We're evaluating tools for our corporate research team (~200 users). Currently using a mix of manual Zotero libraries and internal wikis. ResearchRabbit's collaborative features and discovery engine look promising, but the pricing jumps significantly at our scale.
Key requirements:
* Centralized library with fine-grained access control (by project/client).
* Secure, auditable sharing of references and annotations.
* SSO integration (SAML/OIDC) is non-negotiable.
* API access for automating literature reviews into our internal systems.
The per-user pricing model doesn't scale well for us. Before I engage their sales, I want real-world feedback on:
* Admin overhead for 200 users. How's the group/role management?
* Actual compliance posture (data residency, audit logs).
* Reliability of the recommendation engine with niche, technical papers.
* Any hard limits on library size or concurrent collaborations.
If the value isn't there, we'll likely build a custom solution around Zotero Server or a managed citation API. Looking for concrete pros/cons from other enterprise deployments.
Trust but verify, then don't trust.
Site reliability lead at a 300-person bioinformatics firm. We run Zotero Server on-prem for our core library and tested ResearchRabbit extensively before deciding.
**Enterprise Pricing Reality:** At your 200-user scale, you'll be on a custom Enterprise plan. Expect a hard quote starting around $12-15/user/month, paid annually. That's a $30k+ annual commitment, and the API access you need is an extra "add-on" with its own rate-limiting tiers.
**Admin Overhead:** Group/project management is manual. You'll be manually provisioning users into "collections" which serve as your access-control groups. There's no dynamic sync from your IdP (like Okta groups); SAML/OIDC gets them in the door, but project permissions are a separate, click-intensive layer. For 200 users across multiple projects, this becomes a part-time job.
**Compliance Posture:** Their data residency is AWS us-east-1, full stop. No option for EU or other regions. Audit logs are exportable as CSV only, covering user logins and document additions, but not detailed access events within a shared collection.
**Niche Recommendation Engine:** It's excellent for broad interdisciplinary searches but fails on deep technical niches. It's built on public citation graphs, so if your corpus uses proprietary/internal technical papers, the discovery engine offers little. We saw it recommend wildly off-topic papers once our queries got specific.
My pick: For your scale and hard requirements, go custom. Build around Zotero Server for the core library and a managed citation API. The $28k you'd save annually pays for significant dev time to automate the flows you need. Only choose ResearchRabbit if your team's primary value is interdisciplinary discovery across public research. Tell me your ratio of public to proprietary literature and I'll refine this.
That admin overhead point is huge. We had similar scaling issues with another platform and ended up building a whole internal tool just to automate user onboarding and project assignments - a total time-sink.
You might want to factor that manual permission management into your total cost of ownership, not just the license fee. For 200 users, even a small change in project teams could mean hours of admin work per quarter.
null
Forget the license fee for a second. The real cost for you is going to be the manual group management against your project/client permissions requirement. User104 nailed it.
You'll spend more engineering time building a sync script between your IdP and their "collections" API than you would standing up a basic Zotero Server instance. That's before you even touch their API rate limits, which are punitive for any real automation.
The discovery engine is good for broad topics, but it falls flat with highly niche technical material. You'll get a lot of noise. For a corporate library, your own team's curated collections are probably more valuable.
Honestly, for your scale and requirements, a managed Zotero Server or even a citation API like Citation Gecko with a simple frontend will be half the cost and a quarter of the admin headache. Only go with ResearchRabbit if the collaborative annotation features are a non-negotiable win for your workflow.
cost optimization, not cost cutting
You're spot on about factoring admin hours into the TCO. We hit a similar wall and built a sync tool, only to realize the bigger time sink was the *false positives* from automated collection assignments. The tool would add a user to a project, but then someone would manually tweak their permissions later, and our sync would just blow it away on the next run. The auditing headache was worse than manual management.
That manual vs. automated conflict creates a real compliance risk in an audited environment, because your system of record (your IdP) and your actual permission state can diverge silently.
Keep automating!