The Middle Management Paradox That's Blocking Your AI Adoption
- Nils Brosch
- 4 days ago
- 5 min read
Middle managers were asked to make most purchasing decisions on new tech (SaaS) in the last decades as tech got cheaper. Now they are asked to purchase technology that likely reduces their number of direct reports, which is often equated to a manager's power. If not done right, middle management will (understandably) protect itself, the same way skilled craftsmen did 150 years ago when machines threatened their livelihoods.
The History You Need to Know
Let me paint you a picture from 1816. Skilled textile workers in England were literally smashing machines with hammers. These weren't random acts of violence – they were calculated responses by craftsmen who saw their expertise becoming obsolete overnight.
But here's what's fascinating: it wasn't just the workers. When Frederick Taylor introduced scientific management in the early 1900s, foremen actively resisted because efficiency experts were usurping their traditional control. Sound familiar?
To be precise, we're seeing the exact same dynamic play out in modern companies right now. Your middle managers – the ones tasked with implementing your AI vision – are facing what those foremen faced: external "experts" (your AI team) introducing systems that threaten their core value proposition.
The Data Behind the Human-Problem of AI Adoption
Here are the numbers that kept me up last night:
71% of companies cite organizational culture as the top barrier to AI adoption
72% of the C-suite say their company has faced at least one challenge on their journey to AI adoption. Some of these barriers include power struggles, conflicts, silos, and even sabotage according to the Enterprise AI adoption report 2025 from Writer.
70% of AI pilot failures reportedly stem from people and process gaps, not the technology itself.
68% of middle managers are concerned about AI's impact on their careers
Gartner predicts 20% of organizations will use AI to eliminate more than half of current middle management positions by 2026
But here's the kicker: while you're trying to figure out AI strategy, your middle managers are quietly protecting their turf. They're not refusing to implement AI – they're just... not prioritizing it. Or they're finding reasons why it won't work in their specific context.
The Empire Building Problem
Think about how you measure management success. Bigger teams = more influence = higher compensation = better career prospects. This creates what I call the "Empire Building Trap."
Your operations manager who oversees 15 people knows that if AI automates 10 of those roles, they're suddenly managing 5 people. In most corporate cultures, that's a demotion, not efficiency.
The uncomfortable truth: You've built incentive structures that reward headcount growth, then wondered why managers resist headcount-reducing technology.
A recent observation:
I spoke to a recruiter-friend of mine who said that they are overwhelmed by BDR/SDR vacancies that they need to fill at the moment. It seems like the white-noise of AI outbound and the subsequent lowering of Conversion Rates, is fought by managers by simply throwing bodies at the problem (expanding their Empire), instead of thinking about AI efficiencies.
The Big Finger Pointing Problem
Unlike the Industrial Revolution – where technology adoption was top-down and centralized due to the sheer amount of investments that needed to be made– we're now in the era of what I call "purchasing democracy", due to lower risk recurring software subscriptions the decision-making has moved towards the lower hierarchy levels.
Hence, purchasing tech like AI is foreign to upper management and threatening for lower management.
The Three Scenarios Playing Out Right Now
Based on conversations with 50+ SaaS founders, I see three distinct patterns:
Scenario 1: The Resistance Middle managers slow-roll AI initiatives. They don't say no – they just find reasons for delays. "We need better data quality first." "The team isn't ready." "Let's pilot it next quarter."
Scenario 2: The Fragmentation Teams adopt AI tools independently. You get speed but lose cohesion. Data lives in silos. Tools don't integrate. You're paying for overlapping capabilities.
Scenario 3: The Evolution (This is rare) Middle managers become AI champions because you've redefined what management success looks like.
The Framework That Actually Works
Here's what the companies getting this right are doing:
1. Redefine Management Metrics
Stop measuring managers by headcount. Start measuring by output, efficiency, and team development. One CEO told me: "We now celebrate managers who eliminate their own busy work through automation."
2. Create AI-Native Career Paths
Managing AI agents needs to count as real management experience. Workday recently introduced an "AI Agent system of record", treating AI like digital employees in their org structure.
3. The Hybrid Authority Model
Your middle managers should have budget authority for team-level AI tools, but within governance frameworks. Think "federated decision-making", autonomy within guardrails.
4. Transform Managers into AI Translators
The most successful AI implementations I've seen position middle managers as the bridge between AI capabilities and business needs. They become the experts who know when to use AI, when human judgment is irreplaceable and what good AI output looks like.
5. Focus AI Innovation Responsibilities In One Role i.e. GTM Engineering Per Department
Big companies have AI teams, smaller companies start employing [Function] Engineers i.e. Recruitment Engineering, Marketing Engineering or GTM Engineering. The issue is that the role definition of these titles is still vague and the i.e. average sales experience of a GTM Engineer is only 2 years. So letting folks automate processes or customer facing messaging without guardrails and experience is dangerous. It's paramount to have the quality of output checked rigorously by the responsible manager (AI translator) and to manage AI innovations experiments centrally i.e. in RevOps or Operations.
6. Executive buy-in is not enough, Executive AI mandates are required
Shopify's CEO Tobias Lütke released an internal memo on April 7, 2025 towards the staff of Shopify. In this memo, he highlighted the need to demonstrate AI skills and the need to demonstrate that a problem isn't fixable through AI, but requires extra headcount. This is in line with 18th century machine adoption, which was pushed top-down from the owners of companies. If left to the middle management, the incentives for change would simply be too little or too dangerous.
The Questions You Need to Ask Monday Morning
What percentage of your middle managers' compensation is tied to team size? If it's significant, you're incentivizing AI resistance.
How many different AI/ML tools are teams using across your company? If you don't know the answer, you have a fragmentation problem.
Who actually makes the call on adopting new AI tools in each department? If it's always the department head, you need more centralized strategy.
When you promote managers, do you consider their ability to leverage AI? If not, you're selecting for the wrong skills.
The Uncomfortable Reality
Your middle managers aren't obstacles to AI adoption – they're the key to making it work. But only if you stop treating them like implementers and start treating them like strategic partners in your AI transformation.
The companies that figure this out first will have an enormous advantage. They'll achieve the coordination benefits of the old centralized model with the speed and innovation of the new distributed one.
The ones that don't? They'll end up like those British manufacturers who ignored Brunel's innovations for decades while Americans raced ahead.
Your choice: evolve your management layer or watch your AI strategy die in the middle.
The data in this article comes from recent studies by Gartner, Deloitte, Harvard Business Review, and Korn Ferry, along with historical analysis of Industrial Revolution management patterns. The SaaS procurement insights are based on 2025 industry surveys tracking the democratization of enterprise software buying.
Send me a message to talk about how to modernize your Go-to-Market processes holistically, to make intelligent & structured use of AI.
Comments