In January 2025, when we went full-time with Harnex AI, we had a clear mission: help organisations harness AI as their core technology to stay ahead. Not to replace their people, but to augment them.
We weren’t interested in building shiny demos or selling vaporware. We wanted to solve real problems for real businesses, with the hope that eventually, we’d come across a problem that was truly worth solving.
So we did what any sensible founders would do: we talked to customers. A lot of them.
We spoke to organisations across Aotearoa and Australia, professional services firms, e-commerce companies, startups, corporates, healthcare providers and creative agencies.
We trained their teams, built their workflows and sat in on their strategy sessions. After working with dozens of businesses, we learned something critical: the best problems to solve reveal themselves when you’re close to the actual work.
The Problem Nobody Was Talking About
Here’s the pattern that kept repeating.
We’d train a marketing team to use AI effectively. Within weeks, they’d stop using Google Search and move entirely to ChatGPT and Perplexity for their work. Not because we forced them, but because once they understood how to use these tools properly, they couldn’t go back.
But then, they’d ask us a question we couldn’t answer: "How do we make sure AI engines are finding us? Our customers are using these tools now. Are we even showing up?"
We’d check manually. ChatGPT one day, Perplexity the next. We'd take screenshots, build a spreadsheet and try to spot patterns. It felt ridiculous. Here we were, helping businesses engineer their AI adoption with systematic approaches and clear metrics, whilst simultaneously guessing at their AI search visibility like we were reading tea leaves.
The behavior has already shifted. But the measurement hasn't.
Why This Matters (The Data)
We didn't rely on 6-month-old consulting reports to validate this. We looked at what was happening right in front of us. The intelligence costs driving AI are racing toward zero and the "Google Era" of search is fracturing.
Consider the shift we are seeing right now:
- The Traffic Drop: Organic traffic to news sites dropped 26% between mid-2024 and July 2025 (from >2.3B to <1.7B).
- The Zero-Click Reality: In 2025, 60%+ of searches that trigger AI Overviews end without a click. The user gets the answer and leaves.
- The Opportunity: Brands that are cited in these AI summaries see a 35% lift in click-through rates, and traffic referred by AI engines is highly pre-qualified, unlocking up to 23x higher conversion rates.
If your potential customers can’t find you in ChatGPT, Perplexity, or Gemini, you aren’t just losing clicks. You are invisible to the most valuable segment of the market.
We Don’t Build Solutions Looking for Problems
This is where our approach differs from most startups. We didn’t start with "AI search is hot, let's build a tool." We started with "our clients have a bleeding neck problem that nobody is solving properly."
Client after client would ask variations of the same question:
- "How do I know if AI is citing my content correctly?"
- "My competitor shows up in ChatGPT and I don't. Why?"
- "I updated my pricing two weeks ago, but Perplexity is still quoting the old numbers. What's wrong?"
These weren't hypothetical concerns. These were real business challenges affecting revenue, positioning, and competitive advantage.
Building With Customers, Not For Them
We could have locked ourselves in a room for six months and emerged with a "perfect" product. We chose the opposite path.
Through our network and events like Startup Grind Auckland, hackathons, and our graduate programmes, we found six reputable companies willing to join us on this journey. Not as beta testers. As co-creators.
These weren't just any six companies. They were businesses that:
- Had real AI search visibility challenges.
- Understood the shift happening in search behaviour.
- Were willing to share honest feedback about what worked and what didn't.
The roadmap to MVP? That was planned by our customers. They told us what metrics actually mattered. They pointed us to where the road needs to go. We're just doing the paving.
Why We're Building AuraScope
After 12 months in the trenches, we learned that measurement drives improvement.
When we trained teams to use AI, we measured adoption rates. When we built automation, we tracked reliability. But for AI search visibility? The industry was flying blind.
We are building AuraScope to solve this with engineering rigour, not marketing fluff.
- No Vanity Metrics: We aren't building a dashboard that just says "You're ranking!"
- Causality: We are building a system where, if your score drops, you know why and what to fix.
- Scale: We can't personally consult with thousands of businesses. But we can build the tool that turns the AI "Black Box" into a "Glass House" for everyone.
We're Still Building (And That's the Point)
We’ve validated the problem. We’ve built a pilot group. But the paving isn't done yet.
We are still building, still learning, and still iterating based on what our pilot group tells us actually works versus what sounds good in theory.
I'm bullish that this is the only way to build software in 2025. We'd rather build incrementally with customers who care about the problem than launch with fanfare to crickets.
What Comes Next
If you’ve read this far, you’re probably in one of two camps:
Camp 1: You’re a business facing the AI search visibility challenge and you’re tired of manual checks. You want to stop hoping AI engines will find you and start engineering for it.
Camp 2: You’re curious about how we’re building this. You want to understand the problem space. Maybe you’re even wondering if you should join our pilot group as we continue to MVP.
Either way, we’d love to hear from you.
We aren't trying to sell you something that doesn't exist. We're building AuraScope transparently, with real customers, solving a real problem. If that resonates with you, if you're tired of vibes-based optimisation and want engineering-based measurement, let's talk.
Because the great divide in AI search isn't between businesses using AI and those not using it. It's between businesses who measure what matters, and businesses who collect vanity metrics.
We're building the platform to close that gap.
