What Michelin-star chefs can teach marketers about data
Opinion
In the same way a chef preps ingredients before service, marketers need to ensure their data is fit for purpose before it feeds into decision-making.
There’s a reason Michelin-star chefs obsess over ingredients. They know a simple truth: no amount of skill can rescue poor inputs. If the produce is subpar, the dish will be too.
It’s a mindset marketers would do well to adopt. In modern marketing, data is an essential ingredient. But all too often it’s treated as an afterthought, and outputs suffer as a result.
Marketing and media teams collect vast volumes of data – audience signals, behavioural insights, campaign metrics. But rarely do we interrogate its provenance with the same rigour a chef applies to sourcing.
Do you know where your data comes from?
Do you know who is answering your surveys and whether they’ve answered truthfully?
A recent study carried out by researchers at Electric Twin found that where we conduct surveys substantially shapes results – to such an extent that the same survey pointed to different conclusions depending on where the questions were asked.
Most marketers will have experienced first-hand how poor quality data ultimately leads to poor quality results. We’ve all worked on campaigns that look right on paper but fail to land in-market. Developed strategies built on insight decks that feel convincing but lack star power.
Not every failed campaign can be blamed on poor data. But like cooking, marketing is a craft judged entirely on output. The quality of the work marketers produce is often constrained by the quality of the inputs. These include skill, creativity, technology, design and innovation, but data insight is invariably a critical factor, too.
The challenge for marketers is not a lack of data, but an overabundance of insights that are inconsistent, poorly understood and hard to act upon.
We might have first-party data that’s rich but narrow, or third-party data that’s broad but opaque. Survey data often captures stated intent, not real behaviour. And platform data is often shaped by its own incentives and definitions. Individually, each source has value. But when combined without scrutiny, they can create a distorted picture of reality.
This is where the chef’s mindset matters
Michelin kitchens don’t just source ingredients; they curate them. They understand seasonality, traceability and suitability for purpose. Not every ingredient belongs in every dish.
For marketers, this means moving beyond volume to validity. High-quality, well-sourced data is what enables campaigns to truly resonate – and that principle is becoming even more critical as new technologies reshape how insight is generated.
Synthetic audience modelling promises to compress research timelines from weeks to hours and is enabling teams to scale their research output exponentially.
Brands including The Times, Virgin and Mars are already using synthetic audiences to simulate how different audience segments are likely to respond to creative, messaging or new products. The technology offers a powerful new way to test and refine ideas before they go live.
But there’s a catch. The best and most accurate synthetic audience tools model populations using the seed data an organisation has already collected. They’re not building generic approximations of your audience, but accurate personas based on real data representing real peoples’ thoughts and opinions. That’s what makes them so powerful. It’s also why feeding the system with quality data is critical.
In a marketing landscape already being shaped by AI – where data is collected not just to inform campaigns but to build and scale synthetic populations – the quality of your data will shape the quality of your output more than ever before.
To ensure data is fit for purpose in the age of AI, we must therefore invest not just in curating quality but in preparing it appropriately: cleaning, structuring and contextualising data so it can be understood by AI systems designed to extract even more value from the insights you’ve already got. In the same way a chef preps ingredients before service, marketers need to ensure their data is fit for purpose before it feeds into decision-making.
This requires ongoing evaluation. Whilst the way we make decisions tends to endure, the information, context and reference points that inform decision-making inevitably evolve.
Data decays, audiences shift and culture changes. What was accurate or relevant six months ago may no longer hold true today. The best teams build processes that continuously refresh and validate their inputs, so that they’re in the best position to make the right calls.
None of this diminishes the role of creativity
If anything, it reinforces it. Great chefs don’t just rely on great ingredients; they use them to create something distinctive. But their creativity is grounded in an understanding of how to get the best out of those ingredients, and an honest reckoning with their limitations. The same applies to marketing.
High-quality data doesn’t replace instinct or imagination. It sharpens them, giving teams the confidence to pursue bolder ideas because they know they’re built on a strong foundation.
As pressure on marketing performance intensifies, the margin for error continues to shrink. Media costs are rising, attention is fragmenting and expectations around ROI are only increasing. In this environment, the brands that outperform won’t be those with the most data, but those with the best data discipline who know how to maximise its value.
Michelin-star chefs don’t leave that to chance. Neither should marketers.
Leanne Tomasevic heads up insights and strategy at Electric Twin