How to Roll Out AI Across Your Whole Tour Business Team

The Problem Hiding in Plain Sight: Personal Account Sprawl

Here’s what usually happens by accident. When AI tools first showed up, a few people on your team started experimenting on their own. Someone likes ChatGPT while someone else prefers Claude. If you’re on Google Workspace, someone’s probably playing with Gemini. And your admin has quietly been using her own account for months.

Before long you’ve got six, seven, ten different people, each on their own subscription, each with a private setup, and each with skills and context tied to an account that only they can see.

When somebody leaves, everything they taught their AI, every process they built inside it, can walk out the door with them.”

That’s the real cost. Instead of building a company asset, you’re renting five or ten personal ones. A team plan fixes this at the root: Everyone works under one roof, and the knowledge that gets built belongs to the business, not to whoever happened to build it.

What a Team Plan Actually Gives You

If you’ve only ever used an individual account, a team plan comes down to three things.

Centralized billing under one account. If a guide leaves in October, you turn off that seat and stop paying for it.

Admin level controls. You decide who has an account and what they’re allowed to connect to. A shared project doesn’t have to be visible to every single person in the organization.

A stronger data default, which matters if you’re feeding it real business data. On the team plans for both Claude and ChatGPT, your conversations and connected company files aren’t used to train the model by default. The terms evolve, so it’s worth reading the current ones yourself, but that default matters a lot once you’re uploading anything sensitive.

Pricing changes, but expect roughly $20 to $25 US per seat per month for both Claude’s team plan and ChatGPT Business, usually with some kind of seat minimum or upfront commitment.

One Shared Project Per Department

Here’s the setup principle that tends to work for most teams: One shared project per department, sized to that department’s actual work instead of one catch-all project trying to cover the whole company.

A shared project is a workspace with its own instructions and its own knowledge that the people on your team plan can access. For a tour operator, that might look like a guest communications project for inquiries, complaints, and booking changes. A marketing project for tour descriptions, social content, and newsletters, with your brand kit and sample copy uploaded. An operations project for scheduling, checklists, and supplier notes. And an owner project that stays private, for financials, HR, or strategy.

You don’t need to build all of these on day one. If you’re a small team, start with one, maybe marketing or guest communications, and add people as it makes sense.

Where Your Team’s Knowledge Should Actually Live

This is the question that trips people up. If you’ve been building a personal AI brain, a folder of files on your own computer using a tool like Claude Cowork, that’s still exactly right for you as the owner. Keep building it.

“Your personal AI folder is a cockpit built for the owner. Each department on your team needs its own separate project.”

For the team, each department needs its own project with real context: business information, connectors into the tools they use, and the SOPs or guidelines relevant to their work. For most organizations at this stage, the simplest place for that knowledge to live is whatever wiki or organizational system you already use, whether that’s Notion, Confluence, or a well-organized Google Drive.

Connect your team plan’s projects to that existing source of truth using connectors. Your AI reads those documents live, even as they change. If you don’t have a great wiki yet, that’s fine. Your day one version is simply a shared folder with as much of your core documentation as you have. You can add clarity on pricing, policies, and brand voice over time. A handful of documents in one shared place beats a perfect system you never actually build.

The Two Layers Every Department Project Needs

Every team project needs the same two layers of context, just applied at the department level instead of the individual level.

Layer one is broad business context: who you are as a company, your team, your products and pricing, your brand voice, your values, and information about your guests. This layer is relevant to nearly everything happening in the organization.

Layer two is department specific. For guest communications, that’s refund rules and FAQs. For marketing, it’s your current campaigns, budgets, assets, and brand guidelines. The goal is to get information out of the head of the department lead and into the project, so your team stops walking over to ask questions the AI could already answer.

You’re rarely starting from a blank page here. If you’ve already built a personal AI brain with a tour business profile or customer avatars, drop that into each department’s project as a running start.

Shared Skills That Don’t Walk Out the Door

In a solo setup, a skill is a repeatable instruction set your AI runs, like a recipe you improve over time. On a team plan, you write that recipe once and everyone on the team can run it, and improve it.

Build a review response skill or an inquiry reply skill, and your admin assistant can use it today. If they leave, the next person who steps into that role has it too. It’s preserved inside the structure of your team plan instead of disappearing with the person who built it.

Multiple team members can use the same tour description skill at once, all producing consistent copy based on the same recipe. When someone finds a better way to run it, they improve it and share it with the rest of the team. Over time, your team’s playbook gets sharper instead of scattered.

Getting Your Team to Actually Use It

This is the part most owners underestimate. Introducing the tools isn’t the hard part. Getting people excited to build with them is.

One practice that beats formal training almost every time: a 15-minute weekly show and tell, where anyone on the team shares something they built, a new skill, a new automation. People learn faster watching a coworker solve a real problem than sitting through a tutorial.

“Travis Pitman at TourRadar ran a hackathon: a couple of hours where his whole team opened their Claude team plan and each tried to build one skill or one automation.”

Travis put a bounty on the ideas leadership was most excited about: $1,000 in travel credit toward any trip booked through TourRadar, for whoever built it. Some team members turned out to have more skill than Travis expected, and they built multi-step workflows that saved the organization real time and money. What stood out to him wasn’t the automation itself. It was that once people got a taste of building something themselves, the intimidation mostly disappeared.

You’ll also find skeptics. Give people every chance to succeed, and keep an eye on who’s genuinely stuck versus who just needs more support.

Where to Start This Week

Pick one department. If your team is in the 5 to 20 range, operations or guest communications is usually the highest leverage place to begin. Set up that department’s layer one and layer two context, connect it to whatever documentation already exists, even if it’s messy, and give your team a weekly rhythm for sharing what they build.

Once that context is in place, your team relies less on you for answers and gets to spend more time actually moving the business forward.

If you want help setting this up in your own organization, book a strategy call with our team. We’re working with tour and activity businesses on exactly this right now.