Choose a Lane: Measure or Target

The measurement and targeting functions in marketing technology are not the same thing. They ask different questions, use data differently, carry different compliance exposure, and should live in separate vendor relationships. The industry bundles them constantly — and that bundling creates problems that advertisers are often the last to understand.

Two Lanes, Two Different Jobs

Measurement asks: what happened, and why? It needs sequence. It needs to know what media a consumer encountered before they converted, in what order, across what channels. It doesn’t need to know who that consumer is by name, address, or persistent device ID. It needs the chain of events.

Targeting asks: who should I reach next? It needs identity. It wants to connect a device, a cookie, an email address, a physical location — to build a profile rich enough to make a targeting decision. That profile is the product.

These are different questions requiring different data. Measurement is retrospective: it looks at what happened and assigns credit. Targeting is prospective: it looks at what’s known about a person and decides whether to show them an ad. Running both functions through the same infrastructure, with the same data, under the same vendor contract, conflates two jobs that serve different purposes and carry different risks.

The Bundled Offer and Why It’s a Problem

The appeal of bundling is real. One platform, one integration, one vendor relationship, one contract to manage. Data Management Platforms have made this case effectively: the same identity graph that resolves audiences for targeting can also resolve journeys for measurement. Why pay for two things when one does both?

The structural problem is that a company whose core revenue is targeting has a different relationship with your conversion data than a measurement-only vendor. Conversion signals — who bought what, when, on which device — are the most valuable input a targeting platform can receive. Data services contracts commonly include “platform improvement” provisions that authorize use of client data to improve the underlying platform. In a DMP, that platform is the identity graph. A brand that uses a DMP for measurement may be contributing its conversion signals to targeting infrastructure that serves every other advertiser on the platform.

This is structural orientation, not bad intent. But the advertiser should understand which lane their measurement vendor is actually in — and whether that orientation affects the quality and independence of what they’re receiving.

The Compliance Consequence

The lane you’re in has direct compliance implications. Measurement that operates on aggregate behavioral signals and anonymized exposure records can stay clear of GDPR and CCPA’s most sensitive provisions. Measurement that touches PII to resolve identity — connecting a purchase event to a named individual across devices — crosses into territory those regulations treat more restrictively.

The advertiser typically bears this exposure, not the vendor. The vendor provides a service; the advertiser decides which data to feed into it and signs the contract that governs what happens to it. Understanding which lane a vendor is in determines what the advertiser is responsible for downstream.

There’s a concrete, verifiable check. Under the IAB’s Transparency and Consent Framework, vendors register their data processing purposes on the Global Vendor List — declaring whether they operate for measurement and analytics, for targeting and advertising, or both. CMPs surface those declared purposes to users and enforce consent accordingly. A vendor whose core product is identity resolution and audience targeting registers targeting purposes. A vendor whose product is measurement registers analytics purposes. The declared purpose determines which consent category users see, and what pass-through rate the vendor receives on your site. That registration is auditable by looking up any vendor on the IAB’s Global Vendor List. For a full treatment of how consent categories work as a measurement independence test, see The Measurement Companies That Forgot to Measure Themselves.

First-Party Data: The Durable Foundation

The long restriction of third-party tracking clarified rather than created this problem. Measurement that depended on third-party cookies for identity resolution was always in the targeting lane — using targeting infrastructure to answer measurement questions. Safari began blocking third-party cookies in 2017. Firefox followed in 2019. iOS App Tracking Transparency restricted cross-app identity resolution in 2021. Google reversed its planned Chrome deprecation in July 2024 — third-party cookies remain available in Chrome — but consent enforcement under GDPR and CCPA, and the cumulative effect of Safari’s reach across Apple devices, have nonetheless reduced the reliability of third-party tracking as a measurement foundation. The trend is real regardless of where Chrome’s policy settles.

First-party measurement isn’t caught in any of that. A tag that collects data on an advertiser’s own domain — used exclusively for attribution, not for audience building or identity graph enrichment — doesn’t depend on tracking signals that browsers and devices have been progressively restricting. Server-side tagging, direct integrations with publisher data, cookieless impression capture: these are measurement architectures designed for the environment as it actually exists.

C3 is built on first-party architecture. The tag collects data on your domain and for your measurement program. It doesn’t improve a targeting platform, contribute to an identity graph, or share data across client accounts. That’s not a privacy policy claim — it’s what the MSA says and what the CMP classification confirms.