For more than two decades, companies competed to rank on the first page of Google. Marketing teams spent millions on search engine optimisation, digital advertising and content marketing to ensure their brands appeared before consumers searching for everything from hospitals and banks to skincare products and home appliances.
Now, that battle is witnessing a shift.
Instead of scrolling through pages of search results, consumers have started asking AI assistants such as ChatGPT, Gemini, and Perplexity what to choose. The answer often comes back as a shortlist of just three or four names, instead of pages.
A Business Standard analysis of 375 AI-generated recommendations across five consumer categories found that a handful of brands consistently dominated AI answers. Apollo Hospitals, HDFC ERGO, HDFC Bank and LG appeared in more than 90 per cent of responses in their respective categories, while HDFC Bank and ICICI Bank featured in every AI-generated banking recommendation analysed. Skincare was the only category where recommendations were more evenly distributed across brands.
As AI assistants increasingly become the first stop for consumer advice, companies are beginning to ask a question that barely existed a year ago: Can India Inc influence what AI recommends?
What Business Standard analysis finds
To understand how AI assistants recommend brands, Business Standard tested three leading platforms – ChatGPT, Gemini and Perplexity – across five consumer categories: private hospital chains, health insurance companies, private banks, skincare brands, and home appliance brands.
Five standardised consumer prompts were designed for each category, ranging from broad questions such as “Which are the best private banks in India?” to more specific queries around customer service, quality, value for money, and specialist needs. Each prompt was run five times on every platform using fresh conversations, generating a total of 375 AI responses.
The analysis measured two indicators. The first was recommendation share, or the proportion of responses in which a brand appeared in the primary recommendation list. The second was first-choice share, measuring how often a brand was presented as the first recommendation. The analysis counted only the brands that AI assistants recommended in their main answer to each prompt. Additional examples, speciality-specific suggestions and explanatory mentions were excluded to ensure consistency across platforms.
The exercise was designed to capture recommendation patterns rather than produce a one-off snapshot. AI responses are known to vary across platforms and over time, making repeated testing essential.
The Business Standard findings broadly mirror emerging international research on how AI systems recommend brands. A recent study by researchers at the University of California, Irvine, and the University of Virginia analysed recommendations generated by ChatGPT, Claude and Gemini across 50 brands in five industries. It found that while AI assistants often converged on a handful of leading brands, they agreed on the top recommendation only 41.6 per cent of the time, highlighting that AI-generated recommendations remain probabilistic rather than deterministic. The researchers also found that large, well-known brands enjoyed a clear advantage in AI recommendations, although the extent varied across industries.
The Business Standard analysis extends those findings to the Indian market by examining how leading AI assistants recommend brands across key consumer categories.
Who AI recommends
Among private hospital chains, Apollo Hospitals emerged as the clear leader. It appeared in 68 of the 75 hospital responses, giving it a recommendation share of 90.7 per cent, and was the first recommendation in 57 responses, or 76 per cent of the sample. Fortis Healthcare, Max Healthcare, Narayana Health, and Manipal Hospitals also appeared frequently, but far less often as the first recommendation.
Health insurance showed an even stronger concentration. HDFC ERGO was recommended in 74 of 75 responses, giving it a recommendation share of 98.7 per cent, and ranked first in 72 responses, or 96 per cent of the sample. Aditya Birla Health, Care Health, Niva Bupa, and ICICI Lombard were the other most frequently recommended insurers, although none came close to HDFC ERGO’s dominance.
Private banking produced another striking result. HDFC Bank and ICICI Bank appeared in all 75 responses, but HDFC Bank emerged as the preferred recommendation, ranking first in 49 responses, compared with 22 for ICICI Bank. Kotak Mahindra Bank, IDFC FIRST Bank, and Axis Bank completed the leading group.
Home appliances once again showed strong concentration. LG appeared in 74 of 75 responses and ranked first in 71, giving it a first-choice share of almost 95 per cent. Samsung, Whirlpool, Bosch, and IFB followed as the most frequently recommended brands.
The pattern shifted in skincare. Minimalist recorded the highest recommendation share at 92 per cent, followed by Cetaphil at 84 per cent and CeraVe at 64 per cent. Unlike hospitals or insurance, however, AI assistants did not consistently favour one brand. Minimalist and Cetaphil were almost evenly matched as the first recommendation, suggesting AI views skincare as a more competitive category with several credible alternatives.
A new marketing race begins
Rrahul S Ranjan, managing director of global branding and marketing agency Mrig Sight Media, told Business Standard that AI assistants have evolved from simple question-and-answer tools into decision-making engines that increasingly influence consumer choices. As a result, the agency now routinely checks how brands appear across ChatGPT, Gemini, Claude and Perplexity alongside traditional search rankings.
The change is significant because AI assistants do not rely solely on a company’s website. Instead, they synthesise information from trusted media coverage, expert commentary, review platforms, business directories and brand-owned content. That, Ranjan said, means AI visibility is no longer just an SEO issue. It now requires public relations, content strategy, digital marketing, and analytics teams to work together.
His agency has also moved beyond analysing keywords to testing real consumer conversations. The same prompt often produces different answers across AI platforms because each model relies on different retrieval systems, training data and reasoning methods. Smaller but more authoritative brands sometimes outperform larger competitors, suggesting AI values credibility, clarity and relevance rather than simply marketing spend.
Although AI visibility is not yet a formal marketing key performance indicator, Ranjan said it has become a growing part of strategic conversations with clients. Agencies are increasingly helping brands build what he describes as “AI-ready ecosystems”, where websites, thought leadership, media coverage, product information and digital consistency collectively improve the likelihood of being recommended by AI assistants. Over time, he expects agencies to measure AI recommendation share alongside search visibility and traditional share of voice.
According to Ranjan, marketing budgets are likely to shift towards expert content, independent research, digital public relations and authoritative media coverage because these are the signals AI models rely on most. The greatest risk for brands, he argued, is not criticism but invisibility. If a company does not appear in an AI-generated shortlist, many consumers may never consider it in the first place.
From SEO to GEO: Promise or hype?
As AI assistants become an increasingly common starting point for consumer decisions, marketers are scrambling to understand whether brands can influence AI searches. The emerging discipline, known as Generative Engine Optimisation (GEO), aims to improve how AI models understand, cite and recommend companies. But while agencies are beginning to market GEO services aggressively, experts caution against viewing it as a shortcut to AI visibility.
Veera Ravindra Divi, an independent AI systems researcher, told Business Standard that companies can influence AI recommendations only indirectly. Unlike Google Search, where optimisation improves the likelihood of appearing higher in rankings, AI assistants generate responses by combining information from multiple trusted sources. The objective, therefore, is not to control the answer but to increase the probability of being included in it.
The strongest signals, according to Divi, come from credible third-party reviews, comparison articles, trusted publications, Wikipedia, discussion forums such as Reddit and well-structured factual content. Consistency in how a brand is described across the web also matters because AI systems look for corroboration across multiple independent sources before recommending a company.
“You can influence the probability of being recommended, not the outcome,” he said, warning that any agency promising guaranteed placement in AI answers is overstating what the technology can deliver.
Deepak Gupta, co-founder and technology expert at Style Lounge, believes marketers need to rethink what optimisation itself means. He cautioned that many of the claims surrounding GEO remain exaggerated and there is no formula that guarantees recommendations.
Shipra Jena, founder of AI-first communications firm PitchOne PR, told Business Standard that the biggest misconception is that brands can simply optimise their own websites to influence AI assistants. Research by Muck Rack’s Generative Pulse, which analysed more than 25 million AI citations, found that 84 per cent of citations came from earned media, while paid and advertorial content accounted for just 0.3 per cent.
Infographic created using ChatGPT
The new gatekeeper
The rise of AI assistants could reshape customer acquisition in ways that extend far beyond search rankings. The Business Standard analysis suggests that AI currently tends to favour established market leaders.
Divi argues that while AI naturally favours incumbents because they have richer digital footprints, challenger brands can outperform on highly specific or niche queries where they demonstrate stronger expertise and clearer relevance. The opportunity, he says, lies in the “long tail” of consumer questions rather than broad category searches.
Jaspreet Bindra, co-founder and chief executive officer of AI & Beyond, said AI could ultimately make competition more merit-based because these systems increasingly reward topical authority, expertise and trusted information rather than sheer marketing budgets.
