Meta Ads, Google Ads, TikTok Ads, or even Criteo, each ad network operates based on an advertising algorithm to best reach your marketing target according to your goals. To achieve this, these systems feed on data that you must provide them, and this is where the tracking of your site comes into play. But more and more, marketers are being asked to provide user data in addition to simple site conversion indicators.
Why do platforms request them and what do they serve for?
Basic Data Collection
Let’s start from the beginning, each ad network provides “pixels”, which are scripts that you must implement on your site to be able to transmit information about users’ behavioral actions. To set up these devices, there’s no need to be a developer, Google Tag Manager allows you to install tags without touching a single line of code. It is these same tags that communicate between your site and the advertising solution.
Every time a user visits a page, fills out a form, makes a purchase, or even spends time looking at a product, information is sent to third parties for which you have previously configured specific tags. These can include a wide variety of information, ranging from details about the product or service purchased, the value of the transaction, or even the discount coupon used.
This wealth of information allows ad networks to analyze in detail users’ browsing and purchasing behavior.
The Role of Advertising Algorithms
If Google Ads, Meta, or even Criteo play a major role in the current marketing landscape, it’s thanks to their targeting algorithm coupled with massive audiences. The promise, therefore, is to increase businesses’ profitability by finding the right potential customers within the user audience.
This is where the previously configured tracking comes into play. How could Google Ads know if a click on a sponsored link leads to a transaction if it doesn’t receive the information? How could Meta communicate a cost per lead if no tag returning the number of completed forms was communicated?
It is precisely by transmitting this data that the algorithm will optimize its targeting. In essence, with each conversion, the ad network will feed a database. The fuller this database, the richer and more accurate it becomes, and the more the algorithm can adjust and refine its targeting. This constant accumulation of data creates a virtuous circle of learning for the algorithm. As a result, it can identify more sophisticated patterns and more accurately predict which type of user is most likely to convert.
Each transaction, each completed form, and each significant interaction is a new piece of the puzzle. Together, these pieces help to complete an increasingly detailed picture of user behaviors, preferences, and habits. From this picture, the algorithm can identify trends, make predictions, and determine which type of advertisement is most likely to succeed with each type of user.
Thus, visitor tracking not only benefits marketing platforms by providing them with the necessary data to optimize their targeting. It also benefits you by improving the accuracy and efficiency of your ads, thus increasing your profitability.
No cookie, no party?
To summarise, a visitor clicks on a Facebook ad and then buys a product, the tracking tag sends to Meta the information related to the conversion which is stored in a database. But how does the ad network match clicks on ads with actions taken on your site?
Thanks to cookies!
When a user clicks on an ad, a cookie is dropped on their browser. This cookie contains a unique identifier that allows the ad network to follow the user as they browse the web.
When this user performs an action on your site, such as a purchase, the tag detects this action and sends the corresponding information to the ad network, along with the cookie’s identifier. The ad network can then match the cookie’s identifier with the one it recorded when the ad was clicked. This is how it can attribute the action performed on your site to the click on the ad.
Cross-browser / Cross-device Behaviour
In theory, for the system to work, the user must click on an ad and carry out the conversion on the same browser and the same device. However, behaviors are increasingly cross-browser (between several browsers) and cross-device (between several devices).
For example, a user clicks on a Pinterest ad for a sneaker brand on their smartphone in the morning on public transport. In the evening when they get home, they turn on their computer and place an order on the site they discovered that morning thanks to the ad. We can agree that the conversion should be attributed to the Pinterest ad, yet this will not be the case. Indeed, even if the ad network does receive the signal of the purchase, the click cookie does not exist, and therefore it cannot match the conversion to a campaign.
That’s why advertisers are now being asked to communicate the email address used to make the purchase. This is not transmitted ‘in clear’ but must be hashed by an algorithm (SHA256 or MD5 for example) before being sent.
By collecting data like user email addresses, ad platforms can “link” user activities on different devices and platforms, and also match a social account with a transaction. The Pinterest ad network (like all the others), saving all the clicks associated with a user account, will therefore be able to match the purchase email with the click email if it matches the accounts.
While most platforms are satisfied with harvesting the email address, some go much further by asking for the following parameters:
- First name
- Last name
- Phone number
- Postal address
- Country
- Birthday
So don’t forget to be reasonable about what you decide to send to platforms and consult your DPO/legal service before setting up this type of solution. You can then document your choices in a repository to justify them based on legitimate interest if it is real.
It is important to note that your main goal remains to optimize your campaigns and not blindly enrich the databases of the web giants. That’s why I always advise stopping at an email address and/or phone number depending on the platform used and the client’s context.
What about User Consent
Since 2018 and the introduction of the GDPR in Europe, any deposit of a tracker in a client’s browser must be subject to consent. However, it is important to note that consent for cookies, and consent for the collection and use of user data, are two completely separate things.
Thus, even if a user has given their consent for the use of cookies, this does not necessarily mean that they have consented to all forms of data collection and use. That’s why it’s mandatory to inform users about this data collection, to explain its purpose, and especially to give the choice to refuse the transfer of the email (for example) while leaving the possibility to accept cookies.
Therefore, you must add a step in your Axeptio consent management platform, specific to the collection of personal data. It is also essential to enrich your cookie policy and your privacy policy with detailed information about how you collect and use these user data.
Conclusion
In conclusion, user data has become a vital resource for ad networks, which are all asking for it in 2023.
The importance of this data extends beyond simple ad targeting optimization. With multi-browser and multi-device user behavior, these data also provide a means of tracking increasingly complex purchasing behaviors.
Ad platforms are becoming more and more data-hungry, even if this can sometimes seem on the edge of legality. However, good ideas sometimes emerge, such as the new Google Ads feature allowing the sending of the customer’s Lifetime Value who made the conversion. The goal is to optimize targeting toward long-term profitable customers. Let’s, therefore, strive to track better instead of tracking more!
