Teams on a ChatGPT Business Plan
Groups on ChatGPT Business or Enterprise, or planning to be, who want a shared standard for how it is used.
CHATGPT COURSE
ChatGPT Course for Teams
A practical ChatGPT course for teams: prompts that work first time, images made for work, custom assistants for recurring tasks and small tools built with Codex, with the habits that keep company data safe.
Trusted by teams in best-in-class organisations
Everyone has tried ChatGPT. Few teams use it the same way.
Facing issues such as:
- Prompts that work for one person and fail for the next
- Marketing and internal images that look off-brand or carry errors in the text
- The same task rebuilt from scratch every week instead of saved as a custom GPT
- Spreadsheet answers that nobody has checked before they reach a report
- Staff using personal accounts, with no clear rule on what to paste in
Prompts, images and a custom GPT, built on your own work.
-
Prompt with a method
Write prompts with a role, task, inputs and format, so results are usable the first time and repeatable by colleagues.
-
Create images you can use
Brief ChatGPT for images, edit them in rounds, and check them for errors and brand fit before they go into a deck or a post.
-
Build small tools, even if you are not a developer
Make a custom GPT for a recurring task, and use Codex, OpenAI's coding agent, to build a simple tool or automation.
Who is this course for?
Who the ChatGPT course is for
Marketing, Sales, HR and Operations Staff
People with repetitive writing, analysis and visual work who want to hand the routine parts to a well-instructed assistant.
Managers Who Want Output They Can Trust
Leads who want staff to check results and follow data rules as well as work faster.
Course Overview
What you build in the ChatGPT course
Workshop Agenda
- Write prompts that work first time, with a clear role, task, inputs and format
- Create and edit images for work from a written brief, and use them brand-safely
- Build a custom GPT for one recurring task and test it on real examples
- Analyse a spreadsheet and sanity-check the result before you trust it
- Use Codex to build a small tool or automation, without being a developer
- Follow your team's data rules and check outputs before they leave the team
Each module is a short demo in ChatGPT, a task on your own laptop and a debrief. You try the task your way first and compare it with the method: a structured prompt, a tighter image brief or a tested custom GPT.
Practice material is fictional or non-confidential, including a dataset with planted errors. Before the course we ask which tasks your teams handle, so the exercises match their work. OpenAI changes ChatGPT often, so we teach what your company's plan includes at the time of training.
Course Formats
Format
- Live Workshop
- Hands-on on Your Own Laptop
Course Duration
- Half Day (About 4 Hours)
- Full Day
- 2-Day
Where It Runs
- Usually on your own ChatGPT plan and workspace
What to Bring
- A laptop
- Your company's ChatGPT plan
Tools we cover: ChatGPT, custom GPTs, image creation in ChatGPT and Codex.
ChatGPT Course curriculum (half day, 7 modules)
Roughly four hours in ChatGPT with a 15-minute break; timings are approximate. Longer runs of a full day or 2 days give more build time for custom GPTs, images and Codex tools on your own material.
Module 1: How ChatGPT Works and Where It Fails (25 min)
Talk and Discussion
- Share where you use ChatGPT now, where it helped and where it let you down
- Why it sounds sure when it is wrong, and what that means for checking
- Name the real task you will work on during the session
Module 2: Prompting That Works (35 min)
Talk and Demo
- The prompt shape: role, task, inputs and format
- Add the detail only you know, so output reflects your context
Hands-on
- Take a weak prompt on your own task, rebuild it, run both and compare
- Leave with a prompt cheat sheet
Module 3: Creating and Editing Images for Work (45 min)
Talk and Demo
- Write an image brief: subject, style, format, any text and what to avoid
- Edit in rounds instead of starting over, and keep a consistent look across a set
- Brand-safe use: check text and details for errors, avoid copying other brands or real people, and follow your company's rules on AI images
Hands-on
- Brief and refine an image for a real need, such as a slide visual or a social post
- Edit an existing image through chat and compare the versions
- Put your images into slides, then take the deck further with AI for presentations
Module 4: Custom GPTs for Recurring Work (35 min)
Talk and Demo
- What a custom GPT is: instructions, reference files and one clear job
- Where Projects fit, and when a custom GPT is the better choice
Hands-on
- Build a custom GPT for one recurring task of your own
- Test it on three real examples and fix what breaks
Module 5: Data Analysis, Checked Before You Trust It (35 min)
Talk and Demo
- Upload a spreadsheet, ask for analysis and see what ChatGPT did to get its answer
- Sanity-check first: totals, duplicates and odd values
Hands-on
- Analyse a practice dataset with planted errors, then compare what you caught with the answer key
Module 6: Building Small Tools and Automations with Codex (40 min)
Talk and Demo
- What Codex is: OpenAI's coding agent, which can write, run and edit code for a task you describe
- What non-developers can build: a data clean-up script, a report formatter or a simple internal web page
- Reading what it did, testing the result and knowing when to bring in a developer
Hands-on
- Describe a small tool for your own work and have Codex build a first version
- Review the changes and ask it to fix one problem
Module 7: Safe Use and Next Steps (10 min)
Wrap-Up
- Workspace and data settings, what not to paste, and how to verify output
- Commit to the next task you will turn into a custom GPT or a tool
Contact Us for Pricing
Request a Quote Worked example
Build a custom GPT for one recurring task
What participants do in Modules 2, 4 and 6, shown on a fictional sales team that writes the same kind of follow-up email every week.
-
Pick one task you repeat
Turning call notes into a client follow-up email in the team's format: frequent, written the same way each time and easy to check.
-
Write the instructions
Using the prompt shape from Module 2, the GPT gets a role, the task, the inputs it will receive and the output format, plus one example of a good email.
-
Test it on three real examples
Three sets of practice notes go through it. One usually fails: a missing detail, the wrong tone or a next step the notes never mentioned.
-
Fix what broke
Participants add the rule that fixes the failure, for example "if the notes do not mention a next step, ask for one instead of inventing it", then run all three examples again.
-
Go further with Codex
When a task outgrows chat, such as reformatting a whole export of call notes, Module 6 shows how to describe a small tool to Codex, review what it built and ask it to fix one problem.
Your Facilitators
Learn from experienced practitioners
Our programmes are led by facilitators that bring real, practical industry experience so you get the best outcomes.
Eugene Cheng
Managing Director, HighSpark
Forbes30U30 Asia Alum, seen on CNA, Inc, Business Insider. He's helped clients raise over US$480M in funding. Highly sought after advisor to C-Suites from firms like Nike, Prudential, Oracle.
Frequently Asked Questions
Image creation, custom GPTs and Codex are not equally available on every plan, and limits differ. Tell us what your company has and we will teach the features it includes at the time of training. Business plans generally give your company more control over data settings than personal accounts do, and your IT team should confirm the settings you rely on.
Fees depend on format, group size and tailoring, and certificates are available on request upon successful completion. Cost, data and funding answers for every AI course are in the Generative AI courses FAQ.
Product names and logos are trademarks of their respective owners, used for identification only. HighSpark is an independent training provider and is not affiliated with, sponsored or endorsed by them.
Using Claude or Copilot instead of ChatGPT?
Each tool course follows the same hands-on rhythm. Choose the one your company has approved.
- Claude AI Training: for teams on Claude, with Projects for shared context, long-document work and Claude Code.
- Microsoft Copilot Training: for companies rolling out Microsoft 365 Copilot across Outlook, Teams, Word, Excel and PowerPoint.
- AI at Work: for mixed teams on more than one AI tool, or still deciding.
Curious? Speak to our expert team.
Fill out the brief form and we'll get back to you with an estimate and discuss a plan for your company.
We'll arrange your meeting with a consultant to:
- Provide cost estimates
- Provide specific advice
- Identify any blindspots
- Share best practices
- Discover your needs
Our flagship business storytelling training workshop enables teams to utilise narratives to drive business and team goals.
Equip teams to design clear, compelling visuals that sharpen messages and help audiences grasp ideas quickly and confidently.
Helps professionals turn complex data into clear, persuasive stories that support insight-driven decision-making.
Plan a ChatGPT course for your team
Tell us which ChatGPT plan you use and what your team works on. We will suggest a format and send a quote.
Plan a ChatGPT Course