I Built a Voice OS for Websites — 0 Signups After 20,000 SEO Pages. Here's My Pivot.

When I launched AnveVoice last year, I was convinced the numbers would speak for themselves. A voice operating system for websites. One script tag embed. Agentic DOM actions that navigate, fill forms, click buttons. 53 languages. <700ms latency. It checked every box of what I thought the market wanted.

Six months in, I had 20,253 SEO-generated pages indexed. My content was ranking. The visits were coming in. But the signups? Zero.

Not "low." Not "disappointing." Literally zero paying customers.

This is that story.

The Setup: Why I Built This

I started AnveVoice because I saw a massive gap in web accessibility and user experience. Most websites are hostile to voice interaction. They require clicking tiny buttons, reading dense text, navigating convoluted menus. I thought: what if you could just talk to any website?

The technical challenge was real. Voice-to-action conversion on arbitrary DOM is hard. I built an agent that could listen to natural language, map it to website elements, execute clicks, form fills, navigation — and do it in 53 languages with <700ms latency.

The economics made sense too. Free tier (60 conversations/month), Growth ($36/mo for 2,100), Scale ($120/mo). Low commitment, high upside.

On paper, it was bulletproof. In reality, I was solving for a problem nobody was actively paying to solve.

What Worked

Let me be clear: some things worked.

The SEO strategy was legitimate. I targeted voice UI, accessibility, agentic browsing, multilingual voice — high-intent keywords with decent search volume. I wrote 20,253 pages of content. Google ranked them. Organic traffic came.

I was getting 800-1,200 monthly visitors. Conversion-to-signup rate: 0%.

The product worked too. Every demo to actual users was positive. "This is incredible," they'd say. But saying "this is incredible" and pulling out a credit card are two different things.

What Didn't Work

1. Accessibility Messaging Alone

I led with accessibility. WCAG compliance. Inclusive design. Morally correct. Nobody was buying because of it. Websites optimize for accessibility because they have to — legal pressure, compliance deadlines. Not because they're excited to try new solutions.

2. The "Just Add Our Script" Narrative

I assumed one-line implementation would be a differentiator. What I missed: website owners don't want new dependencies without proof of ROI. A one-line embed doesn't eliminate objections around performance, security, and support burden.

3. Enterprise Sales Without Enterprise Resources

The actual buying decision lives with product managers, accessibility officers, CX teams. These people don't find you through SEO. They find you through industry recommendations, RFPs, and sales conversations. I had one founder (me).

4. Horizontal Positioning

"Voice OS for websites" is so broad that nobody sees themselves in it. E-commerce, SaaS, healthcare — the answer is yes to all. Which meant no answer felt compelling to any of them.

The Data That Broke Me

At month 5, I ran the numbers:

20,253 indexed pages
1,047 monthly organic sessions (average)
0 signups
~200 hours writing content
$3,200/month infrastructure
Revenue: $0

Cost per potential signup: undefined (there were none).

The Pivot: From Horizontal to Vertical

I'm repositioning AnveVoice to serve three urgent verticals:

1. Healthcare & Telemedicine — The WCAG 2.1 AA compliance deadline is weeks away. Telemedicine platforms face legal exposure if their patient intake forms aren't accessible. Voice-powered form filling solves this in days, not months.

2. Government Websites — Same deadline. Medicare.gov, VA.gov, SSA.gov — all scrambling. 200+ million seniors on Social Security alone. Voice interface is table stakes for accessibility compliance.

3. International E-Commerce — Shopify stores selling globally. Voice product search and checkout in 53 languages is a competitive moat.

These verticals share three things: urgent external deadline, high pain cost, and clear decision makers.

What I'm Asking the Community

Have you faced accessibility compliance deadlines? What made you choose one solution over another?
If you run a telemedicine, government, or e-commerce platform: Would voice-powered form filling + navigation be useful?
Pricing feedback: Is $36-120/month the right range for these verticals?
Distribution: Where do accessibility officers and compliance directors actually look for solutions?

The Honest Truth

I built something good. It just solved a problem nobody was urgently paying to solve. Accessibility is a "should." Legal deadlines are a "must."

I'm chasing the "must."

If you've been in this situation — good product, wrong market positioning — I'd love to hear how you got out of it.

If you're in healthcare, government, or international e-commerce and face April's deadline, let me know. I'd rather help you solve the problem than blog about it.

—Adarsh

AnveVoice is live at anvevoice.app. Free tier: 60 conversations. No credit card required.

Founding community members get lifetime 50% off.