I audited my own product and found 10 things wrong. Here's every fix — Phase 1 + 2 done.
Two weeks ago I did something uncomfortable.
I stopped adding features and actually audited what I had built.
What I found was embarrassing. Not because the product didn't work — it did. But because basic trust and technical signals were broken. I was driving people to a landing page that was quietly failing them before they even understood what JewelViz does.
Here's every fix — honest and complete.
Phase 1 — Technical fixes:
→ Sitemap pointing to wrong domain (lovable.app instead of jewelviz.com) — Google was indexing the wrong URLs
→ Fake aggregateRating in JSON-LD schema — 4.8 stars with zero real reviews. Google policy violation. Removed.
→ Broken "Watch Demo" button — no video existed. Dead CTA on hero. Now scrolls to How It Works.
→ Meta title rewritten — from generic to conversion-focused: "AI Jewellery Photoshoot for Indian Jewellers — 60 Sec, ₹33/Photo"
→ Keyword meta trimmed — 10 keywords → 5. Less spam signal.
Phase 2 — Conversion fixes:
→ Real proof section — 6 actual AI-generated output images on homepage. The product was asking for trust without showing evidence.
→ WhatsApp floating button — my customers are small jewellers. They don't fill forms. They WhatsApp. Now they can reach me directly.
→ Mobile sticky CTA — "Start Free — 20 Images" always visible on mobile. 80% of my audience is on phone.
→ Hero carousel — 7 real showcase images. First impression is now proof, not copy.
What I learned:
Most of these were invisible to me because I built them. Fresh eyes — even my own after stepping back — caught things I had normalized.
The fake star rating was the most alarming. I had added it "just to see how it looks" during early development and completely forgot it was live.
If you haven't audited your own landing page recently — do it. Pretend you're a first-time visitor who has never heard of your product. You'll find things.
Current numbers:
— Paying customers: 0
— Website visitors: slowly growing
— Conversion rate: unknown (fixing tracking next)
Phase 3 next — city and jewelry-type SEO landing pages.
Building in public from Saharanpur, India 🇮🇳
jewelviz.com
I like that you audited your own product. Most founders spend so much time building that they rarely step back and experience the product like a stranger would.
One thing I've noticed with self-audits: you usually find the things you already suspected were wrong. The harder problems are the ones you don't see because you already know the product too well.
A new visitor doesn't have your context. They are silently asking:
"Is this for someone like me?"
"Why should I trust this?"
"What makes this better than what I'm already using?"
Those gaps are often hiding between a good product and a good conversion rate.
The fact that you found 10 things already means there are probably more opportunities in the outside perspective — especially around the first impression and messaging.
This kind of teardown work is one of my favorite parts of SaaS. Small clarity changes can sometimes have a bigger impact than adding another feature.
Thanks so much for the thoughtful feedback. I completely agree. Building the product every day makes it easy to miss things that a first-time visitor notices in seconds.
Your point about the silent questions every visitor asks really stood out to me. That's exactly the perspective I want to keep improving—making it immediately clear who JewelViz is for, why it's trustworthy, and why it's different.
Really appreciate you taking the time to share this. It gives me a lot to think about as I continue refining both the product and the messaging. 🙏
Your audit uncovered something many founders miss: fixing broken trust signals is only the first layer.
Looking at JewelViz from a first-time buyer perspective, the biggest opportunity I see is around the "value gap" between seeing AI-generated jewellery images and understanding why a jeweller should pay for them.
The visitor may like the output, but they still need the bridge:
"How does this help me sell more, launch designs faster, or reduce photography costs?"
That connection between product capability and business outcome is often where good products lose conversions.
Your technical fixes are a great foundation. The next interesting step would be pressure-testing the homepage from a customer's buying perspective — the questions, doubts, and objections a jeweller has before trying it.
Would be interesting to see how the next iteration performs.
That's a really thoughtful observation. I completely agree—the outputs alone aren't enough. The real challenge is helping jewellers immediately understand how those images translate into faster launches, lower photography costs, and ultimately more sales. That's exactly what I'm working on in the next iteration of JewelViz. Thanks for taking the time to share such detailed feedback
One small thing I'd pressure-test in the next iteration:
Your homepage does a good job explaining what JewelViz does, but I'd also look at whether a jeweller can immediately recognize themselves in the first few seconds.
For example, are you speaking to someone launching new collections every week, a custom jewellery store, or a retailer trying to reduce photography costs? The more specific the first impression feels, the easier it is for visitors to think, "This was built for a business like mine."
It's a subtle messaging shift, but it often has a big impact on conversions.
Thank you so much for taking the time to share such thoughtful feedback. I really appreciate it. You're absolutely right—while I've focused on explaining what JewelViz does, I now realize the homepage should immediately make a jeweller feel, "This is built for businesses like mine." I'll definitely refine the messaging to speak more directly to specific jewellery businesses and their workflows. Thanks again for helping me improve the product. It means a lot! 😊
What stood out to me is that none of these are “growth tweaks”—they’re trust repairs. Most founders optimize for traffic first, but you did the harder thing: fixed the signals that decide whether traffic should convert at all. Especially the fake schema rating—that’s the kind of silent credibility leak that can distort everything downstream without you noticing.
Thanks Aryan, that distinction actually reframes it well for me too — I was calling these 'fixes' in my head, but you're right, most of them were removing things that were actively working against trust, not adding anything new. The fake schema rating especially — I built it during early testing and just... forgot to remove it. Scary how something like that sits invisible until someone points it out. Appreciate you catching the nuance here
Glad it resonated.
Your reply made me think there's one strategic implication of that shift which becomes much more important as the product evolves, but I don't think I can explain the reasoning properly in a thread without oversimplifying it.
If you're interested, what's the best email to reach you on?
Appreciate that — happy to keep the conversation going. Could you give me a quick high-level sense of what the implication is? Even a rough version helps me figure out if it's something to dig into over email or if we can just hash it out here
Fair question.
The short version is that I don't think the implication is about those individual changes at all.
It's about the strategic decision they quietly point to as your product evolves.
I don't think I can explain that properly in a few lines without oversimplifying it. If it resonates, I'm happy to walk you through my reasoning over email.
No worries, I'll pass on the email — happy to keep discussing here if you want to share more anytime."
This comment was deleted 2 months ago