation can happen in a few ways, including:
- IP-based geolocation: an internet user’s IP address will give a general idea of his or her location and can be used to great effect for country, city or regional targeting
- Hyperlocation (with mobile apps): for even more precise geolocation targeting from mobile apps based on GPS. When users enable location services (as many apps request or require) it is possible to target the app with location-specific ads.
Hyperlocation often employs geofencing to hone in on very specific locations (down to
a few hundred meters in some cases).
Incidentally, hyperlocation – and geolocation targeting in general – is another form of contextual targeting. Ads for certain products can be shown based on known data about the surroundings (average household income, for instance).
Along with geolocation, device targeting can be especially useful for pushing mobile ad campaigns. This has obvious implications for serving local ads, contextual ads or even promotional coupons for stores/restaurants in the immediate vicinity of the mobile user.
Akin to device targeting is the method of targeting internet users by connection type i.e. whether a site visitor is on Wi-Fi or using a GSM data connection.
This may not be so important when it comes to finding a high-value customer, but it could affect what type of ad creative is served.
Facebook, for example, even offers advertisers multiple options when it comes to mobile users using their phone’s GSM connection – 2G, 3G or 4G.
So if a user is on a slow mobile connection, he or she might not be the best target for a media-rich ad.
Ad blocker targeting
Yes, it’s true. One of the newest, and perhaps most ironic, methods of pushing targeted ad content is by focusing on users of ad blockers.
There are several reasons why this method can be very valuable for brands:
- Users of ad blockers automatically “identify themselves” by have ad blocking software installed. Since publishers (think ad giants Google and Facebook) are keen on detecting and circumventing ad blockers, the very fact that someone has an ad blocker installed makes them an easy target.
- A majority of those using ad blockers tend to be tech-smart and in general younger. Again, this automatically divides them into a segment and makes it easy to target them.
- Because using an ad blocker indicates a user wants to avoid intrusive ads, publishers can leverage this fact to offer high-end ads to publishers with the budgets to create top-quality ad content.
And since many sites (like Forbes) have turned to “bargaining” with ad block users (access to content in exchange for an “ad-light” experience) they can then raise the prices on those “light” ads to garner increased profit.
Last, but not least on our list of AdTech targeting methods is retargeting. In other words, serving ad content to visitors that have interacted in some way with our brand before.
The possibilities for retargeting are quite broad and could be set at one interaction or a series, depending on how a brand wants to judge the value of the potential customer.
It’s main focus is customer retention and conversion of potential customers that have already engaged with brands – which is why it proves effective for large and small companies alike especially when used in conjunction with other campaigns aimed at acquiring new leads.
Retargeting can be accomplished by setting a cookie when a visitor completes a certain action on your site or engages with your content in some way.
Some platforms such as Facebook offer retargeting ad opportunities based on lists of email addresses which have been collected either from newsletter subscriptions, ecommerce account setups or even in person and later entered into CRM databases.
And there you have it – an in-depth breakdown of the most common AdTech targeting methods. Now you don’t have any more excuse to “miss the mark” with your next ad campaign.
Ad Tech / MarTech Guide
Download our Guide & get an in-depth look at the process of developing successful advertising and marketing solutions.