Francel Van Dyk Humaine Head of Operations
TL;DR Building AI capability is an operational problem before it is a tooling one. Here is what Humaine learned through our process of adopting Notion as our operations tool of choice:
AI amplifies what is already there. If your team’s foundation is inconsistent, AI makes it worse faster.
A shared operating environment is the prerequisite, not the reward, of good AI adoption.
Notion is our digital home for training, tools, pipeline, and governance. The tool is not the point. The system is.
Human judgement still decides where AI helps and where it does not. That boundary is documented, not assumed.
Small teams can operate with the discipline of larger ones. The system is what makes that possible.
Before we adopted AI properly, we had to answer a harder question first: where does the work actually live?
Not the just the outputs but also the thinking, process. and institutional knowledge that makes a team consistent when the pressure is on. For most small teams, that knowledge lives in people’s heads, in email threads, in group chat messages that nobody can find six weeks later.
At Humaine, we solved this by building a single operating environment in Notion. Not a document store. Not a project tracker. A digital home for everything the team needs to do good work: training, tools, pipeline, project delivery, and the governance layer that makes AI adoption work in practice.
As Humaine’s Operations Lead. I built this system and I maintain it. This is how it works, why we built it this way, and what other teams can take from it.
The short version: the operating model is the point. Notion is the enabler.
The Problem We Were Solving
We are not a large team. But even small teams can develop a big information problem.
Before we built the Notion environment, our knowledge was scattered. Brand guidelines in one place. Process notes somewhere else. Onboarding happening informally, depending on who had time to explain things. Projects tracked in tools that only some people used consistently.
The result was predictable: uneven output, duplicated effort, and a constant reliance on whoever happened to know where something was. When someone was away, work slowed. When someone left, knowledge walked out with them.
The deeper problem was not disorganisation. It was uneven capability. Some team members knew exactly how to approach a brief. Others were guessing. Some understood the AI tools available to them. Others were not using them at all, or using them poorly.
This is the problem most teams face when they try to adopt AI. The tool is not the barrier. The absence of a shared operating system is.
You cannot build AI capability inside a team if the team does not have a clear, shared way of working first. AI amplifies what is already there. If the foundation is inconsistent, the amplification makes it worse.
What Does Your Team Need Before Adopting AI?
Most teams approach AI adoption as a tool problem. They find a tool, they give people access, and they wait for productivity to improve. It rarely does, at least not consistently.
The reason is that AI adoption is not a tool problem. It is a systems problem.
Consistency
When everyone on the team is working from the same knowledge base, using the same approved tools, following the same process documentation, AI output becomes consistent. The brand voice does not drift. The quality floor does not drop. A junior team member using a well-designed system can produce work that meets the same standard as a senior one.
This is how you scale content with AI without going generic. The system holds the standard. The tool executes against it.
Quality without resistance
One of the most common questions we hear is: how do you get staff to adopt AI without resistance? The honest answer is that resistance usually comes from fear, not laziness. People worry that AI will replace them, or that using it signals they cannot do the job themselves.
The way we addressed this was by framing AI as part of the operating system, not as a replacement for the people inside it. The Notion environment makes clear where AI helps and where human judgement is irreplaceable. That distinction reduces anxiety and increases adoption.
When people understand the boundaries, they use the tools more confidently.
Speed and accountability
With a structured operating environment, AI tools can be deployed faster and with less risk. A brief is already in Notion. The brand context is already documented. The approval process is already mapped. AI can accelerate the work because the scaffolding is already in place.
And because everything is tracked, accountability is built in. If something goes wrong, we can trace it. If something works well, we can replicate it.
Visibility
Leadership can see what is in progress, what has been delivered, and where AI was used. This is not surveillance. It is the kind of operational clarity that allows a small team to make good decisions quickly, without constant check-ins or status meetings.
How We Built Our AI-Ready Operating System
The Notion environment is built around five areas. Each has a clear owner, a clear purpose, and a clear relationship to the work we deliver.
Knowledge and training
This is where we hold everything a team member needs to understand how Humaine works. Brand positioning, tone of voice, process documentation, tool guides, and training materials. When someone joins, they do not need to ask twenty questions. The answers are already there.
This matters for AI adoption specifically. One of the most common reasons AI use goes wrong inside a team is that different people are working from different assumptions about the brand, the audience, the standard. The knowledge hub creates a shared reference point that keeps AI output consistent with who we actually are.
Tools library
We maintain a curated list of the tools we use, why we use them, and how. This is not a showcase. It is a decision record. When a new tool is considered, we document the rationale. When something is retired, we note why.
The tools library also holds our AI guidance: which tools are approved for which tasks, what the human review step looks like, and where AI should not be used without additional oversight. That clarity is what makes adoption consistent rather than accidental.
Pipeline and project management
New business, active clients, project timelines, deliverable status. Everything is tracked in one place. Account leads update it. Leadership reads it. Nobody needs to chase a status update in Slack.
For AI adoption, this layer is more important than it looks. When project information is visible and structured, AI tools can be used to support reporting, summarisation, and briefing without the risk of working from incomplete context.
People and operations
Contracts, onboarding documents, team structure, HR processes. The administrative layer that keeps the business running cleanly. It is not glamorous, but it is what allows a small team to operate with the discipline of a larger one.
AI governance
This is the layer most teams miss. We document which AI tools we use, what tasks they are assigned to, and where human review sits before and after. That makes our AI use auditable, consistent, and improvable over time. The distinction between AI that genuinely changes decisions and AI that only speeds up production is worth understanding clearly – our guide on how to evaluate a marketing agency’s AI capability covers this in detail.
“The question is not whether AI can do the job. The question is whether the team knows when to use it, how to use it, and when to override it.”
What My Role Actually Looks Like
People sometimes assume that a system like this runs itself once it is built. It does not.
My job is to keep the Notion environment honest. That means reviewing it regularly, updating documentation when processes change, retiring tools that no longer earn their place, and adding new guidance when the team encounters something we have not mapped before.
It also means being the person who asks the uncomfortable question when something looks inconsistent. If a process is documented one way but being done another, that is a signal. Either the documentation is wrong or the process has drifted. Either way, it needs resolving.
The system is only as good as the discipline behind it. A Notion environment that is not maintained becomes a liability, not an asset. Outdated documentation is worse than no documentation, because it creates false confidence.
The other part of my role is helping the team understand why the system exists. Not just how to use it, but what it is for. When people understand that the operating environment is what allows them to use AI confidently and safely, they invest in keeping it current. When they see it as admin, they avoid it.
That shift in framing is, I think, the most important thing I do.
What Your Team Can Do Before Adopting AI
We are not suggesting every team should use Notion. The tool is not the point. The principles behind how we use it are.
Build the system before you adopt the tool. AI rewards clarity and structure. If your team does not have a shared way of working, a shared understanding of the brand, and a shared record of what good looks like, AI will not fix that. It will make the inconsistency faster and more visible.
Make the operating environment feel like an asset, not a chore. The Notion environment works at Humaine because the team understands what it is for. It is not a filing system. It is the infrastructure that makes good work repeatable. When people see it that way, they maintain it. When they see it as admin, they abandon it.
Separate the governance from the tools. Knowing which AI tools to use is less important than knowing when to use them, what human review looks like, and where the boundaries are. That governance layer is what makes AI adoption safe and sustainable, not just fast.
Document the decisions, not just the outcomes. When a tool is retired, note why. When a process changes, update the record. The value of an operating system is not just what it contains today. It is what it tells you about how you got here and why things work the way they do.
These are not Notion principles. They are operating principles. The tool just makes them visible.
The Operating Model Is the Point
We did not build this system to impress anyone. We built it because the work required it.
A team that wants to use AI well cannot afford to be disorganised. The two things are in direct conflict. AI rewards clarity, consistency, and structured context. Disorganisation produces the opposite. The tool will faithfully reproduce whatever chaos it is given.
What Notion gave us was a place to put the discipline. The training, the tools, the process, the governance. Not as a statement of intent, but as a working environment that the whole team uses every day.
If you are thinking about how to build AI capability inside your team, or how to create a cleaner operating system for the work you already do, we are happy to talk through what we have learned. Not as a consultancy pitch. As a team that has done the building and is still doing it.
The system is never finished. That is the point. If you want to understand how this operating model connects to the broader commercial framework behind Humaine’s work, the Six Labs Model explains how strategy, content, and technology compound into durable advantage.