WARC projected that global advertising spend would exceed $1 trillion in 2024 for the first time. At that scale, platform activity is easy to count, but business impact remains difficult to prove.
A credible measurement strategy needs more than one identifier or attribution model. Brands need governed first-party identity, supported exposure and conversion data, consistent outcome definitions, and a mix of observational and experimental methods suited to each channel.
Key Takeaways
The $1 trillion milestone was a WARC forecast for global advertising in 2024; use it as historical context, not a timeless measure of current spend.
Authenticated identifiers such as UID2 can support addressability and measurement across participating environments, but their coverage depends on access, implementation, permissions, and adoption.
Hashing or tokenizing an email address protects the source value in specific data flows; it does not by itself establish consent, lawful use, anonymity, or privacy compliance.
Identity matching can connect supported exposures and outcomes, while experiments and other methods are needed to assess whether advertising caused incremental results.
Why advertising measurement needs a stronger foundation
Advertising platforms report delivery, impressions, clicks, conversions, and modeled results inside their own environments. Those metrics are useful for operating campaigns, but they do not automatically reconcile customers, purchases, returns, loyalty activity, or offline transactions across a brand's systems.
A brand-level view requires consistent definitions and identity. Teams need to know which outcome counts, which time window applies, how duplicate events are handled, whether returns are included, and which customer or household received the treatment.
What authenticated identifiers do
Authenticated identifiers are derived from information a person provides in a direct relationship, such as an email address or phone number. Within Unified ID 2.0, normalized email addresses or phone numbers can be mapped to raw UID2 values, and encrypted advertising tokens can be used in supported advertising workflows.
UID2 is deterministic within its framework, but it is not universal. It operates across participating publishers, advertisers, data providers, and demand-side platforms, subject to credentials, policies, opt-out handling, regional requirements, and technical integration.
An authenticated identifier can improve the chance that supported events refer to the same person or account. It does not reveal every touchpoint, work inside every platform, or prove that an ad caused a later purchase.
Hashing is not the same as consent
Hashing and tokenization can reduce exposure of the source email address or phone number in a workflow. The resulting identifier still represents a person and must be handled according to applicable law, contracts, participation policies, consent signals, and the brand's intended use.
The UID2 framework includes participant requirements and opt-out processes. Brands must still determine whether they have permission to collect, transform, share, activate, retain, and measure the underlying data in each market and channel.
Why identity resolution matters before activation
Customers often use multiple email addresses, phone numbers, loyalty accounts, devices, and household relationships. Sending each identifier independently can fragment audiences, suppressions, frequency decisions, and measurement.
Identity resolution connects records using governed rules and evidence before a brand maps eligible identifiers into an advertising framework. Teams should test false merges, missed matches, shared contact information, household logic, token refresh, opt-outs, and destination-specific requirements.
Use several measurement methods
No single method provides a complete answer across the advertising ecosystem. A practical measurement design uses each method for the question it can support:
Platform reporting helps teams operate delivery, pacing, creative, bidding, and platform-defined conversions.
Identity-resolved analysis connects supported exposure, audience, customer, and outcome data to examine who converted and what else is known about them.
Test and control groups estimate incremental impact when assignment, sample size, contamination, and outcome data support a valid experiment.
Attribution models allocate credit under stated rules and assumptions; different models can produce different answers.
Marketing mix modeling evaluates aggregate relationships over time and can complement user-level methods where granular coverage is incomplete.
Clean rooms and platform-specific analyses can support measurement where raw event-level data cannot move freely, but the available joins and outputs remain subject to each environment's controls.
Connect online and offline outcomes carefully
A brand with the necessary data and permissions may connect supported media exposures or audiences to ecommerce, point-of-sale, loyalty, or other outcomes. That view can reveal purchases that platform conversion reporting does not contain.
A match is not automatically an attribution result. Teams still need a documented lookback window, deduplication rules, eligible populations, return handling, identity confidence, and an appropriate method for distinguishing correlation from incremental impact.
Build a measurement loop, not another dashboard
Measurement becomes valuable when it changes the next decision. Bring available customer, campaign, media, conversion, and transaction data together; investigate performance in customer context; then refine audiences, signals, journeys, tests, and campaign choices.
Amperity's Measurement and Optimization capabilities connect supported marketing performance and outcome signals to trusted customer context. Teams can analyze what happened, identify what deserves investigation, and make the next decision while retaining control over interpretation and action.
See how Amperity can support identity-resolved advertising measurement using your available platforms, outcomes, permissions, and test design. Request a demo to evaluate the complete measurement path.