Good user data tells a story. At the heart of any effective programmatic advertising solution lies detailed and reliable data on a consumer’s demographics, their preferences, and their interests. High quality personal data helps profile consumers, predict their response to an advert and optimise how they are targeted. However, effective personal data is not static. It flows throughout the supply chain, across publishers, service providers, platforms, and advertisers, all the while describing consumer behaviour, reactions to different advertising campaigns and, ultimately, the consumption journey.
This is the ultimate definition of “good” data; it tells a story. It helps the advertiser evaluate a consumer’s journey through their marketing funnel, from brand awareness through to interest, evaluation, commitment and, ideally, a purchase. Capturing reliable data throughout that journey is the holy grail of marketing. Everything you learn about the existing advertising technology stack slots into some part of that journey. Because the more advertiser knows about a consumer, the consumer’s fit to a particular target audience, their propensity to spend, and their position within the marketing funnel, then the higher the probability that every incremental dollar of advertising spend will result in an eventual purchase. But how can that data be collected? When is that data volunteered with consent, and when is it appropriated without consent?
- Three ways to collect personal data
- Cookies
- Tying cookies together to build user profiles
- Cookies and privacy
Three ways to collect personal data
Broadly, there are three ways to collect personal data; data is sometimes volunteered (typically when creating an account, registering for content or services, or making a purchase), data is observed, and data is inferred. Volunteered data is often referred to as first party data. Data that is observed and inferred can be either first party or third party. It depends on the means with which data is observed or inferred. This is where the conversation around cookies starts.
Cookies
A cookie is file that drops into your device storage (for instance the hard drive on your PC) when you visit a website. Some of these cookies are useful; a cookie that tells the website operator who you are, so you do not have to logon every time you visit the website. But not all cookies are designed to help you. Some cookies are designed to tell the website operator, or one of their associates, information about what you are looking at, and what you might be interested in. There are two types of cookies, first party and third party. First party cookies are installed on your browser by the website domain you are visiting while third party cookies are served by third party domains. Often, cookies allocate you and ID which helps the issuer of the cookie to track your behaviour over time.
Tying cookies together to build user profiles
The main advertising technology challenge has been to tie different cookie IDs into one customer profile, which, since cookie IDs come often originate from different third-party domains and across different ad-tech platforms, can be challenging. Ideally, advertising technology platforms combine user browsing histories across many domains from multiple devices, creating more accurate user profiles. Those profiles are typically packaged into target audience segments by data management providers (DMPs). Brands looking to develop or grow their audience purchase target audience data from DMPs.
Advertisers seek to minimize advertising spend and maximize their return on investment. This is best done by developing and then targeting, and re-targeting, audiences, with as much accuracy as possible. The more accurate, and stable, the user profile, the higher the return on advertising spend and the more valuable the data. The more effective the stitching together of different cookie IDs, and the richer the associated data, then the more effective the advertising.
Cookies and privacy
Digital advertising has a purpose (develop audiences and convert users to consumers) and advertisers need to track whether their advertising is effective. Cookies help track user actions, and responsiveness to advertising campaigns, while also ensuring individuals do not get annoyed by excessive impressions from the same advertiser (controlled retargeting). While personalised content and targeted advertising can be beneficial, the degree to which personal data is shared across multiple domains, without permission, is alarming. According to Clearcode, a leading programmatic advertising consultant, 90% of an individual’s user history is typically known by more than 90 companies, and up to 600 companies know more than 50% of that user’s history. Cookies present a huge user privacy challenge. Yet removing cookies creates a new set of problems; a lower quality consumer experience, and wasted advertising spend. While the adtech stack has endeavoured to fill the gap with Fingerprinting IDs, the most innovation has come from Apple and Google, and their focus on cohorts and differential privacy.
The demand for tracking individuals across the Internet will always exist. Several regulators, most notably the European Parliament with their General Data Protection Regulation (GDPR), have tried to limit the amount of consumer data collected and monetized, without consumer consent. Google and Apple have stepped up their efforts to limit how cookies are used, and how consumer data can be captured through Chrome and Safari. Apple has also limited the way mobile apps use, store and share consumer data. This has left the programmatic advertising industry somewhat in disarray. It has also further concentrated power with the walled garden operators, like Facebook and, of course, Google, Amazon and Apple.
