Can I get feedback on my landing page and Demo flow?

Hey Dumebi,

Your concern about Banned Lexicon not being a known term is exactly why it SHOULD lead the page. Unknown terms aren't a weakness, they're a positioning opportunity. When HubSpot coined inbound marketing or Drift coined conversational marketing, nobody knew those terms either. The landing page's job is to educate.

What I'd change:

1. Hero: Lead with a side-by-side comparison. ChatGPT output with delve, tapestry, embark highlighted in red vs Ozigi output clean. This proves the value in 3 seconds, no jargon needed.

2. Subhead: Powered by Banned Lexicon. We block 200+ AI-telltale words at generation, not after the fact. This frames it as technology, not buzzword.

3. Replace the placeholder logos. Fake company names kill trust faster than showing nothing. Either use real customer names with titles, or swap for 100+ founders use Ozigi with real Peerlist avatars.

4. The live demo is great, but it's doing too much work. Add a 15-second explainer GIF or short video above it showing: paste raw notes, get 3 polished outputs for LinkedIn, newsletter, X thread.

The product is clearly good. The page just needs to prove it visually instead of describing it with words.

I run an AI-powered landing page studio. If you want, I'll generate a free sample hero section for Ozigi showing this comparison layout. No strings, just curious if the AI output would be useful to you. 48h turnaround.

Either way, fix the hero hierarchy and the signups will follow.

The product is stronger than the current landing page frame. “AI content that sounds human” is clear, but it still puts Ozigi in the same crowded bucket as every anti-ChatGPT writing tool. The sharper angle is that you’re turning raw context into multi-channel content in one workflow while preserving voice.

The live demo is a good move because people can test the output fast. I’d make the first screen prove the before-and-after quicker: raw notes or URL in, LinkedIn/X/newsletter output out, with the “your voice, not AI voice” difference visible immediately.

One thing I’d watch is the Ozigi name. It is short, but with the .app extension and AI writing category, it may feel more like a side tool than a serious content operating system. If this moves toward teams, personas, newsletters, and publishing workflows, Beryxa.com would give it a cleaner SaaS brand.

If people reach the landing page but don’t continue into the demo, I’d look for a mismatch between the hero promise and what visitors expect the demo to prove.

The page probably needs to make the demo feel like the obvious next step, not another commitment.

A few things I’d check:

• Does the headline make the outcome concrete enough?
• Is the demo CTA framed around value, not effort?
• Does the page explain what the user will see in the demo?
• Are there enough trust signals before the demo CTA?
• Does the page handle the “why not just use ChatGPT?” objection?

I’m building a free AI landing page diagnosis tool that checks headline clarity, CTA friction, trust, and buyer objections. Happy to run your page through it and share what it finds.

Hey — feedback on the LP since you asked specifically.

Your headline does three things: "Find Leads, Run Outreach, Publish Content." Your reviews do one thing: every single testimonial is about how the writing sounds human, not like AI. Every quote mentions human-like output. None mention leads. None mention outreach.

So visitors read three claims and your own users only vouch for one. The headline argues with the social proof.

Framing fix (no redesign):

Before: "Ozigi — Find Leads, Run Outreach, Publish Content"
After: "AI content that sounds like you wrote it — not a bot"

Then position leads and outreach as secondary benefits of that core promise. The testimonials will actually support the headline.

Second thing while I'm here: every testimonial on the page is labeled "Peerlist reviewer" — not a single paying customer quoted anywhere. One line from someone who paid you, describing a specific outcome they got, outweighs eight reviewer quotes. Even one real customer sentence above the fold would shift how the social proof reads.

Both are copy changes. If either lands, I'd love to hear.

The demo flow question is often more important than the landing page
itself — most SaaS tools lose people between "I'm interested" and
"I understand what this does."

One thing worth checking: does your landing page set the right
expectation for what the demo will feel like? If the page promises
simplicity but the demo shows complexity, that gap kills conversions
even if both are individually good.

What's the drop-off point you're seeing — are people leaving the
landing page before clicking demo, or starting the demo and not
finishing?

Nice work on the "Banned Lexicon" concept - that’s a much sharper technical differentiator than just saying "human voice."

I think you're right that people don't know the term yet, but that's actually why it's a powerful hook. You don't sell the name; you sell the problem it solves.

The problem isn't "I want a human voice" (that's the desired outcome). The problem is "ChatGPT uses words like 'delve', 'tapestry', and 'embark' that make me look like a bot."

If you show a live comparison of "Standard AI" vs "Ozigi (Lexicon-Cleaned)" right in the hero, you prove the value without needing them to understand the jargon first.

Also, +1 on the logo trust leak. If those are real teams, add their specific roles or real avatars. If they’re placeholders, better to remove them or show real user avatars from your Peerlist reviews instead.

I actually did a deeper dive into your above-the-fold hierarchy (the newsletter/blog/social split below the hero is strong, but the pricing jump from $0 to $15 might be scaring off solo creators). If you want the full breakdown of the 3 biggest "Trust Leaks" on the current page: https://roastmysite.io/go.php?src=external_manual_ih_ozigi_lexiconhook_may19_usd_presell_hv

Good product, the landing's just selling it wrong.

"AI content that sounds human" is everywhere now. Undetectable.ai, WriteHuman, StealthGPT, half the newsletter tools all say this. The phrase doesn't differentiate anymore.

Your real wedge is hiding in the footer: Banned Lexicon. Most "humanize AI" tools detect AI patterns and rewrite after the fact. Yours bans them at generation. Different mechanism, and it should be the hero, not buried in feature notes.

Also — the logo strip ("meridian, STACKBASE, Cortex"). Are those real customers? Names look like placeholder brands, and that kills trust faster than no logos at all.

Pricing's weird too. Free → $15 → $39, but $15 unlocks just enough to feel limited. I'd hit free limits and skip to $39, or churn.

Banned Lexicon is your moat. Put it on top.