Our Senior Director, Industry Development & Marketing, Marie-Clare Puffett shares highlights from her IAB Connected Commerce breakout session 'Measurement Deep Dive: What Really Moves the Needle'.
At the IAB Connected Commerce Summit in New York, I had the pleasure of moderating a session with Michael Lin, Senior Manager of Measurement Science at Instacart, who opened with a sharp, timely message: marketing measurement is shifting from correlation to causation. And that shift is no longer theoretical - it’s becoming a requirement.
For years, marketers have relied on patterns that look meaningful. When revenue rises alongside ad spend, it’s easy to assume the ads worked. But as Michael pointed out, those trends are often shaped by everything except marketing performance - distribution changes, pricing, competitive dynamics, or even internal budget biases. Correlation can tell us what happened, but not why it happened.
Other industries solved this problem long ago. Medicine uses randomised controlled trials to determine what truly works. Marketing is finally catching up. The IAB’s measurement framework makes this explicit, ranking methods by their ability to establish causality, with RCTs at the top, followed by model‑based approaches, and legacy tools like Marketing Mix Modeling (MMM) and attribution further down the ladder. The takeaway: not all measurement answers the same question, and treating them as interchangeable leads to flawed decisions.
This is where Retail Media changes the game. Platforms like Instacart can directly link exposure to purchase behaviour, enabling true experimentation at scale. Instead of inferring impact, marketers can observe incrementality directly, using deterministic data and controlled ad exposure.
Causal measurement is simple in concept: compare outcomes between a group that saw ads and a group that didn’t. The difference is the incremental lift. But doing this well requires rigor - proper randomisation, validated control groups, and clear statistical thresholds. When executed correctly, it delivers the metrics that matter most: sales lift and incremental ROAS.
MMM still has a role, especially for long‑term planning and cross‑channel strategy. But it was built for a world without granular, user‑level data. In today’s Retail Media environment, its limitations are more visible. The future isn’t either/or, it’s MMM for strategic context, experiments for ground truth.
As Michael emphasised - and as the audience Q&A reinforced - adopting causal measurement is as much a mindset shift as a methodological one. Start small, test incrementally, and let the evidence build. Once you see causality clearly, it becomes hard to rely on anything less.
Marketing doesn’t suffer from a lack of data. It suffers from a lack of clarity. Causal measurement brings that clarity, separating signal from noise and grounding decisions in evidence rather than assumption. As Retail Media continues to scale, this won’t just be a competitive advantage; it will be the expectation.
Want more insights into the IAB Connected Commerce event? Check out Marie-Clare's reflections from the event in our blog post here.
