I'm evaluating research tools for my team, and Iris.ai's pricing page mentions projects but isn't specific about concurrency limits. Before we commit, I need to understand the operational ceiling.
From a platform perspective, a "project" likely maps to a distinct workspace with its own document sets, saved searches, and AI model contexts. The practical limit for a single license won't be just a number—it's about system resources and user workflow. Based on scaling similar SaaS tools, I'd expect bottlenecks in:
* Active AI researcher contexts (memory/CPU on their backend)
* Indexed document volume per project
* Simultaneous user sessions within a project
Has anyone stress-tested this? For example, if my team of 5 needs to juggle:
* A long-term literature review for compliance (Project A)
* A fast-paced competitive intelligence sprint (Project B)
* An internal knowledge base curation effort (Project C)
Can a single "Company" license handle all three being actively used, edited, and analyzed in parallel daily? Or would we see performance degradation or need to archive/disable one to work on another effectively?
Sharing your actual experience with multiple concurrent projects—especially if you've hit limits—would help us architect our research workflows correctly.
Great breakdown of the bottlenecks. I was looking at this for my last role and found the real limit often wasn't the number of projects, but the data churn within them.
> A long-term literature review for compliance (Project A)
This one, with heavy doc uploads, could be the anchor that slows the others if you're all working at once. For your team of five, you'd probably be okay with three active projects, but you might feel lag during peak analysis in Project A.
I ended up asking their support for a sample resource usage report. Might be worth a shot?
MartechStruggles