Hey everyone! I'm diving into Iris.ai for the first time and feeling a bit overwhelmed. The workspace looks super powerful, but I don't want to waste my first hour clicking around aimlessly.
I'm coming from a martech/analytics background, so my brain naturally wants to compare features and map out workflows. For those of you who are seasoned users:
* What's the **one core workflow** I should set up first to get a real "aha" moment?
* Is it better to start with feeding it my own research PDFs, or should I play with the tool's own discovery features?
* Any key settings or filters I should configure right away to avoid noise?
I learn best by doing, but a nudge in the right direction would save me so much time. My end goal is to use it for customer journey research and competitive analysis, but I want to nail the basics first.
— alex
Data > opinions
Hey alex, I'm actually in the same boat as you, started with Iris.ai last week. For that "aha" moment, I'd say skip your own PDFs at first and use the Discover tool with a really specific query related to your goal, like "customer journey mapping touchpoints 2023". That gave me a concrete list of papers to see how it connects concepts.
I made the mistake of not setting filters early and got a ton of outdated stuff. Definitely click that "Publication Year" filter right away and maybe narrow the "Document Type" to reviews or articles, not patents, unless you need those.
Question for you, since you're from analytics: did you find the data visualization in the workspace intuitive for comparing papers? I'm still figuring that part out.
Learning by breaking
I think you're spot on about the filters - the year filter is a lifesaver. I'd add the "Review Articles" filter to that list for a new user. Starting with a well-structured review paper from the Discover results gives you a solid foundation of key concepts before you dive into narrower primary research.
On the data viz, I didn't find it immediately intuitive for comparison either. It's powerful, but the learning curve is real. I found it clicked after I used the "Extract Data" tool on a few papers first to create a structured table. Then the visualizations built from that table made a lot more sense.
Keep it constructive.
Absolutely agree about the structured table being a prerequisite for the visualization. The system needs that clean, extracted data layer before the charts can generate anything meaningful. It's a two-step workflow that isn't immediately obvious.
I'd add that for competitive analysis, the "Extract Data" tool works best if you pre-define your own custom columns - like "Methodology", "Sample Size", "Key Finding" - rather than relying on the auto-suggested ones. Feeding it those column headers forces a consistent structure across papers, making the subsequent comparison charts actually useful instead of just pretty.