Apple SKAdNetwork & Private Click Measurement - Next Advertising Era
Let's dive into Apple's world of measurement post-third-party cookie: the only solutions that have been active and used in the market for years (not without issues)
Apple was the first company to disable third-party cookies in its browser by introducing Intelligent Tracking Prevention. They were also ahead of the game in limiting the use of Mobile Advertising IDs (MAID), the equivalent of third-party cookies in the smartphone app world, on iPhones through the App Tracking Transparency framework. The Cupertino-based company deserves major props for bringing privacy issues into the mainstream conversation, turning it into a powerful marketing tool and forcing all players in the Ad Tech space to play catch-up.
Why this fierce focus on privacy?
Apple's stance on privacy might be seen by some as just a savvy marketing move, but the reality is, Apple has always shied away from invasive advertising in its products. Back on June 21, 2010, they rolled out the now-defunct iAd advertising program, which eventually fizzled out under a barrage of criticism by June 2016. In its place, Apple Search Ads emerged, currently the only advertising venture by the tech giant.
Apple has consistently steered clear of complex targeting options in its ad tools. The targeting features in Search Ads, for instance, remain pretty straightforward. Yet, the company has always placed a high value on performance metrics, pioneering alternative frameworks for ad measurement beyond traditional third-party cookies and MAIDs
What is SKAdNetwork?
The SKAdNetwork was created to measure advertising campaigns within smartphone apps, enabling tracking of conversions post-impression and post-click. The core concept involves notifying the operating system, iOS or iPadOS, about the advertising activity the user is exposed to. After a few days, iOS or iPadOS will handle sending any conversion information to the advertising platform and, if desired, directly to the advertiser.
SKAdNetwork, akin to Google's Attribution API for event reporting, operates under a similar framework. It anonymously gathers events like clicks and impressions through the operating system, later dispatching conversion data. This data is ‘noisified’ to ensure anonymity, and the granularity of details (like campaign, group, keywords) increases with the number of conversions tracked. The key goal here is to prevent user identification through the transmitted data. Contrary to Mountain View's solution, Cupertino's approach does not provide as precise aggregated reporting. Instead, the responsibility of managing these details is shifted to the Ad Tech industry.
The initial response to SKAdNetwork in the market was tepid. Its first iteration didn’t quite hit the mark for Ad Tech needs. However, its fourth release in 2022 marked a significant improvement, earning nods of approval from industry players. Despite these advancements, Apple's solution isn’t without flaws, leading companies like Meta to roll back to the earlier release 3
What is Private Click Measurement (PCM)?
Private Click Measurement is Apple's solution designed specifically for measuring post-click conversions in a web environment. However, it hasn't gained much traction in the market because it only covers one measurement scenario. Post-click measurement can be achieved with simpler and more established solutions. The latest updates in Safari 17, which involve removing click tracking parameters, might push the market towards adopting PCM.
Implementing Private Click Measurement involves modifying the advertising link by adding two attributes to the HTML <a> tag: 'attributionsourceid' and 'attributeon'. The first attribute identifies the advertising action, and the second specifies the advertiser.
At conversion time, the advertiser needs to make an HTTP call to Apple's servers, indicating the domain where the advertising occurred, along with a numeric identifier for the conversion.
I don't want to spend time on analyze these solutions like I would have in the past, because over the last year, there have been some interesting developments in the Apple sphere that might suggest a shift in direction from the tech giant
What is happening in Cupertino?
Apple and Google's measurement solutions, though pitched as open standards, only work in their respective closed ecosystems – Apple on Apple, Google on Google. SKAdNetwork isn't even a proposed standard. Both solutions are moving towards similar ideas, yet remain insular. Neither Microsoft nor the Mozilla Foundation have declared support for these technologies. In fact, the folks behind Firefox, along with Meta, have put forward an alternative standard: Interoperable Privacy Attribution (IPA).
It's evident that this situation is unsustainable: new proposed standards are popping up like mushrooms in the market. However, there's a silver lining – many minds are brainstorming potential solutions, and they've all come together under one roof in the W3C's Private Advertising Technology Community Group. Their goal is to find shared standards, starting with measurement.
Inside the PATCG in September (the only working group supported by Apple), something unusual happened: Cupertino put forward a proposal for a measurement standard before its implementation, called Private Ad Measurement.
At the same time, another odd event occurred in September: iOS 17 came with no updates related to SKAdNetwork – no version 5 of the framework.
Summing up Apple's moves in 2023:
They're not updating their measurement infrastructure.
They're pitching a new standard in the only truly global working group.
Especially noteworthy, their new proposal is receiving excellent feedback.
Conclusions on the Future of Measurement
We're in an exciting time for measurement, perhaps even more than for targeting. We're seeing new proposals and structures gaining ground. Many have feared in the past that we'd lose the ability to measure , but that's not happening. The digital market is responding to new limitations with strength and vitality. So, what should we do in the meantime?
Enhance your company's ability to collect first-party conversion data: make easy to link a transaction to a user ID. This will simplify the implementation of solutions like Google's Enhanced Conversion and Meta's Conversion API.
Collect click tracking parameters from different advertising platforms in your backend: when a user converts, along with conversion information, bring in Google's gclid and Meta's fbclid, preferably with a timestamp of when you collected the parameter. This will improve conversion tracking
Identify secondary conversions: they can help gauge interest in your brand and products/services. This will be useful for future use of the Protected Audience API.
Enrich user profiles. Start with obvious information like birth date, gender, job, but then identify elements related to your business. For example, if you're in the pet-related products, find out how many pets they have, their species, breed; if you're in fashion, gather preferences in size and color; in beauty, information about skin type, allergies, etc. This will help profile your customers.
Ensure easy linkage between product information in your online CMS and backend product data.
Ensure your analytics system collects user information in a privacy-compliant manner, enabling use of navigation data for profiling.
Collect click and/or impression information from advertising platforms for marketing mix modeling. If you haven't already, activate Google and Meta's MMM exports.
Easily obtain return and canceled purchase information in exportable formats by user, transaction, and product, to have precise real transaction data (don't believe those who say a conversion is just as valuable even if the user returns the products: if it's not valuable on the balance sheet, in fact, it's a cost, it can't be revenue at the marketing level).
Until now, we've used dirty conversion data over which we had little control. New privacy-oriented developments are leading us towards more precise data, directly linked to the company's backend. Sometimes, it feels like digital is waking up from the lethargy brought by the ease of data collection with a pixel and is finally striving for real transaction data: those valid for a balance sheet.
Previous Episode of: Next Advertising Era
Reading Suggestions
Be Data Literate by Jordan Morrow that is driving my professional choice
Present Beyond Measure: Design, Visualize, and Deliver Data Stories That Inspire Action by Lea Pica that is supporting my data storytelling
IAB Tech Lab Identity Solution guide from IAB to understand Alternative ID
Marketing in the messy middle Google's document to guide us through the Messy Middle, Mountain View's perspective on summarising the user's decision-making process.
IPA-PAM-ARA Comparison and Tradeoffs: This document contains the original comparison presented at TPAC. A working item from that was to create a steelman version for each proposal which we’ll do in this document: