CUNY paid Oracle $600M for its HR software (2013)

Original Article ↗ Hacker News Discussion ↗

New York’s public university system CUNY is criticized for allegedly spending hundreds of millions of dollars on an Oracle/PeopleSoft-based HR and student system that many users find unintuitive, inflexible, and wasteful. Commenters debate whether such eye‑watering figures are even accurate, how much is vendor cost versus padded institutional overhead, and why large public institutions repeatedly buy feature-bloated enterprise software that poorly fits their needs. The thread broadens into a critique of academic and government procurement, the incentives that favor big vendors like Oracle, and whether organizations should adapt their processes to off‑the‑shelf tools or invest in custom or open‑source alternatives.

Cost and scale of the CUNYfirst deal

  • The oft-cited $600M Oracle figure is heavily disputed in the thread.
  • Several participants with higher-ed/ERP experience argue that a $600M single software deal is implausible relative to CUNY’s total budget and typical higher-ed pricing.
  • Others dig into CUNY budget and tender documents and conclude it looks more like ~$300M total over ~10 years across ~26 institutions, including HR, ERP, compute, and staff, i.e. ~$30M/year system-wide, with only a fraction going to Oracle itself.
  • Some note that universities and agencies routinely pad multi‑year IT budget requests by 3–5x to cover staffing, risk, and under‑funding, which can make quoted numbers look inflated or opaque.

Enterprise / government procurement dynamics

  • Many comments describe large organizations (academia, government, Fortune 50) as structurally prone to waste, dysfunction, and “no one gets fired for buying big-name vendor” thinking.
  • Incentive problems are emphasized: people spending others’ money, resume-padding, budget-protection, “lowest bidder” rules, and occasional hints of corruption or “incentives.”
  • Vendors like Oracle, Deloitte, IBM, etc., are framed as specialists at extracting money from such environments.

Oracle / PeopleSoft and software quality

  • Multiple anecdotes depict Oracle/PeopleSoft systems as archaic, unintuitive, and painful to implement and use, sometimes never working properly despite huge spend.
  • CUNYfirst is described as visually and functionally obsolete and process-worsening, with “configure only” constraints forcing CUNY to change course numbering and workflows to fit the software.

Build vs buy vs open source

  • Several argue that the same money could have funded a greenfield system or an in‑house/contractor build with far fewer people and better UX.
  • Others counter that selling into higher-ed is so painful and “boring” that only vendors like Oracle tolerate it, and sustaining custom systems is non‑trivial.
  • There is recurring support for government‑funded open source platforms that many institutions could share, but skepticism about political feasibility and governance.

Process change vs customization

  • One camp insists organizations should adapt processes to standard ERP workflows to avoid endless costly customization.
  • Another camp reports repeated failure and poor UX when shoehorning unique or complex processes (universities, healthcare, payroll) into rigid off‑the‑shelf systems.
  • Consensus: customizing ERPs is risky and expensive, but blindly forcing all processes to fit the tool can also be disastrous.

Broader critiques of academia and bureaucracy

  • Several participants generalize from CUNY: entrenched inefficiency, weak accountability, difficulty firing underperformers, and bloated internal IT or consulting layers.
  • Others push back that similar or worse dysfunction exists in large private companies; incompetence and misaligned incentives are seen as systemic rather than sector‑specific.