‘Too much data, too little magic’: Why it is important to question blind faith in algorithms

New Delhi: Marketers today sit at a crossroads. On one side lies a world driven by dashboards, algorithms and performance numbers that promise sharper targeting and higher returns. On the other hand lies the enduring craft of marketing, built on creativity, storytelling and human intuition, that has always shaped how brands connect with people. 

The tension between the two is no longer theoretical. It plays out every day in campaigns that either strike the right balance or fall flat by leaning too heavily on numbers.

This tension was laid bare at a recent roundtable hosted by MMA Global’s Marketing Leaders' roundtable, where senior marketers spoke candidly about the risks of letting data take over completely. 

While nobody dismissed the importance of analytics, several participants warned that an obsession with numbers can drain the life out of marketing. 

Over-analysis, blind faith in algorithms and misplaced confidence in cohorting were cited as traps that threaten both creativity and consumer trust. At the same time, the discussion highlighted how zero-party data and a sharper sense of context can help marketers rediscover balance.

Some argued that the industry is at risk of losing the “magic” of marketing by being overwhelmed by data. Others cautioned that platform-driven open targeting has reduced the role of marketers to mere bystanders. Throughout the conversation, one key reminder stood out clearly: data can guide decisions, but it cannot replace context, creativity, or long-term brand building.

When too much data takes the magic out of marketing

The risks of leaning too heavily on numbers came up early in the discussion. Anupam Tripathi, former GM - Marketing and Media Head atLenskart.com, pointed out that over-analysis can hollow out the creative work that makes marketing memorable. 

“When you over-analyse the data and take the end numbers, you end up losing the magic of marketing,” they said. “I truly believe if you do too much analysis of data, and primarily from brand love and awareness, loyalty and positivity, you kind of lose the whole point of marketing,” he said. 

Tripathi added, “You might shoot ten films, confidently pick the top three, and they end up being the least successful. I believe that while expectations should be high, doing too much data analysis, especially for things like brand love, awareness, loyalty, and positivity, can cause you to lose sight of what marketing is all about.” 

The point struck a chord that while data can inform, it cannot fully explain why a message resonates. Without space for creativity and instinct, marketing risks becoming mechanical.

Platforms, algorithms and the erosion of control

From over-analysis, the conversation shifted to the growing influence of platforms in deciding who sees what. A brewing consensus at the table inclined towards a feeling that it has weakened marketers’ control over strategy.

This was highlighted in the forum by Damandeep Singh Soni, Chief Business Officer, Astrotalk. “What has happened is that marketers' thinking has become a bit depleted. The platforms have taken the choice away from us, so all you have to do is turn on open targeting, and Google or Meta will decide who to serve the ad to,” said Soni. 

Soni stated that young marketers are going down this path of open targeting very rapidly. This contrasts with traditional methods where marketers meticulously chose demographics, interests, and other audience specifics. According to Soni,  this approach diminishes a marketer's role in understanding consumer needs and desires.

Soni stressed that a thoughtful marketer can craft a single, impactful headline that performs better than a bunch of randomly combined lines optimised by an algorithm. 

Why context cannot be ignored

Even when the data is clean and the algorithms are sharp, campaigns can falter for one simple reason: context.

Soni was quick to point out that while there is no escape from the need for data, one cannot let go of context. Soni argued that the problem with data is that it often lacks context. It's nearly impossible to quantify or “tag” a data point with all the external factors that might influence it, such as the time of day, a holiday, or a major news event.

To illustrate this point, Soni used an analogy: a bad meal at a restaurant might just be a result of the chef having a “bad day”, not a sign that the restaurant is always bad. Similarly, a creative that performs poorly once might do well in a different context. And vice versa is true as well. 

“Because people want to succeed faster, they might also re-run a best-selling creative and not see the same results because the context has changed,” said Soni. 

In other words, data shows outcomes, but it does not explain conditions. And without the ability to read conditions, marketers risk misinterpreting what the numbers really mean.

Zero-party data as a foundation

If over-analysis and over-automation were the villains of the discussion, zero-party data emerged as one of the heroes. Unlike third-party data scraped from browsing or purchase behaviour, zero-party data is information consumers willingly share. It offers both clarity and trust.

Ruchira Jaitely, Chief Marketing Officer, Diageo India, spoke about her past experience with a campaign which used AI to analyse stool patterns to reassure parents about gut health. “It resulted in absolutely fantastic zero-party data,” Jaitely said. “It created a whole new foundation for the next 50 years of growth for the brand and the business,” she added. 

The point was not about the novelty of the product but about what it achieved: data that was volunteered, not extracted, and insights that could fuel brand growth for decades rather than days.

Cohorts that overlap, communities that unite

But not all data-driven segmentation is so dependable. Mridul Kr Verma, Senior Director, Nielsen, cautioned against assuming too much precision in cohorting.

“The cohorts you define are not as unique as you think. There is a huge overlap between them, so be very sure what you are cohorting,” he said.  

The alternative, Verma argued, was to think less about demographics and more about communities. “Community is a great idea because it is an action you can rally people around, rather than saying this person is a biker or this person is into trekking. It could be the same person; you don’t know that,” he said. 

Communities, unlike cohorts, are built on shared interests and behaviours that people choose to express. They offer marketers a way to connect on purpose rather than on probability.

Tackling the paradox

By the end of the session, the conclusion was not that data should be abandoned but that it should be handled with restraint. Analytics can measure outcomes, but they cannot define meaning. Algorithms can optimise, but they cannot imagine. Cohorts can target, but they cannot inspire.

As Jaitely from Diageo put it, marketing today risks a “dangerous tilt” towards forgetting what it is supposed to do. The antidote is balance: using data intelligently while protecting the creativity, context and long-term vision that numbers alone cannot provide.

The paradox remains. Data is both a gift and a threat. It can unlock growth and precision, or it can flatten marketing into lifeless optimisation. What separates the two, as the roundtable showed, is not the size of the dataset or the sophistication of the tool, but the judgement of the marketer using it.