The real estate industry is rushing to embrace artificial intelligence. In summer 2025, Morgan Stanley estimated that 37% of processes in commercial real estate could be automated through AI, saving the industry $34 billion. According to JLL’s 2025 Global Real Estate Technology Survey, 88% of real estate investors are already piloting AI, up from just 5% in 2023.
Yet early results are sobering. Research from MIT shows that a large proportion of generative AI pilots fail to deliver tangible P&L improvements, while Gartner predicts over 40% of agentic AI projects may be cancelled by 2027 – more often than not, the reason comes down to data.
Real estate’s structural complexity
Unlike vertically integrated industries, real estate operates through a distributed ecosystem. Property managers, fund administrators and other service providers each contribute data to the same underlying assets, often using disparate systems, inconsistent naming conventions and varied accounting structures, whether in Excel, PDFs or other fragmented formats.
As trial balances and general ledger data flow from property managers to fund accountants, granular details get consolidated into broad cost categories. By the time information reaches the investor, critical distinctions such as rental income split by tenant or associated service charges are no longer visible.
As a result, investors often set up separate, manual and error-prone data collection processes, pulling together rent rolls, charts of accounts and arrears directly from property managers -just to produce the polished investor reports that make it all look seamless.
Beyond standards: The case for data infrastructure
The industry has responded by advocating for better data standards – common definitions for charts of accounts, lease structures, ESG metrics and more – in an attempt to bring consistency to this fragmented landscape and increase interoperability and data exchange. However, the challenge runs deeper than standards alone: it requires technology that can collect, map and standardise data across a fragmented ecosystem.
In light of this, real estate asset managers are proactively seeking systems that automate data collection end to end: structured onboarding portals for scheduled submissions, automated notifications, robust validation rules that catch errors at source, and a single, versioned, auditable dataset that the entire chain can trust.
By centralising this flow into a purpose-built real estate data platform where data is harmonised, versioned and always available to downstream processes, organisations create a reliable foundation for meaningful analysis, such as investor and fund level reporting and budget versus actuals.
What this means for asset and investment managers
In an industry increasingly looking to technology as a source of competitive advantage, a solid data foundation is not a back-office concern. It is a strategic one. Firms that invest in getting this right will be better positioned to move faster, report with confidence, and unlock the analytical capabilities, including AI, that will define the next generation of asset management.
Julian Vogel is co-CEO and founder of Tower360, a data and workflow infrastructure platform for institutional real estate asset management.
