The numbers come first. Hitachi, ranked No. 197 on the Fortune Global 500 with $70 billion in annual revenue and nearly 290,000 employees worldwide, has achieved a 40% efficiency gain for its sales and marketing team and a 20% reduction in operating costs after deploying a data-fabric solution built on enterprise software vendor Appian. No legacy database was replaced. No massive migration was required.
Bala Krishnapillai, senior vice president and chief information officer of Hitachi's Americas division, inherited a data problem that was years — arguably a century — in the making. Founded in 1910 as a single mining machinery repair shop, Hitachi grew into a sprawling conglomerate spanning rail systems, digital products, industrial machinery, power and renewable energy, and medical systems. That growth came largely through acquisitions, including an $11 billion purchase of ABB's power grids business and a $9.6 billion deal for U.S. software vendor GlobalLogic.
The price of that dealmaking history: data scattered across more than 150 different customer relationship management systems, including Salesforce, SAP, and Microsoft platforms. When sales and marketing teams needed to build a new business proposal, it could take weeks to produce an accurate analysis report. 'We have massive data stored in our ecosystem,' Krishnapillai said. 'It was unmanageable from an IT enterprise standpoint.'
The Appian platform connected data on top of the existing legacy infrastructure rather than forcing consolidation into a single database. The result, according to both companies, is faster project creation and sharper analytical output. 'When we create a proposal now, it becomes stronger, more compelling, and very competitive,' Krishnapillai said.
Appian CEO Matt Calkins argues the data-fabric architecture also positions Hitachi for the next wave of AI deployment — autonomous, agentic systems that explore enterprise data in unpredictable ways. 'They'll be like humans, exploring the enterprise and making decisions,' Calkins said, 'and so they need to go places' that traditional siloed infrastructure cannot support.
On the broader AI front, Krishnapillai has deliberately avoided deploying a single enterprise-wide AI tool across all 607 subsidiaries operating in 190 global markets. His strategy organizes adoption into three tiers: everyday productivity tools such as Microsoft Copilot and Google Gemini; job-specific applications vetted by IT and business leaders together, including AI-enabled content creation and competitive analysis; and developer-focused coding assistants, where Hitachi works closely with Anthropic. 'From the enterprise AI strategy standpoint, there is no one solution,' Krishnapillai said. Token consumption is actively monitored and controlled, though divisions like research and development face no usage restrictions given their competitive sensitivity.
CEO Times take: Hitachi's approach is a quiet rebuke to the AI maximalism that has burned through corporate budgets with little to show for it. Research consistently shows a high failure rate for enterprise AI pilots, and cost overruns have forced even the largest employers to scrutinize token spend. What Krishnapillai has demonstrated is that disciplined architecture — connecting what already exists rather than replacing it — can deliver measurable margin improvement before a single autonomous agent is ever deployed. Capital rewards clear rules, and in this case, clear data. The 20% cost reduction did not require a blank check or a moonshot. It required an engineer who understood the ledger.
