- Nearly 80% of enterprise data is unstructured, with over half classified as dark data
- AI systems struggle without accessible data scattered across many legacy and software systems
- India hosts 2,117 Global Capability Centres generating $98.4 billion, many AI-first operations
Every time you shop online, send an email, make a payment or chat with customer support, you create data. Businesses spend billions of dollars collecting that information. Yet a surprisingly large share of it is never used.
This hidden pile of unused information is now emerging as one of the biggest roadblocks in the race to adopt artificial intelligence (AI). Experts say it is quietly costing companies millions in lost opportunities, slower decisions and lower productivity.
A Business Reporter analysis estimates that nearly 80 per cent of enterprise data is unstructured. More than half of it is classified as "dark data" -- information that organisations have collected but rarely analyse or use.
In other words, companies are sitting on a treasure chest. The problem is that they cannot find the key. This problem is even more pronounced in India.
'AI Can't Deliver Results Without Data'
As businesses rush to deploy AI, many are discovering an uncomfortable truth. Even the smartest AI systems cannot deliver meaningful results if the data they need is scattered across hundreds of software applications, trapped in legacy systems or buried inside documents that nobody can easily access.
"The multinational that has gathered petabytes across two hundred systems has not collected gold. It has collected ore. And ore, however abundant, pays no dividends," said Chandan Mishra, Vice President of Marketing and Sales at SCIKIQ.
According to Mishra, the solution is not ripping out old software and starting from scratch. Instead, companies should connect their existing systems so both employees and AI tools can search information through simple, natural-language questions.
Experts say the challenge for Indian companies is no longer about collecting more data. It is about making existing data usable.
"The most successful AI systems are not necessarily those with the largest datasets. They are the ones with the best context," said Ashish Chandra, a global AI expert.
He believes organisations need a common knowledge layer that connects information across departments instead of leaving it trapped in separate databases. Without that context, even advanced AI models struggle to produce reliable answers.
The issue is becoming increasingly important for India's rapidly expanding Global Capability Centre (GCC) ecosystem.
'India's GCC Boom Depends On Data'
According to the latest Nasscom-Zinnov GCC Value Orbit Report, India is home to 2,117 Global Capability Centres that generate $98.4 billion in annual revenue. The sector has expanded 32 per cent since FY2021.
The report also found that 506 companies from the Forbes Global 2000 list now operate GCCs in India. Nearly half of the centres established since FY2021 have been designed as AI-first operations, underlining India's growing role in building enterprise AI solutions.
Chandra believes this shift will redefine India's competitive advantage. "The next generation of GCCs will not compete on labour arbitrage. They will compete on intelligence arbitrage," he said.
Industry experts say companies are beginning to realise that competitive advantage will come from extracting more value from existing information rather than endlessly collecting new data.
"AI has brought a long-standing enterprise challenge into sharper focus. As organisations accelerate their AI investments, the conversation must shift from data collection to data activation," said Ayush Sarvaiya, Co-founder of Plus91Labs.
He said the winners in the AI era will not necessarily be businesses with the biggest databases. Instead, they will be the ones that can turn information into faster decisions, better customer insights and measurable business value.
The urgency is growing as companies pour billions into enterprise AI. Industry estimates value the global AI agent market at anywhere between $1.81 trillion and more than $3.5 trillion by 2030.
Ramya Chatterjee, CEO of Solitaire Brand Business and Director at Prointek Global Innovations, believes many organisations still face the same problem.
"Organisations are data-rich but insight-poor," she said. "The most are no longer asking how much data they possess. Instead, they are asking how effectively they can activate that data to generate business value."
