I spent 30 years building telecom infrastructure.

I have watched every major shift in the industry — analog to digital, digital to VoIP, VoIP to cloud. What is happening now is bigger than all of them combined. And almost nobody is talking about the part that matters most. The AI conversation has been dominated by chatbots, copilots, and large language models. Important, yes. But they are the visible layer. Underneath, something quieter and far more consequential is happening: the voice infrastructure that every business depends on is being rebuilt from the ground up. Not upgraded. Rebuilt. Every phone call a business makes or receives contains a signal — about the customer, the employee, the market, the moment. Until now, those signals vanished the second the call ended. We started Nucleus to change that. Today we are launching The Signal — a newsletter where we share what we are seeing from inside this shift. Market data. AI trends. The insights that most businesses are missing because they are too busy answering the phone to realize how many calls they are not answering. First edition is live. Every word is a signal. The question is whether anyone is listening.

30 years in telecom taught me one thing. Every phone call contains a signal. About the customer. About the business. About the moment. Until now, those signals vanished the second the call ended. We built Nucleus to capture them. Today we are launching The Signal — where we share what most businesses miss. First edition is live. Link in bio.

30 years in telecom. Watched every major shift. What is happening now is bigger than all of them. Every phone call contains a signal. Until now, they vanished when the call ended. We built @NucleusAI to capture them. Launching The Signal — our newsletter. First edition live.

I need to tell you a number that is going to bother you.

$56,700. Per month. Gone. From a single 6-bay auto repair shop. Not from bad marketing. Not from a failed campaign. From a phone that rang while every advisor was busy — and nobody picked up. Here is the math: 30 calls a day. 23% miss rate. That is 7 calls per day that go to voicemail. At $450 average repair order and a 60% conversion rate — that is $56,700 every month that called, got silence, and drove to the competitor. The worst part is not the money. It is that nobody counts it. There is no line item on any P&L for "revenue that called us and we did not answer." So it stays invisible. Month after month. Year after year. We see this number because Nucleus processes 600 million minutes of voice traffic every month. When you sit on that much data, the patterns are impossible to ignore. And this pattern — the silent revenue leak — is everywhere. Not just auto repair. Healthcare. Retail. Financial services. Any business where the phone still matters (which is most of them). The first step to fixing it is counting it. We wrote about this in The Signal. Link in comments. Source: Numa 2024 Industry Trends, ASA Surveys, Nucleus network data.

$56,700. That is what a 6-bay auto repair shop loses every month to missed calls. Not bad marketing. Not lost leads. Customers who called. Got voicemail. And left. The worst part? Nobody counts it. There is no line item for "revenue that called us and we did not answer." The first step to fixing it is seeing it. That is what we do at Nucleus. Read more in The Signal. Link in bio.

$56,700/month. Revenue that called a 6-bay auto shop, got voicemail, and left. 30 calls/day. 23% miss rate. $450 avg ticket. There is no line item on any P&L for "customers who called and we did not answer." So it stays invisible. Until you count it.

A shop owner told me something I have not been able to forget.

"We are too busy doing the work to answer the phone that brings us more work." He said it like a fact. Not a complaint. Just the reality of running a service business in 2026. His bays were full. His advisors were writing estimates, reviewing inspections, managing walk-ins. And the phone was ringing. And ringing. And ringing. Every one of those calls was a customer who had already decided to choose him. They picked up their phone. They dialed his number. And they got silence. That is the problem we built Nucleus to solve. Not because it is a good market opportunity (it is). Not because the technology is ready (it is). But because these people deserve better. The shop owner who opened at 6 AM. The advisor who stays late to make sure the brake job is done right. The tech who takes pride in diagnosing what nobody else could. They are not faceless businesses. They are families. The best thing we can do is serve the people who serve others. That is not a tagline. That is the reason Nucleus exists. If you run a business where the phone still matters — and it does — read The Signal. We wrote it for you. Link in comments.

"We are too busy doing the work to answer the phone that brings us more work." A shop owner said this to me. Not as a complaint. As a fact. His bays were full. His team was working. And customers were calling, getting voicemail, and leaving. Every one of those callers had already chosen him. They dialed his number. And got silence. The best thing we can do is serve the people who serve others. That is why Nucleus exists.

"We are too busy doing the work to answer the phone that brings us more work." A shop owner told me this. Not a complaint. A fact. His customers were calling. Getting voicemail. Leaving. They had already chosen him. They dialed his number. And got silence. We built Nucleus for people like him.

Three numbers you need to understand about 2026:

$9 billion — the agentic AI enterprise market this year. 171% — the average return on enterprise AI agent deployments. 80% — the share of enterprise apps expected to embed AI agents by December. Read those again. This is not a forecast. This is capital being deployed right now. But here is the part that most of the AI conversation misses: The value is not accruing to the applications. It is accruing to the infrastructure. Think about what happened with cloud computing. Thousands of apps were built on AWS. Most of them came and went. Amazon kept the infrastructure revenue. The same pattern is emerging in AI. Thousands of AI applications will launch this year. Most will struggle to differentiate. But the companies that own the underlying infrastructure — the voice networks, the data pipelines, the orchestration layers — those are the ones building something that lasts. At Nucleus, we own the telecom backbone that processes 600 million minutes of voice traffic every month. We did not build an app on top of someone else's platform. We built the platform. Applications come and go. Infrastructure compounds. More in The Signal, Investor Edition. Link in comments. Sources: Deloitte 2026, Tech Insider, Gartner/Salesmate

The AI shift in three numbers: $9B — agentic AI enterprise market 171% — average ROI on AI agent deployments 80% — enterprise apps embedding agents this year But here is what most people miss. The value does not accrue to the apps. It accrues to the infrastructure. Think about cloud. Thousands of apps built on AWS. Most came and went. Amazon kept the infrastructure. Same pattern. Happening right now. In AI. Applications come and go. Infrastructure compounds.

Three numbers that define 2026: $9B — agentic AI enterprise market 171% — average ROI on AI deployments 80% — enterprise apps embedding agents this year The value is not accruing to applications. It is accruing to infrastructure. Applications come and go. Infrastructure compounds.

Most businesses are running their phone system on technology that was designed before the iPhone existed.

It does one thing: route calls. Dial 1 for sales. Dial 2 for support. Hold music. Transfer. Voicemail. That is the entire capability of a traditional PBX. And it has not fundamentally changed in 30 years. Here is what it does not do: It does not transcribe calls in real time. It does not analyze customer sentiment. It does not tell you that your Tuesday morning miss rate is 34%. It does not detect that your best sales rep is burning out based on vocal stress patterns. It does not surface the $56,700 in monthly revenue disappearing through voicemail. An AI-native PBX does all of this. Not as an add-on. Not as a dashboard bolted onto the side. As the foundational layer. Every call — inbound, outbound, internal — flows through AI from the first second. Every word becomes data. Every data point becomes a signal. Every signal becomes an insight the C-suite can act on. This is the category Nucleus is building. And the upgrade cycle from legacy PBX to AI-native PBX is one of the largest infrastructure replacement opportunities in enterprise technology right now. Cloud replaced on-premise servers. Streaming replaced physical media. AI-native voice will replace the PBX. The question is not whether this happens. It is who captures the value when it does. We break this down in The Signal. Link in comments.

Your phone system was designed before the iPhone. It routes calls. That is all it does. It does not listen. It does not learn. It does not tell you that 23% of calls go unanswered. It does not detect burnout in your best rep's voice. An AI-native PBX changes everything. Every call — transcribed — analyzed — intelligence. Cloud replaced servers. Streaming replaced media. AI-native voice replaces the PBX. The question is not if. It is who captures the value.

Your PBX was designed before the iPhone. Routes calls. That is all. Does not listen. Does not learn. Does not tell you 23% of calls go unanswered. AI-native PBX: every call — transcribed — analyzed — intelligence. Cloud replaced servers. AI-native voice replaces the PBX. Who captures the value?

A few years ago, I hit a wall.

Not a financial wall. Not a business wall. An identity wall. I had co-founded one of the largest telecom companies in Canada. Eight-figure revenue. 560 employees. Awards. Recognition. Everything that was supposed to mean I had made it. And I was disengaged. Because somewhere along the way, my identity had become the company. I was the founder. The president. The builder. Take those away, and I did not know what was left. So I stopped. And I asked the most dangerous question a successful person can ask: Who am I beyond my performance? That question changed everything. It led me to write a book. It led me to transform the culture of a 560-person company. It led me to a standing ovation at IMPACT15. And it led me, eventually, to Nucleus. Because here is what I discovered: the greatest threat to any business is not technology, competition, or market shifts. It is leaders who do not lead from who they are. AI is rising. The world is shifting. Leaders everywhere are scrambling. But the ones who know who they are — not what they do, but who they are — those are the ones shaping what comes next instead of reacting to it. That is why we built Nucleus the way we did. Not to compete on features. To build from identity. Every word is a signal. And the most important one comes from within. The Signal newsletter is live. Link in comments. — John

A few years ago I hit a wall. Not financial. Not business. Identity. I had built an eight-figure company. 560 employees. Awards. Everything that was supposed to mean I had made it. And I was disengaged. Because my identity had become the company. So I asked the most dangerous question a successful person can ask: "Who am I beyond my performance?" That question changed everything. It led to a book. A culture transformation. A standing ovation. And eventually, to Nucleus. The greatest threat is not AI. It is leaders who do not lead from who they are. — John Stix

I co-founded an 8-figure telecom. 560 employees. Awards. Recognition. And I was disengaged. Because my identity had become the company. I asked: "Who am I beyond my performance?" That question changed everything. The greatest threat is not AI. It is leaders who do not lead from who they are. That is why we built Nucleus.