From Nuggets to Crowns: The Next Era of Experimentation

“I remember we had done something similar a couple of years back, but I cannot quite recall the outcome.”

That line came from a client strategy meeting late 2024. We were reviewing audience behaviour through our Brand Patterns model when the client recalled an old experiment. After some digging, we found the study; it validated the inference we had just made.

What stayed with me was not the validation. It was how easily the learning could have been truly lost.

This is not isolated. Eight in ten marketers say they struggle to reconcile results across fragmented tools (MMA, 2024). Valuable insights vanish in decks, emails or memory, rediscovered only by chance.

I often say: “Running a test is like refining ore: it requires rigour, but what emerges is pure gold. The tragedy is how often that gold slips through the cracks.”

When Nuggets Multiply 

Globally, the share of marketers using structured experiments doubled in a single year, from 18% to 36% (WARC, 2024). In EMEA, nearly nine in ten UK leaders say experimentation is essential to achieving 2024 goals (Optimizely, 2024). 

In an environment of rising costs, privacy-driven signal loss and budget scrutiny, proof is not a luxury; it is the ticket to investment.

MENAT is following this trajectory, though under more complex conditions. Business optimism remains (HLB, 2024), but 72% of marketers globally expect economic conditions to impact strategies in 2025, with only a third expecting budgets to rise (WARC, 2024). 

Uncertainty strengthens the case for experimentation. When markets move fast, messaging, platforms and media mixes must be stress-tested more often. One area where budgets may hold or grow is test-and-learn activity, because it helps brands adapt without committing large sums blindly.

MENAT shows signs of this shift. The A/B testing tools market, though still small, continues to expand at double-digit rates (Cognitive Market Research, 2024). Appetite grows, not because the environment is easy, but because it is unpredictable.

Though golden nuggets multiply, the question remains: are they enough?

Are Nuggets Enough? 

Leading platforms report tests increased by more than 50% in two years (Optimizely, 2025). Returns can be spectacular: 92% of retailers using structured test-and-learn programmes report at least a tenfold ROI (Mastercard, 2024). 

Yet cracks show. Nearly half of ideas tested fail to validate their hypothesis (Mastercard, 2024). Too often, experimentation focuses on surface tweaks — button colour or call-to-action placement. What matters more are sophisticated experiments that address incrementality: isolating the true impact of channels, audiences and mixes. While many marketers run A/B or lift tests, fewer than half (45%) use econometrics or marketing mix modelling to measure impact at scale (WARC, 2024). 

Another gap looms: most marketers struggle to piece results together across siloed tools and teams (MMA, 2024). My spreadsheet days proved logging hypotheses, outcomes, and lessons was messy but powerful; together, entries revealed patterns a single test never could. But spreadsheets do not scale. Beyond one analyst or team, nuggets scatter before they shape strategy.

That is the paradox: adoption has accelerated, but quality, sophistication, and consolidation have not kept pace.

From Scatter to Strategy: Building the Operating Model 

The answer is consolidation — not just software, but an operating model.

Aggregated learnings evolve from anecdotes into firm, comparable evidence. Benchmarks form, enabling leaders to answer questions like, “What’s the average lift from retargeting in this sector?” with confidence. Aggregation moves decisions from instinct to proof, creating governance, guardrails and accountability.

Essentials are consistent across organisations:

• Roles & governance: owners curate repositories; reviewers ensure quality.
• Shared schema: standard fields — hypothesis, metric, minimum detectable effect (MDE), audience, variants — make results comparable.
• Decision rules: clear guidelines on when to scale, retest or pivot.
• Meta-analysis: regular reviews to surface repeatable success factors.

Industry moves show this shift is underway. Platforms like Meta and Google maintain archives of brand-lift studies. Retailers such as Walmart and Amazon run structured test-and-learn programmes. Within UM, our Experimentation Lab centralises designs, guardrails and outcomes — part of a movement to turn scattered nuggets into crowns.

This discipline elevates tests from tactical to strategic assets and guards against test theatre, vanity metrics, duplication and tool sprawl.

Crafting the Crown 

One experiment may be a nugget of gold. But when gathered, organised, and shaped with intent, those nuggets become something greater: a crown of strategic advantage.

Globally this shift is underway; in EMEA it is becoming standard; and in MENAT the appetite is clear though the challenge remains: not just to run more tests, but to ensure their wisdom compounds.

The opportunity is simple: craft crowns, not just collect nuggets. How is your team approaching this? I would love to hear how others are capturing and scaling their own insights in practice.

Sources: MMA, 2024 | WARC, 2024 | Optimizely, 2024 | HLB, 2024 | Cognitive Market Research, 2024 | Mastercard, 2024

Rishad Alavi - Manager - Data & Analytics, UM UAE