In the past 15 years, the customer journey has become a multi-platform, multi-device bonanza, with data flooding in from laptops, desktops, mobiles, tablets, TVs and smart wearable devices. Thanks to a number of tracking methods – cookies first and foremost among these – marketers have learned to follow their customers throughout the journey and tailor their offering to match their interests and preferences. But in the past 7 years, customers have started to realize that they’re sharing this data for very little return, and that it’s travelling further and faster than they could have ever imagined. All of a sudden, data privacy and protection have become very real public concerns.
The dawn of ad-blocking was the first sign that the public were getting sick of marketers feeding them a stream of branded messaging, then came a number of well publicized data breaches, such as the Snowden leaks about US and UK surveillance programs, the many revelations made by Wikileaks, and most recently the Cambridge Analytica/Facebook scandal and a number of personal data breaches by large corporations. Finally, as the cry for protection rose, Schrems vs. Data Protection Commissioner placed data privacy at the top of the agenda for public discussion.
Public pressure led to tough regulations such as the GDPR and the CCPA being passed, which made the tech giants started to sit up and take notice of their customers’ demands. This led to Firefox allowing users to block tracking cookies in 2018, after which Google shook the marketing world to its core by announcing in January 2020 that it intended to phase out third-party cookies by 2022. More recently, Safari and iOS update beat Google to the punch by blocking cookies in March 2020.
The decline of cookies, especially Google third-party cookies, which track roughly 90% of all search engine users, will have a massive impact on the marketing landscape in the next two years, reducing the reach and value of certain tools within the marketing stack, especially Data Management Platforms (DMPs). But brave marketers will see this development for what it is – an opportunity to evolve and develop better, more customer-friendly methods of identity resolution and brand journey personalization.
Why third-party cookies matter to marketers
A cookie is a small file that retains data. When you go onto a website, that site can write data to the cookie, to remember things about you or things about what you’re doing on the site. If you come back to the site, they know it’s the same person, hence they can tailor the browsing experience for you, because they’ll know what you did before.
A Data Management Platform (DMP) is a software platform used to collect and manage data. The platform allows any brand to upload their customer data, which will be stripped of all personal data (such as telephone numbers, email addresses, names and so on). That data is then tipped into a joint bank with data from hundreds of other brands, forming a deep data well for contributing brands to leverage in order to direct advertisements to specific audience segments.
For instance, they may wish to advertise to 35-year-olds from Georgia who like Nike trainers and drink Gatorade. They will also be able to create lookalike audiences that mirror their best customers, and push adverts out to those people to. It is a way of expanding the reach of every brand investing in the DMP and advertising to new, primed audiences with content that will pique their interest. Only when someone clicks, and then hits their website, will the brand start to be able to track them using a first party cookie, and there will be some evidence in the tracking to say that they clicked on a certain ad, from a certain site, from a certain audience that was created originally in the DMP.
Meanwhile, the DMP sets third-party cookies (that is, cookies created not by a specific website but by an external domain) on all the websites signed up to its service, in the form of scripts or tags. These cookies are used to track the online behavior of customers who visits these sites, and show them tailored adverts. When a user clicks on the ad, the cookie will be loaded on their computer, so it will be possible to see where they go next and what they do. If they open up multiple tabs, that counts as a single ‘session’, and all information browsed becomes fair game.
A profile can be built up of the user’s favorite topics, brands, sites and products, alongside the original data loaded by a brand about the customer, with all this information being merged into an anonymous ‘identity’. The customer could hit 100 websites, and each would drop a cookie on them, leading to a continual collection of browser ID, device ID, IP address or any other form of digital ID, so the DMP can recognize a particular person from that individual ID.
It isn’t only DMPs that use third-party cookies either, data vendors use them and sell the information and anonymous profiles harvested to interested companies. If the ‘person’ captured in the anonymous data suddenly identifies themselves to the brand – for example, they sign up to a newsletter and it turns out their profile is already held by the company, then the anonymous behavioral data can be added to the known information, to enhance the customer profile.
What will happen when the third-party cookie hits the fan?
Of course, when third-party cookies are blocked, this tracking will stop. There will be tick-box on Google, Safari and so on that allows the search engine users to turn on these cookies if they want to, but most probably won’t, partly because of concerns about personal data being accessed and used by companies, and partly because filling in any box takes action, and most people need some form of motivation to take action. So, third-party cookies will, in effect, be switched off like a light all over the globe.
Already harvested data, connected to devices, virtual identities or known profiles, will continue to be available, but no new information will be added. From that day on, every new device, browser or unique ID will not be tracked, so the DMP will start to diminish as time goes by, and the associated advertising will start to become less targeted.
It might be that, right now, a company spends $1m on advertising through a DMP, and they reap a $5m return. Once that decline in targeting begins, however, that $5m will decline in tandem, until a point is reached where the return equals or drops below the investment. At that point, if not before, subscribing brands will cease to invest in DMPs and will take their data with them, shrinking the already stagnant data pool. Marketers will need to start looking elsewhere for solutions to their targeted advertising needs.
When cookies are gone, what’s next?
Luckily for marketers, third-party cookies aren’t the only option for recognizing their customers and the online behavior of those customers. For a start, there are first-party cookies to fall back on. First party cookies live on a brand’s own website, and track the actions of a user within a session on that website. When a new, anonymous person hits the website for the first time, then leaves without subscribing or leaving any personal data, they’re lost. However, if they come back a second time they bring the cookie with them. That means that the brand can collect the details of that second browse session, and the third, fourth, or any others from then on.
An anonymous profile can again be built up, encompassing settings preferences, products browsed, items placed in the cart. This information will only be available to that brand, and won’t be shared with any other companies, giving a competitive advantage. Each time the brand will need to try its best to authenticate the amassed behavioral data record, by offering the visitor something that will make them leave some form of personal data.)
All browsers support these cookies, so there’s no risk of them disappearing any time soon. What’s more, because they are self-set, brands can rely on the relevance of the information gathered, not to mention understand just how recent or how valuable it is, something that is harder to judge will a pool of information collected by somebody else. Identity graphs can also be used to bring as much brand data together as soon as possible to create device or browser IDs.
Then, in the same way as third-party cookies (and more easily, since the customer is already ‘on the spot’), the information gathered by the first party cookie has the potential be turned into an identified record. A visitor can be tempted by offers, adverts and other means to input personal information into a form, which can be attached to the previously anonymous behavioral data. If a record is already held by the brand about that person, this offline information can be blended with the online data.
In 2021, we are already aware of many initiatives where marketers are exploring investment in a Customer Data Platform (CDP) to capture and store the first party cookie to link with all the interactions with their other platforms, whether that be through the CRM, social media, website, eCommerce platform or app. The CDP acts as central repository for the marketing stack, and having data from lots of different systems improves the likelihood of identifying an individual. Email addresses, telephone numbers and other personal information from different systems can be merged and blended together.
Then, algorithms and heuristic rules sets (rules of thumb for decision making) can be used to determine which bits of information belong together in one profile. It might be that a single person has ended up in your database five times, across multiple systems, and that in some cases misspellings have crept in or pieces of PII have changed. The CDP will map the ingested data together, so the different fields from different systems are mapped onto single fields, and data within those fields is standardized and normalized, to make it consistent. It can then be cleansed, to remove likely errors, deduplicated, then presented as a view of a single customer, with all of their page visits, transactions, interactions and so on.
The Single Customer View can then be used to understand customer requirements and up to date communications preferences, personalize an individual’s brand experience using knowledge of their past behavior, or create segments of customers that can be shown adverts that might be of interest or personalized recommendations based on what similar people have looked at. All of this can be achieved using unique, relevant and accurate information that a person has willingly shared with you, removing the possibility of mismatched identities or the customer being unnerved by ‘creepy’ tracking, and making it much more likely that, as an engaged follower that knows and trusts your brand, the individual can be tempted to make a purchase.
By using first party data in this way, to match against an audience in Facebook or other customer-facing platforms, advertising platforms and DSPs will be able to make ad targeting more effective and reduce wastage, whilst CDP algorithms will make attribution between online activity and offline purchases more transparent, enabling brands to stop advertising a product to customers that have already purchased it.
Making the most of available resources
Of course, for the moment, third-party cookies are very much still alive, so it makes sense for brands to widen their advertising scope to as many suitable but unknown customers as they can while there’s still time. The real trick is to try and get completely anonymous prospects registered in the database, before the third-party cookies die out. One way to do that is to offer materials that will be of great value to those customers, such as gated research papers, tempting offers aligned to their interests, or the offer of rewards if they sign up to a loyalty card, to make them feel that providing their personal information is a fair swap.
For example, one apparel brand that BlueVenn spoke with on this very subject earlier in the week is already planning ways to authenticate anonymous new visitors to their website. We thought their idea of offering a free, personalized, low-cost T-shirt in exchange for a sign up was a good one.
Another way to resolve identities before the loss of cookies is through identity graphs. BlueVenn combines business-held first party data with behavioral data gathered by third-party cookies, along with all sorts of other third-party data such as electoral rolls and financial systems like Experian, through third-party identity graphs.
One newsmedia customer of BlueVenn has an offline subscription system, for people who want get papers delivered, as well as an online system, where people can subscribe via the website. The online and offline data is blended together within their Customer Data Platform, then a third-party identity graph is used to not only join that data together, but link the data of anonymous online customers of the news group to users known within the identity graph (for example, where a cookie has followed that person over 5 different sites, then they’ve provided an email address on one of the sites). This extra information provides an enhanced view of the customer, allowing better targeting with advertising, based on information that the newsmedia organisation hadn’t previously been privy to.
So, a CDP can be used to squeeze every drop of goodness out of third-party cookies while they are still around, making it worthwhile investing in one sooner rather than later. Then, when third-party cookies die out, brands can continue to build up customer profiles using first-party cookies and the CDP. There is life after DMPs and third-party cookies, and advertising agencies that manage DMPs for their clients or use them to help sell advertising will need to start considering how they can work in conjunction with clients’ CDPs, or advising them that they need a CDP instead.
If you’re planning to move towards phasing third-party cookies out of your advertising strategy, and are curious to know more about the benefits CDPs can offer, then sign up for one our demos today.