The Slow Hollowing of PeopleSoft: What I'm Seeing in the Field

PeopleSoft is dying. Everyone knows it. But here’s what they don’t tell you: it’s going to take a very long time to actually die.

Replacing an ERP system takes three to five years minimum—sometimes longer. Some organizations never pull it off at all. So what I’m witnessing across multiple clients isn’t a wholesale rip-and-replace. It’s something more interesting: a gradual hollowing out, where organizations chip away at PeopleSoft functionality by bolting on modern point solutions and integrating them back in.

I have a unique vantage point here. I’m delivering vendor API integrations into PeopleSoft systems, which means I see firsthand how organizations are nibbling at the edges of their monolith.

The Talent Crisis Nobody Talks About

It takes years to become a competent PeopleSoft technical resource. Functional folks ramp up faster, but the really good ones have technical chops too—and they built that expertise through implementations and upgrades. Those projects barely happen anymore. The learning environment has essentially collapsed.

Meanwhile, the experienced people are leaving. Some are retiring. Some, unfortunately, have passed away. I’ve lost several colleagues who were excellent PeopleSoft resources. The net effect: you’ve got this massive ERP system at the center of your organization, supported by shrinking teams with declining experience levels.

There are twenty ways to do anything in PeopleSoft. Eighteen of them are subpar. Knowing which two are optimal takes project experience that fewer and fewer people have.

The Grass Isn’t Greener

Everyone thinks their current system sucks. I have news: the new system will also suck. It might solve a few pain points. The UI might be prettier. The reporting might be marginally better. But it will bring its own problems.

Take Workday Student. I’ve worked with Workday HR, and honestly? It already looks like legacy software to me. Don’t get me started on the reporting limitations and lack of SQL support. They decided to reinvent the wheel there, apparently ignoring the Lindy effect that’s kept SQL relevant for fifty years. But that’s another post.

I am not sure if Oracle will ever build a true PeopleSoft replacement. If they do, it will take years and probably be expensive, and ugly, and a Frankenstein system of bolted-on modules. If they don’t, competitors will try to fill the gap, but those systems will also have their own issues.

The Cambrian Explosion of Point Solutions

Here’s the pattern I keep seeing: an IT director or stakeholder discovers some vendor product. It’s modern. The sales team is slick. The demo is impressive. It has “AI” in the name somewhere. It might even deliver real value.

But then comes the integration.

These vendors provide granular APIs for their own systems sometimes. What they don’t provide is any enterprise-level orchestration. That’s your problem. So now instead of a few monolithic systems, you’ve got dozens—maybe hundreds—of disparate applications that all need to talk to each other.

This is where something like Temporal could help with orchestration. But most organizations aren’t there yet. They’re just trying to stay stable. I have seen clients try to use Mulesoft, Boomi, and other iPaaS solutions, but the complexity often overwhelms their teams.

When you start seeing how effective vibe-coding is getting starting in late 2025, you might find that building these integrations becomes easier but you may also see an explosion of new point solutions that need to be integrated. I was a doubter until around 2025-October when Anthropic and others made significant strides. Now, I can see vibe-coding being a real game-changer for integration tasks and creating new applications. I have personally used AI to write several new applications outside of PeopleSoft in just weeks, something that would have taken months or years before.

The Path Forward: Integration as Core Competency

Whatever your long-term ERP strategy, your PeopleSoft team needs to master integration. Not flat files—real-time web services. You need to:

• Push data out to modern systems (student records, employee data) in near real-time
• Accept data back from those systems cleanly
• Expose PeopleSoft functionality through proper APIs

This is where the industry is headed. Eventually something will replace PeopleSoft—maybe Oracle’s next thing, maybe a competitor. But in the meantime, the organizations that survive will be the ones who treat integration as a first-class capability, not an afterthought.