Naoma AI Demo Agent - The video AI demo agent for B2B SaaS for immediate demos

@wasil_abdal Really glad the experience landed well — and great question, this is one we get a lot.

The short answer: yes. Naoma can pull from your product documentation, knowledge base, and custom content we configure during onboarding. So when a prospect asks something specific or off-script, it's not guessing — it's drawing from the actual source material your team has built.

The depth depends on what we load in during setup. Our team works closely with each customer to map out the key scenarios, edge cases, and technical questions their ICP typically asks. The more thorough that foundation, the better the demo handles complexity.

That said — we're honest about the limits. For truly unusual edge cases or highly technical architecture questions, Naoma is designed to recognize when a human needs to step in and route accordingly. The goal isn't to replace that conversation, it's to make sure it happens with a prospect who's already warm and informed.

Would love to dig into your specific use case if you want to pressure-test it further. 🙏

@wasil_abdal Thank you!

@wasil_abdal yes, we support uploading the knowledge base and other documentation

@wasil_abdal really glad the experience landed well, and great question on the knowledge depth.

Naoma pulls from a knowledge base we configure during setup - product documentation, FAQs, common objections, pricing, edge cases. The more context you give it, the sharper it gets. It’s less about a “standard script” and more about how well you’ve briefed it, similar to onboarding a new rep.

For real-time dynamic data: live pricing changes, feature flags, account-specific info - that requires integration with your existing systems, which we support.

And when it hits a genuinely complex edge case it can’t answer confidently? It hands off gracefully rather than guessing. In our experience that moment, done well, actually increases trust rather than breaking it.

Would love to hear what is your specific use case.

@joao_seabra Thank you, and great questions, these go straight to the core of what we've been working through.

On the sample: the 10-20% is from early pilots across a handful of B2B SaaS companies - small sample, so we're careful not to overgeneralize. But the directional signal is consistent: when you remove the scheduling friction and replace it with an instant conversation, a meaningful share of visitors who would have bounced actually engage.

On handling unknowns I fully agree. A confident hallucination is worse than an honest handoff. Naoma is designed to recognize the boundary and route gracefully to a human when it hits it. That moment, done well, actually builds trust rather than breaking it.

On early-stage products: Naoma works best when there's enough product to show, but it doesn't require a polished, static flow. When features change, you update the agent. It's closer to briefing a new rep than rebuilding a product tour.

@joao_seabra Thank you!

@joao_seabra Joao, thank you for your support!

@mauricevv thank you! Wanna try it?

@mauricevv Many thanks!

@mauricevv thank you

@mauricevv thanks! Impressive indeed!

@alena_korpula Thanks so much for the feedback!

@alena_korpula thank you for your trust and support!

@alena_korpula thanks a lot for your support

@alena_korpula thanks for the feedback!

@mehmet_kerem_mutlu Thank you!

@mehmet_kerem_mutlu Great question — and one we track closely.

We don't share specific metrics publicly, but here's what we can say:

There is a bounce rate — some prospects drop off in the first 30 seconds. But that number is consistently lower than the average no-show rate for traditional scheduled demos. A prospect who never shows up costs you just as much as one who clicks away in 20 seconds — except with the AI demo you at least got them in the door.

For prospects who engage past that point, average session length runs over 6 minutes depending on the product. That's a real, active conversation — not passive video watching.

And to your point about drop-offs — we actually built our pricing around this. We only charge for demos over 3 minutes. If someone bounces early, it doesn't count. We think that's the only fair way to do it.

@mehmet_kerem_mutlu thanks

@mehmet_kerem_mutlu thanks for the comment and questions! See that Dmitry has already addressed this

@islam_midov hi Islam and thank you for your question. Get demo now and if U'd like - book a demo with our sales team. We'll show you our pricing.

@islam_midov Thank you!

@islam_midov thanks for the support - to be honest, we are still trying to figure it out. But the target price is $10 per demo from 3+ minutes

@islam_midov Many thanks!

@jozsef_orsos thank you!

@jozsef_orsos we connect a client's knowledge base. So Naoma can successfully handle even difficult technical questions.

@jozsef_orsos Naoma can use the knowledge base and sales playbooks to handle objections the same way the human sales-rep would

@jozsef_orsos thanks! It handles all the questions it has in knowledge base. If someone is trying to simply goof around or jailbreak it, the agent will return them back to the topic - we implemented some guardrails for that.

@davitausberlin Yes, good points! Thank you!

@davitausberlin Thank you!

@davitausberlin, fair points on both - appreciate the candid feedback.

On languages: you're right to flag it. To be precise, Naoma currently handles 33 languages well. Beyond that we'd rather be honest than overpromise.

On the human touch tradeoff: I'd slightly reframe it. It's not AI demo now vs human demo later - it's AI demo now vs no conversation at all for most visitors. The majority of people who hit "Book a Demo" never show up to the call. Naoma captures that lost conversation, not replaces the human one. The rep still handles the high-value stage.

And yes, if you're drowning in volume it's an obvious fit. But the more interesting case is actually the opposite: companies with decent traffic and almost no inbound pipeline, because the friction is killing intent before it ever reaches a human.

Thanks for the thoughtful pushback - this is exactly the conversation worth having.

@davitausberlin Thank you!

@thegoodb0rg hi and thank you for your question. It's a usage based model with monthly packages

@thegoodb0rg thank you. We have different pricing options, I guess it would be better to discuss this with @dzmitry_ivanouski

@thegoodb0rg happy to chat in details on how we can tailor our agent for your products

@thegoodb0rg Thank you!

@dzianis_yatsenka Hi, thank you. We support 30+ languages, including English, Spanish, French, German, Portuguese, Hindi, Japanese, Korean, Chinese, Arabic, and more

@dzianis_yatsenka Thank you!

@dzianis_yatsenka thanks for your great question, Naoma supports 30+ languages.

@dzianis_yatsenka thanks for the support, first of all!
I was thinking if we can collaborate on the data from non-native English speakers? Just an idea - saying it out loud :-)