You probably saw some of the news reports earlier this year: media companies across the country, drunk on the hype coming out of Silicon Valley, fired their writers and editors, declaring themselves “AI-first.” We saw the same happen with engineers at technology companies. But within weeks, many were quietly rehiring their critical staff when the technology failed to deliver.
You may have enjoyed a silent chuckle at the hubris but there’s a powerful lesson here for anyone running a $100-500 million company: your competitive advantage – in fact, your very survival – depends on understanding that technology adoption follows predictable phases and knowing how to navigate them successfully.
IBM reports that 61% of CEOs say competitive advantage now depends on who has the most advanced AI capabilities. But most companies are implementing AI in ways that virtually guarantee they’ll lose to competitors who understand the real game.
The reality is stark. If you’re stuck optimizing your current processes while competitors are reimagining their entire business model, you’re not just falling behind – you’re facing an existential threat. But as a mid-market CEO, you have a unique advantage: you’re more nimble than large enterprises while more resourced than startups. The question is whether you’ll leverage this advantage before it’s too late.
The four-phase transformation framework that determines winners and losers
Technology adoption isn’t random. Whether we’re talking about AI, data analytics, or any transformative technology, successful companies traverse a predictable pathway. Understanding these phases, and where you stand in each, can mean the difference between leading your industry and becoming irrelevant.
Phase 1: “Sprinkle on top”
This is where most companies start with implementing any new technology, and it feels like progress. You add AI chatbots to your website, implement robotic process automation tools, or layer new capabilities onto existing processes. It’s the “we added AI to our customer service” phase.
Take robotic process automation. The technology lets you sprinkle automation on top of what you already have. You can accomplish some efficiency gains in a relatively painless way, but that’s as far as that technology will ever take you. Your people are still fundamentally going through the same processes, just with digital assistance.
The trap here is obvious once you see it: you’re putting lipstick on a legacy pig. It feels modern, but you’re still doing the same things you always did, just with fancier tools.
Phase 2: “Integration and efficiency”
Here, technology starts making your operations genuinely smoother. You automate reports, streamline workflows, maybe reduce headcount or cycle times while improving repeatability and consistent quality. This is where agile transformations typically happen – development teams adopt new methodologies to deliver faster and more efficiently.
But there’s a crucial insight here: Phase 2 isn’t just about efficiency. You’re subtly training your people to work with the new technology. Think about how Apple gradually trained us to use touch interfaces. They didn’t throw the equivalent of an iPhone 14 at us in 2007. They moved us from physical buttons on iPods to a single button on early iPhones to no buttons at all. By the time we got to complex multi-gesture interfaces, we were ready.
The limitation? You’re still fundamentally doing the same things, just faster or cheaper. That’s optimization, not transformation.
Phase 3: “Reimagining from first principles”
Now things get interesting. You stop asking “How can this new tech help us do what we already do?” and start asking “If we started from scratch, what problems could we solve today that we couldn’t before?”
This is where real disruption lives. You’re not just optimizing, you’re re-inventing.
A while back, we worked with a tape and film-based production company that was responsible for a number of well-known TV channels. They operated massive production facilities with studios, editing rooms, mastering, and distribution facilities. When they wanted to distribute a show internationally, they had to create different versions for different markets – varying commercial break placements, different video formats (NTSC vs. PAL), and different show lengths. Everything ended up on physical media that got couriered around the world.
Working with them over a decade, we helped transform them into a digital media company. They shut down all their production facilities. Everything became digital and cloud-based. But more importantly, they shifted their entire business model from being a production company to being a digital media company. Independent producers now create the content while the business focuses on distribution and leveraging their vast digital media assets.
The breakthrough came when we figured out how to make all their old video content searchable by indexing closed caption transcripts with time codes. Suddenly, their people could ask questions like “Which episodes featured recipes for chicken?” The system would find those episodes and show the exact time index. Digital editing tools could cut out the segment, create a promotional spot, and distribute it over the internet – all with a few clicks.
That’s reimagining from first principles. They didn’t just digitize their old processes; they created entirely new capabilities that were impossible in the analog world.
Phase 4: “Native technology”
The technology becomes “dial tone” – just how business is done. If you were starting a software company today, would you build on-premise enterprise software or cloud-based software? The answer is obvious: 98% of the time, you’d build for the internet in the cloud. Most of the time, we don’t even contemplate doing things offline anymore.
In the Generative AI wave, the leading tech companies are aiming at making the new capabilities that AI agents unlock just as unremarkable (from an operations perspective). Last month, the Amazon CEO put out a company-wide memo revealing that the company has over 1,000 GenAI services and applications in the works, and urged his people to “use and experiment with AI whenever you can.” Marc Benioff of Salesforce has claimed that 30-50% of all code is now written by AI agents and that they have reduced support cost by 17% using agentic AI.
These enterprises are going all in on the belief that AI agents will change how we all work and live. This is the remarkable thing about AI: there’s now a generation of people walking the earth who will never know a non-AI world. They’re five and six years old right now, but some are already using devices with AI built in. For them, AI is Phase 4 from day one. The rest of us are somewhere back in phases two or three.
The competitive threat reality check
Why phases 1-2 put you in mortal danger
While you’re making existing processes 10% more efficient, your competitors might be inventing entirely new approaches to serving your customers. This isn’t about avoiding technology: Blockbuster, Borders, and taxi companies all used technology. They lost because they only used it to optimize business models that were already becoming obsolete.
The startup threat is real. New companies can build with Phase 3 or even Phase 4 thinking from day one. They don’t have legacy systems, established processes, or embedded organizational thinking to constrain them. If you’re anywhere less than Phase 3, a competitor will eventually eat your lunch.
Your mid-market advantage window
But here’s your superpower: you’re more nimble than enterprises and more resourced than startups. Large companies are paralyzed by “shadow IT” concerns and compliance bureaucracy. I know of companies right now blocking network traffic to online AI services to shut down unauthorized AI use while simultaneously building internal AI systems.
You can move faster than that. But this advantage is temporary: you have to act decisively, or you’ll lose it.
The CEO’s strategic transformation playbook
How to identify your current phase
Start with an independent assessment. You might be in Phase 2 with data analytics while still in Phase 1 with AI. The key question isn’t “How much faster is this?” but “What new capabilities does this create?”
Warning signs you’re stuck: implementing technology without a clear vision of what success looks like. As I coach executives, there are two kinds of goals you can set: point goals and process goals. Point goals have definitive endings: “I want $700 million in revenue in 2025.” Process goals are more open-ended: “I want to be a leader in AI driven customer success.”
If you’re not really clear on what problems you’re solving or what capabilities you’re creating – and not everything is a problem to solve; sometimes it’s a new capability to create – then you have no way to know what phase you’re in or how to measure your progress.
When and how to advance phases
The majority of the time, you have to traverse this pathway in order. It’s very difficult to jump straight to Phase 3 when something new comes along. Mostly, first movers don’t win. Think about the railroads. The very first people who decided to build trains across the US all lost their money. It was the second generation, once the infrastructure was established, who became the robber barons. More recently, the same thing happened with Web 1.0 – a lot of people lost their shirts in the dotcom crash by going all in on technology that hadn’t found its feet yet.
Be an early adopter in Phase 1. When it’s time to start sprinkling new technology on top of existing processes, get in there and see how it works. This prepares you for when reimagining time comes. It’s hard to reimagine what you can do with a technology if you don’t understand what that technology is capable of.
Use Phase 2 to prepare for Phase 3. You’re not just making things more efficient; you’re training your people and starting to imagine what the third phase might look like. What can you do that you’re not doing now?
Only move to Phase 3 when you have a hypothesis for how to reimagine and a clear vision of what the outcome should look like. If you’re in the $500m-1B range, you can afford to conduct experiments. Don’t spray technology around thinking it’s going to magically make everything better but make multiple bets and double down on what works.
Avoiding common transformation pitfalls
I’ve seen companies implement agile transformations by telling the development organization to “go be agile” without thinking about how that changes the interface with the rest of the company. Product management is still waterfall, thinking about five-year roadmaps, while development can’t see anything further than two weeks away. You create impedance mismatches that actually make things worse.
The lesson: don’t focus on one area in isolation. Look at how everything interconnects. This is often where the opportunities lie in any case.
Another trap: technology without purpose. Remember the mid-’90s when everybody and their brother wanted a web page? Every company had to have a homepage with a picture of their building, phone number, and address. Nobody knew why they wanted to be on the internet or how they could leverage its power to reach customers, service customers, or make their brand stand out. They just wanted to be there because everybody was there.
AI is the latest version of this. People say, “We need AI in our product” so they can tell Wall Street or VCs that they’re an AI product or service. But if you’re not clear on what success looks like, you have no way to measure or understand the impact. As the Cheshire cat told Alice, “If you don’t know where you’re going, any road will take you there.”
Your AI-specific action plan
Current state assessment
The biggest concern I’m hearing at the C-suite level is the privacy and security around AI: the potential loss or escape of proprietary information, customer data, financial information. If you process credit card information, you need PCI compliance. How do you maintain that while leveraging AI capabilities?
Meanwhile, your employees want AI in their lives. This leads to shadow IT issues where people are using consumer AI tools with corporate data. Now, you could bring in the technology that monitors whether corporate email addresses are being used with services like ChatGPT, so you can block it. But realistically, you’ve got to find ways to allow your employees to leverage the benefits without the risks.
Building your competitive AI advantage
Start with controlled Phase 1 experimentation. Some AI SaaS companies are offering private subscription plans. At Ten Mile Square, we have a Google Workspace Pro plan, and when I log into Gemini, it tells me that because of our account level, none of our data will be used for training or retained. Note that I would never provide or use any client data or confidential information with this or any service not explicitly sanctioned for such use by the client.
For companies with the resources – and this is where mid-market companies have an advantage – consider running your own foundation model in a walled garden. I’m working on a project right now where we’re running a foundation model on AWS, connecting it to internal data sources, source code repositories, and engineering documentation. It’s becoming an expert technology consultant for internal developers and technology workers.
So now they can ask it: “I need a library that does X,” and it searches all your source code to say, “Here’s one that does what you want.” Or they can upload buggy code and ask it to fix the problem. All without worrying about proprietary information escaping to third party control and ever-changing usage policies.
The timeline reality
Very few companies are in Phase 3 with AI yet. The AI companies themselves are all about Phase 3. They’re trying to reimagine the world directly. Some media companies are attempting it, though with mixed results. But for most businesses, we’re still in early days.
The models are still immature, and this stuff takes time. But that’s also your opportunity. The companies that reach Phase 3 first will create new business models and revenue sources, and set new industry standards for years to come.
What got you here won’t get you there
Each phase requires different thinking. Just like a company can’t do the same things to get from $500m to $1B that it did to get from $100m to $500m, the same applies to technology adoption.
Maybe you’ve been running the same stable manufacturing operations system for decades. Bugs are few, outages are rare, and the talent pool while aging is stable for now. Yet none of it will work with AI capabilities like dynamic pricing or real-time product recommendations.
The approaches that make you efficient at your current business model might prevent you from seeing opportunities to fundamentally reimagine how you serve customers.
The survival imperative
Here’s the bottom line: companies that reach Phase 3 first in AI will define new competitive landscapes for everyone else. As a mid-market CEO, you have a unique window of opportunity. You’re more nimble than large enterprises but more resourced than startups.
The question is whether you’ll use this advantage to lead the reimagination of your industry and invent new forms of value, or spend the next decade playing catch up to competitors who moved faster.
Technology adoption is a journey, not a destination. You have to traverse its phases more or less in order, but you can move through it faster and more strategically than your competitors. The companies that understand this and act decisively will survive and thrive. Those that don’t risk becoming cautionary tales.
Understanding where you are in the technology adoption cycle is just the first step. At Ten Mile Square, we help mid-market companies assess their current capabilities, identify transformation opportunities, and develop strategic roadmaps that turn technology adoption into competitive advantage. Our five-step technical assessment process helps you define the real problems, analyze gaps, and develop actionable recommendations that align your technology infrastructure with your business goals. Because the biggest risk isn’t moving too fast, it’s moving too slowly while your competitors reimagine your industry.