While three-quarters of brands still fly blind on platform metrics, more advanced teams are triangulating attribution, MMM and experiments into one intelligence layer. And for many, the measurement maturity gap is quickly showing up as a significant growth divide.
The economics of digital performance have shifted. Paid search costs are climbing, attention is fragmenting across channels and formats, and privacy-driven signal loss is eroding the neat, click-based narratives marketers once relied on. Meanwhile, boards are asking more pointed questions, including where the next unit of revenue actually comes from, and how confidently it can be repeated.
“You can’t achieve results you can trust if the measurement layer isn’t in play,” said Jaco Lintvelt, MD of Incubeta ZA, hosting the recent Future of Measurement webinar. “We’re seeing a big shift from asking if a campaign really worked on a particular platform, to whether it created profitable growth we can take into the boardroom.”
The three speeds of measurement
According to Eddie Schutte, marketing intelligence consultant at Incubeta, most South African organisations are still stuck in the early stages of measurement maturity, moving at basic speeds.
At the lowest level, brands count what is easiest such as platform attribution, surface-level analytics, and reporting that simply keeps the lights on.
The next level adds stronger analytics and cloud integration, with more sophisticated dashboards and some multi-channel attribution. It is useful, but still largely descriptive and not a reliable guide to what should happen next.
The real breakaway is at what Incubeta calls the pioneering speed. Here, measurement stops being a rear-view-mirror analysis and becomes a live growth intelligence layer. Brands at this level run advanced and predictive analytics in near real time, track total sales attribution rather than just last-click, and bring brand health and creative performance into the same frame.
Crucially, they triangulate between attribution, media mix modelling (MMM) and structured experiments so that each method continuously audits and calibrates the others.
SA faces a growing maturity gap
According to Schutte, triangulation is vital. “Like-for-like paid search costs have jumped, AI overviews are siphoning off clicks, and only a tiny fraction of users ever get to your traditional ad. In response, about 80% of enterprise marketers have handed the keys to platform automation, including bidding, targeting, and even creative, often on top of fragmented, incomplete signals. If your measurement is shallow, you’re using world-class AI to automate capital waste at an unprecedented velocity,” he warned.
Schutte also warned that there is a growing maturity gap between a small group of marketers who have moved beyond single-source, platform-led reporting to unified, triangulated measurement, and the majority who are still optimising on incomplete, biased signals.
A three-phase solution
When looking at the solution, Incubeta believes brands must first lock down their commercial reality by joining real-time web and app behaviour with offline CRM and point-of-sale data, while piping in campaign costs, clicks, impressions and impression-level exposure from all major platforms into a single analytics environment.
Second, they must maximise signals through a durable identity spine. User IDs, hashed User-Provided Data (UPD) data and transaction IDs, combined with disciplined event hygiene, give modelling and bidding algorithms resilient, high-value signals that can withstand ongoing privacy changes. Schutte warned that without this, “the machines are forced to guess” which is sub-optimal.
Third, they need to activate this data strength across the organisation. Schutte advises that analysts, media buyers and finance should share one calibrated decision engine, not a patchwork of conflicting dashboards.
Proof that it works
Activating data strength bridges the gap between commercial strategy and deep statistical validation. It is the exact type of uncompromised data ecosystem that Fred Kimokoti, director of data science at Incubeta, uses to deploy advanced, open-source Bayesian models, and the company already has definitive proof that this framework works.
A recent project for German tax-filing company, Taxfix, shows this process is effective.
Until recently, the business had been using a black-box model that couldn’t clearly distinguish baseline demand from incremental media impact, inflating the apparent contribution of certain channels.
Rebuilding the MMM using Google’s Meridian and integrating full marketing activity, owned media, external factors and search query data improved baseline estimation by more than 10 percentage points and cut channel contribution inflation by 54%.
Practically, Taxfix stopped paying for wins that were really organic, reallocated more intelligently into the upper funnel, and reshaped budget flows around the tax deadline by using MMM as a live instrument to engineer ROI, not just report on it.
Looking through the leadership lens, Lintvelt wrapped up by pointing out that rapid market and technology changes are reshaping what it means to be a marketing decision maker. “If the platforms and models become more intelligent, our roles have to evolve with them. The value we bring is not just in operating the tools, but in deciding which signals matter, how we integrate brand and performance, and how we use this intelligence to unlock profitable growth.”
