From shared TV moments to personalised connections
In the past, the media mix of a big brand looked broadly similar for decades. TV sat at the centre, supported by a handful of mass channels. When everyone watched the same programmes at the same time, counting eyeballs, reach, and frequency was often considered good enough. Only a few advanced advertisers went further with econometric models to prove impact.
Today, the landscape is very different.
Ads behave more like an invisible concierge, responding to context, signals, and intent in real time. Consumers interact with brands in countless micro moments across devices and platforms. Conversions can happen within seconds of exposure, often in the same environment where the ad was seen.
This creates a new expectation. Marketers are no longer content to know what was seen. They want to know what made a difference. That means understanding incrementality, what sales or outcomes were truly caused by media, as opposed to what would have happened anyway.
At the same time, this shift coincides with tightening privacy rules and heavy use of AI in platforms. Algorithms decide which combinations of creative, audience, and bids to deploy. Assets are bundled into packages that perform as a whole, making it harder to isolate the effect of a single element. Cookie deprecation and consent choices reduce the availability and reliability of user level data.
Marketers want more nuance and certainty from media effectiveness measurement at exactly the moment when granular tracking is becoming less feasible.
A modelled future
Moosman’s key point was that media effectiveness measurement is inevitably becoming more modelled. Perfect, continuous, user level tracking across all channels is no longer realistic, and clinging to that ideal is unhelpful.
Instead, advanced marketers are embracing:
Modelled conversions, using statistical methods to infer which conversions can be attributed to ads when direct, one to one tracking is not available due to privacy constraints.
Marketing Mix Modelling, MMM, to understand the total impact of all media and external factors on business outcomes over time, at an aggregate level.
Incrementality experiments, controlled tests that isolate the causal effect of specific campaigns, channels or strategies by comparing exposed versus non exposed groups.
None of these are new. MMM has existed for decades. What is changing is how they are combined. Rather than picking one method and declaring it the single source of truth, leading organisations use triangulation. They look for consistent patterns across MMM, attribution and experiments, and use discrepancies as prompts for deeper investigation or calibration.
Moosman stressed that the real bottlenecks are rarely technical. The tools and methods are there, from Google and other partners. The gap is usually organisational.
The invisible enablers
To reach a state where MMM, attribution and experiments work together, organisations need four underlying enablers.
Solid data foundations.
Consent based, high quality first party data. Proper tagging and implementation across digital touchpoints. Clear governance over who owns which data, how it is validated and how it can be used. Without this, even the best model will produce brittle or misleading results.
Clear KPI frameworks.
A shared understanding of which KPIs matter at strategic, tactical and operational levels. A clear mapping between media metrics, for example impressions, reach and ROAS, and business outcomes, for example revenue, profit and customer value. This ensures that media performance is discussed in a language that resonates with the C suite, particularly the CFO.
A measurement owner and team.
A named head of measurement who is accountable for the overall approach. Access to specialist skills in MMM, incrementality and tracking, whether in house or via partners. Enough authority to set standards, challenge questionable metrics and say no when colleagues want to cherry pick flattering numbers.
Embedded processes.
Regular learning reviews where insights from MMM, experiments and platform data are discussed. Clear links between those insights and optimisation cycles, such as changes in media mix, creative strategy or budgeting. A defined cadence for budget reviews where measurement evidence is expected and used.
Without these enablers, measurement remains a series of projects rather than a capability. Reports get shared, but little actually changes.
A people problem, not a technology problem
Moosman’s conclusion cut through much of the noise around AI, cookies, and dashboards.
Measurement is not a technology problem. It is a people problem.
The technology, from modelled conversions to MMM platforms and experimentation tools, is already available. What is often missing is ownership, power, process, and discipline. Someone whose job it is to care about measurement end-to-end. The mandate to standardise KPIs, push back on weak claims, and influence budget decisions. Regular forums where evidence is reviewed and turned into action. The will to look beyond short-term platform metrics and keep a long-term view of effectiveness.
In that sense, the future of media effectiveness is both modelled and human. Models do the heavy lifting in the background, filling in gaps that privacy and AI create. People decide which questions to ask, how to interpret the answers, how much weight to give different sources, and ultimately how to change course.
For CMOs and marketing leaders, the message is clear. You do not need to wait for the next big tool to improve media effectiveness measurement. You need to build the organisational muscle to use the tools you already have and to let them influence how you plan, spend, and defend your marketing budgets.
If you are ready to move from fragmented reports to a unified view of impact, Nepa is here to help.
Watch a summery from the event
Get in touch to explore how our media effectiveness solutions can support your next step.
Published on: 24TH FEB 2026