Audience targeting 101: How to collect and use audience data
As you move around the Internet, the sites you visit drop cookies, or lines of code, on to your browser, which start to develop a profile of you. You may be labelled as a “sports enthusiast,” for example, after visiting 15 pages of related content within a week.
When advertisers combine this behavioural information with geo-targeting, dayparting, device and other data sets, they can optimize their campaigns and reach audiences most likely to engage with their ads.
There are three main kinds of audience targeting data available, defined by how and where the information is collected. Each have their pros and cons, and often advertisers need to use a combination to achieve their goals. Here’s a brief overview of the options, and why you would, or would not, use them for your campaigns.
Audience targeting: First party data
This is data you collect directly, including site or mobile app analytics, CRM or e-mail databases. This may also be first party data that belongs to publishers you are working with.
Pros: It is considered the most accurate and relevant data because it is collected directly. Because you own the data, you outline the taxonomy and control how it should be segmented, making it unique. It can also be very cost-effective – there aren’t additional data usage costs, although there may be related ad tech costs (for example, if you have your own data management platform, or DMP).
Cons: Since your site visitor or customer base is limited, your data set will be, too. You may not have the scale you need to run large campaigns, and there will be a limit to the depth of your data – a clothing company may not have data on pet owners, for example.
Audience targeting: Second party data
This is another company’s first party data that has been shared with you for your use.
Pros: This can add more depth and scale to your data. If we continue with the clothing company example, let’s say you were launching a new line of pet fur-repellant pants – you could partner with a pet supply company to reach a new audience, or match your data sets to see which of your visitors have pets. Since you are getting data directly from the source, there is still a high level of quality to this data set.
Cons: Since this includes the sharing of data, you may run into issues with privacy (such as company policies or Canadian law), data ownership and usage costs. Depending on how the data is shared, you may also run into ad tech integration issues between platforms.
Audience targeting: Third party data
This is aggregated audience targeting data purchased from a collector or vendor, and it may include site visitation behaviour (inferred or observed data) or registration information (declared data).
Pros: The high availability of data providers, which can achieve tremendous scale and depth of data that you may not have access to otherwise.
Cons: There is a perception that this audience targeting data is of lower quality. Because it belongs to third party vendors, the definitions of these segments may not be fully transparent. A C-Suite segment may include people who have indicated on a site registration form that they are executives, but it may also include people who consume a lot of business content of interest to executives. It’s also not unique – your competitors may have access to and be using this C-Suite segment, hitting the same audience with a similar message. It can be very costly. In addition to any ad tech costs, you will also be paying an incremental CPM to use this data, and those costs vary from pennies to a few dollars depending on the segment.
When should you use these different kinds of data?
That depends on your goals and what is available to you. If you don’t have any first party data of your own, you’ll have to rely on a publisher’s first party data, or second or third party data to fill those gaps.
Let’s say your key goal is to increase revenue from your existing clients. Since these are your customers, and you know their current value, this is a first party data set. You can segment it to target customers who may be interested in complementary products, such as notebook buyers who want pens or sticky notes.
But it may not be enough to focus on existing customers – you may also need to acquire new ones. These people may not have visited your site in the past or even heard of your product, so how do you reach them? We’ll use another example of a new line of all-natural baby products. Your audience targeting definitely includes new parents, but it could also include those who are shopping for baby gifts, or who are interested in eco-friendly products.
In order to reach these audiences, you may need to tap into second or third party data to find people who visit parenting sites, who have bought similar products in the past, or who are interested in “green living.”
Want to start collecting or using data?
Electronic data collection is carefully regulated in Canada, and subject to the Personal Information Protection and Electronic Documents Act (PIPEDA). Any data collecting you do must comply with these rules, which include notifying users about what data you are collecting, why, and how you will use it.
If you are planning to use audience targeting data in your advertising campaigns (whether it’s your own or someone else’s), consider working with a partner that participates in AdChoices, a self-regulatory program for online interest-based advertising that provides users with greater control over their data collection and the types of ads they see. It’ll make everyone a little more comfortable.
Marce Bylinska is a digital strategist at The Globe and Mail. She can be reached @mmmarce.