When good people don’t click… what’s it really costing you?

johan@avenix.com.au (Johan Meyer) — Fri, 12 Jun 2026 03:29:37 GMT

Have you ever sat in a leadership meeting and thought:

“We’ve got a strong team… so why does this feel so hard?”

There’s no blow-up. No obvious failure. Just a quiet friction that shows up everywhere. Decisions take longer than they should. The same conversations keep resurfacing. People stay busy, but progress feels… slow.

The problem founders see (but can’t quite name)

It rarely presents as a “team issue.” Instead, it looks like:

Projects that drift or stall
Leaders pulling in slightly different directions
You stepping in more often than you should
Work being redone, or not quite landing right the first time
Subtle tension, nothing dramatic, just enough to slow things down

On paper, everything looks fine. Good people. Solid experience. Strong intent. But underneath, something isn’t clicking. And over time, that friction compounds.

Why it doesn’t show up clearly in the numbers

Here’s the challenge: you won’t find this problem in your P&L. There’s no line item for:

Confusion
Delayed decisions
Misaligned priorities
Unspoken frustration

You will see salaries. You will see overheads. But what you don’t see is:

Lost productivity from duplicated or stalled work
Revenue that never materialised because momentum slowed
Margin erosion from rework and inefficiency
Opportunities missed because no one truly owned the decision

It’s not a cost problem. It’s a conversion problem.

Effort → output

Opportunity → revenue

Strategy → execution

When teams don’t click, that conversion rate drops, quietly. You’re effectively paying for full capacity… and only realising part of it.

The real shift: it’s not about better people

This is where many founders go wrong. They think: “Maybe we need stronger people.” But more often than not, it’s not capability, it’s structure and alignment. Two consistent patterns show up in great businesses:

1. Clarity of roles and decisions

Airbnb’s Brian Chesky has spoken about the need to clearly define who owns what as the business scales. Without it, even strong leaders compete instead of align.

2. Intentional team design

Sara Blakely (Spanx) built her team by hiring against her weaknesses, not in her own image. Yvon Chouinard (Patagonia) prioritised values alignment and trust, creating teams that could operate with autonomy.

Different approaches, same underlying principle:

Strong teams aren’t built by chance.

They’re designed to work together.

One thing you can do this week

Pick one key initiative in your business and ask a simple question: “Who is ultimately accountable for this, and who makes the final call?” If the answer isn’t immediate and consistent across your team, you’ve found friction. Clarity here creates momentum almost instantly.

A final thought

Most founders don’t have a people problem. They have a system that quietly forces good people to work against each other. Fix that, and you don’t just improve culture. You unlock growth that was already sitting inside the business.

From Personal AI to AI-Native: Where does your vision sit?

johan@avenix.com.au (Johan Meyer) — Fri, 29 May 2026 07:46:27 GMT

Most firms are talking about AI. Far fewer are redesigning how work actually gets done.

That matters, because there is a big difference between a team member using AI to get through their own workload faster, and a firm that has rebuilt its operating model around connected AI systems. One creates individual productivity. The other creates leverage, consistency and scale.

The shift is happening fast. What looked advanced six months ago can feel basic today. In the right environment, a firm can move from personal AI use to platform-level orchestration in a matter of months. But speed without vision creates chaos. The firms getting real value from AI are not just buying tools. They are making deliberate choices about culture, workflow design, governance and leadership.

I think about AI adoption across four stages: Personal AI Use, Process Improvement, Platform Orchestration, and Autonomous Agents / AI-Native. Each stage creates value. Each stage also has a ceiling. Knowing where that ceiling sits helps leaders decide what to invest in next, and what needs to change in the business to support it.

Stage 1: Personal AI Use

This is where most firms begin. Individuals use AI to write faster, think more clearly, summarise information, brainstorm ideas, structure their day, create lists, or get a first draft on the page. It is useful immediately because it helps people move quicker through work that would otherwise take longer. For many professionals, this stage creates a genuine lift in energy and confidence because the blank page disappears and momentum improves.

But the value created here has a hard ceiling. Every employee is still doing all of the work themselves, just with a better tool in hand. The gains are personal, not institutional. They depend on the curiosity, discipline and prompting skill of the individual. That means quality varies, knowledge stays fragmented, and very little compounds across the firm.

The benefits are obvious: faster drafting, quicker synthesis, lower mental load, and better personal organisation. The pitfalls are just as real: inconsistent outputs, overconfidence in weak answers, no standard method, and no meaningful leverage of the employee’s deeper skills. Privacy and security matter even at this early stage. If staff are pasting client or personal information into public tools without guardrails, the firm is creating risk before it has created much strategic value. Australian privacy guidance is clear that personal information used in AI systems still attracts privacy obligations, and public tools should be treated with care, especially where sensitive data is involved.

The plateau at Stage 1 arrives when individual productivity improves, but the firm itself does not become more scalable, more consistent or easier to manage.

Stage 2: Process Improvement

Stage 2 starts when AI is built into a specific repeatable task. Instead of asking staff to figure out their own prompts, the firm creates a defined agent or workflow for a known job: reviewing documents, drafting a standard client response, extracting information from forms, checking a workpaper, preparing a first-pass summary, or handling another repetitive activity that consumes time but adds little strategic value. The employee is still responsible for the outcome, but some of the grind has been lifted from their plate.

This is often where firms first see measurable return. Low-value work reduces, turnaround time improves, and people get more room to focus on clients or judgement-heavy tasks. But the ceiling is still there. The employee remains the owner of every step, every review and every handoff. There is still limited leverage of that person’s true talent. And because different people often continue to do the rest of their job in different ways, scaling remains difficult.

The benefits at this stage are better throughput, fewer manual touches, and stronger consistency inside the specific workflow. The pitfalls are agent sprawl, duplicated logic, weak documentation, and overestimating what one well-built agent can do for the broader business. Security becomes more important because these agents often need access to business data or internal systems. Least-privilege design, logging, clear ownership and human review are essential if you want the gains without creating blind spots. As AI use matures, governance, security and operational discipline stop being optional and become the foundation for trust.

The plateau at Stage 2 shows up when isolated wins do not translate into a more connected, more teachable, more scalable firm.

Stage 3: Platform Orchestration

Stage 3 is where the whole system starts to join up. Instead of building one-off helpers, the firm develops connected roles across workflows. Employees stop being only doers and start becoming managers of multiple agents that handle the daily grind they used to perform themselves. Those agents can work across several systems, with carefully controlled permissions, shared context and a clearer understanding of what good looks like.

The productivity lift here is material because work is no longer trapped in silos. A connected platform can move information, trigger steps, standardise methods and present review-ready outputs far more consistently than a team of individuals all working from memory and habit. The employee still owns the final output. They still review, apply judgement and take responsibility for what goes to a client or into the business. But their role has shifted upward. They are spending less time producing and more time directing, validating and improving.

The benefits are significant: stronger productivity across teams, more consistent service delivery, less dependency on individual memory, and far better leverage of human judgement. The pitfalls are complexity, weak change management, unclear accountability between humans and agents, and the temptation to connect too much too quickly. Privacy and security have to be designed in, not added later. As agents gain access across systems, the question is no longer only whether the answer is right, but whether the action is authorised, observable and reversible. Firms need audit trails, role clarity, escalation paths and disciplined access controls if they want to orchestrate safely at scale.

The plateau at Stage 3 arrives when AI can support whole roles and connected workflows, but still waits for a human to initiate the work and sign off every outcome.

Stage 4: Autonomous Agents / AI-Native

Stage 4 is where the operating model changes properly. Work no longer needs to begin with an employee deciding to start it. AI can prepare work because another event has triggered it: a document arrives, a deadline approaches, a threshold is breached, a transaction looks unusual, or a client event requires follow-up. Because the system is already connected across the firm, information moves without the silos that usually prevent scale. Employees become AI managers, exception handlers and owners of judgement rather than primary producers of routine work.

This is what an AI-native firm starts to look like. Capacity expands without a matching increase in headcount. Quality can become more consistent because the system is operating from shared rules, connected data and persistent memory. The employee role rises again: fewer manual tasks, more oversight, more intervention where judgement is actually needed, and more time spent on client relationships, commercial thinking and strategic decisions.

The benefits are transformative, but so are the risks. The pitfalls include automation without accountability, agents acting beyond intended authority, hidden failure modes, poor exception design and cultural resistance from teams who feel the ground moving under them. Security and privacy become board-level issues here. As autonomy increases, governance must mature with it: agent identity, permission boundaries, human override, monitoring, incident response and compliance controls all need to be explicit. The challenge is no longer only whether AI can help. It is whether the firm can trust autonomous action inside regulated, client-sensitive workflows. Recent guidance and maturity models all point in the same direction: autonomy without governance is not maturity, it is exposure.

The plateau at Stage 4 is less about technology and more about governance, trust and the organisation’s ability to keep redesigning itself as the tools continue to change.

Culture, Vision and Leadership Matter More Than the Tools

The biggest mistake leaders make is treating AI as a software project. It is an operating model decision. It changes how work is designed, how people are trained, how quality is reviewed, how risk is governed and what the firm expects from its professionals. If the culture rewards heroics, protects inconsistency and tolerates messy processes, AI will simply amplify that mess. If the culture values discipline, openness, learning and shared ways of working, AI can become a serious advantage.

Vision matters because these transitions can happen quickly. A firm can move from personal AI use to platform orchestration in months if leadership is clear on the destination and ruthless about priorities. But moving fast does not mean skipping foundations. The firms that will win are the ones building trust, standards and capability at the same time as they build automation.

So, what is your vision?

The most useful question is not whether your firm is using AI. It almost certainly is. The better question is where the value is being created, where it plateaus, and what has to change in your culture, systems and leadership to move to the next stage.

Personal AI use is a good start. Process improvement is where measurable ROI begins. Platform orchestration is where scale and consistency start to compound. Autonomous agents are where the firm becomes genuinely AI-native. The gap between those stages is now moving quickly. The opportunity is real, but so is the need for judgement.

Established firms will often find this transformation hardest, simply because they have more legacy systems, habits, structures and history to work through. By contrast, new businesses with far less baggage are building AI-native models from day one because they do not have the same layers to unwind. What they often lack, though, is the judgement that comes from years of battle scars and hard-won lessons. In the long term, the greatest value will be created by businesses that can bring both together: the speed and design freedom of an AI-native model, with the commercial wisdom, pattern recognition and discipline that only experience can bring.

Want to know more, please reach out. Would love to discuss this topic with you, no expectations.

Avenix Pty Ltd blog