
Half-day (4hr), in-person, hands-on workshop taking non-technical corporate knowledge workers from zero to running a supervised Claude Cowork task on their own work, using live demos on Ray's own operating businesses (edragon.asia training, durian delivery, bird nest delivery) instead of borrowed case studies.
Programme Outline (4 hrs)
9:00–9:15 Welcome & AI Context 2026 (lecture)
9:15–9:45 Module 1 – What Claude Cowork Is (lecture + live demo)
9:45–10:30 Module 2 – Live Case Study Demo on a real business (live demo)
10:30–10:45 Break
10:45–11:30 Module 3 – Hands-On Lab 1: First Task (hands-on)
11:30–12:15 Module 4 – Department Sprint breakout (Sales/Marketing/Ops-Finance/HR) (hands-on)
12:15–12:45 Module 5 – Governance, Security & Rollout (lecture + Q&A)
12:45–1:00 Close-out: 30-day action plan, upsell offers
Class size: 12–20 pax hands-on. No coding required.
Companies across Singapore are asking the same question: does AI training actually move the needle, or is it another half-day out of the calendar with nothing to show for it? This FAQ answers the questions heard most often from HR leads, department heads and business owners evaluating a half-day Claude Cowork workshop for their team. It covers what the training actually does for productivity and sales, how the session is structured, what it costs, and how it compares to the other AI training options available in the Singapore market. If a question isn’t covered here, that’s what the discovery call is for.
Q: What is Claude, and how is it different from ChatGPT or other AI tools?
Claude is Anthropic’s AI model, available through a chat interface much like ChatGPT — but the workshop isn’t about the chat interface. It’s about Claude Cowork, Anthropic’s agentic product that moves beyond answering questions to completing multi-step work directly inside your team’s files, folders and connected tools. The practical difference: instead of copy-pasting a draft out of a chat window, Claude Cowork opens your spreadsheet, pulls the numbers, drafts the report and flags what needs your sign-off — the way a capable junior staff member would, not a search box.
Q: What is Claude Cowork specifically, and why does the workshop focus on it?
Claude Cowork is the part of Claude built for real work: it connects to tools like Google Drive, Microsoft 365 and Slack, runs multi-step tasks with visible steps and approvals, and keeps working even after you’ve closed the laptop. The workshop focuses on it — rather than generic “AI literacy” — because most Singapore teams have already tried ChatGPT for drafting. The gap is turning that familiarity into tasks Claude actually executes, not just suggests.
Q: Do we need to already be using Claude to benefit from this workshop?
No. Most participants arrive having used ChatGPT or Claude for the odd email or summary, and nothing more. The first module is built for exactly that starting point — you leave with an active, working Claude Cowork setup connected to at least one of your tools, not just theory.
Q: Is this training only for tech-savvy staff, or can non-technical teams use it too?
It’s built specifically for non-technical teams. No coding, scripting or IT background is required. The people who get the most out of it are the ones who deeply understand their own work — sales, ops, finance, HR, marketing — because they’re the ones who can tell Claude exactly what “done” looks like for their tasks.
Q: How exactly does Claude Cowork improve day-to-day productivity?
It removes the manual assembly work that eats a knowledge worker’s week: pulling data from three systems into one report, reconciling a spreadsheet line by line, drafting the same type of email for the tenth time. Claude Cowork does the multi-step assembly and hands you a reviewable result, so your time shifts from doing the task to checking and deciding on it.
Q: What kinds of tasks can Claude Cowork actually take off my team’s plate?
Recurring, multi-step, file-and-data-heavy work is the sweet spot: weekly performance reports pulled from multiple sources, invoice and expense reconciliation, meeting-note summaries and follow-up tracking, document and contract review against a checklist, inventory and scheduling lookups, and first-draft content or correspondence. The workshop’s department-sprint block is built to identify exactly which of your team’s tasks fit this pattern.
Q: How much time can a team realistically save after this training?
It depends on how much of the role is repeatable, file-based work versus judgment calls that need a human in the room regardless. Teams with a lot of reporting, reconciliation or drafting in their week see the fastest, most visible time-back; teams doing mostly relationship-driven or judgment-heavy work see smaller but still real gains. The workshop is built around your team’s own tasks so the estimate you leave with is specific to you, not a generic industry number.
Q: Will this replace staff, or just make them more productive?
The workshop is built around augmentation, not headcount reduction. Claude Cowork still requires a human to define the task, review the output and approve significant actions — it isn’t autonomous in the sense of running unsupervised. In practice, most companies use the freed-up time to absorb growth without proportionally growing headcount, rather than cutting existing roles.
Q: What happens to work quality when AI is doing more of the first draft?
Quality depends on how well the task is set up and reviewed — which is exactly what the workshop teaches. Claude Cowork shows its steps and asks for approval before consequential actions, so nothing goes out the door unreviewed. Teams that treat the AI output as a strong first draft, not a final answer, tend to see quality hold or improve, because reviewers spend their attention on judgment calls instead of on typing.
Q: How does a productivity tool like Claude Cowork actually help drive sales?
Sales performance is bottlenecked as much by admin time as by selling skill — account research, proposal drafting, CRM updates and follow-up tracking all eat into selling hours. Claude Cowork compresses that admin layer, which means more of a salesperson’s week goes back into calls, meetings and relationship-building — the activities that actually close deals.
Q: Can Claude Cowork help with sales prospecting and lead research?
Yes — this is one of the strongest agentic use cases. Given a list of target accounts, Claude Cowork can pull together account research, recent news and relevant context from your CRM or connected tools into a briefing document, instead of a rep manually researching each account for twenty minutes before a call.
Q: What about proposal writing, quotations, and follow-ups?
Claude Cowork can draft proposals and quotations from a template and the specifics of the deal, and can be set up to flag or draft follow-ups when a deal has gone quiet. The workshop’s sales breakout builds exactly this kind of task using your own templates and past proposals as the reference material.
Q: Can it help with customer service and retention, not just new sales?
Yes. Customer support triage, drafting first-response answers from your knowledge base, and summarising a customer’s history before a renewal call are all Claude Cowork tasks — and retention work is often higher-ROI than new-lead generation because the customer relationship already exists.
Q: Is this only useful for B2B sales, or does it work for retail/e-commerce too?
It works well for both. In consumer and e-commerce businesses — the workshop draws live demonstrations from Ray’s own durian and bird nest delivery operations — the highest-value tasks tend to be order and delivery reconciliation, customer WhatsApp/email triage, and marketing content generation, rather than classic B2B account research.
Q: Why a half-day format instead of a full day or multi-day course?
A half-day is long enough to get every participant from zero to a working, supervised Claude Cowork task on their own work, but short enough that a company doesn’t need to write off a full working day per participant. Multi-day courses in the Singapore market tend to spend the extra time on generic theory across multiple tools — the half-day format works because it stays narrowly focused on Claude Cowork and your team’s actual tasks.
Q: What actually happens during the 4 hours?
The session runs five blocks: a short context-setting module on what Claude Cowork is and how it differs from chat-based AI, a live demo on a real business task, a hands-on lab where each participant sets up Cowork and runs their first supervised task, a department-sprint block where teams work on function-specific tasks in breakouts, and a closing module on governance and a 30-day action plan.
Q: How many people can attend, and who should be in the room?
The hands-on format works best for 12–20 participants, with an executive-briefing option (no hands-on lab) for larger groups up to 40. The right people are those who own the work day-to-day and understand why it’s done a certain way — department heads and the staff who actually do the recurring reporting, admin or drafting work, not just observers.
Q: Do participants need to bring anything or prepare beforehand?
A laptop and an active Claude account (Team or Enterprise seat, arranged in advance if the company doesn’t already have one) are the only hard requirements. Participants are also asked to bring a short list of real recurring tasks they’d like to spend less time on — the workshop is built around solving those, not generic exercises.
Q: Can the workshop be run on-site at our office, or only at a training venue?
On-site at your office is the default and preferred format, since it keeps the session anchored in your team’s actual systems and files. A live virtual delivery is also available at the same price if on-site isn’t practical.
Q: Do participants need coding skills or a technical background?
No. The entire workshop is designed for non-technical business users. If a task turns out to need real software development — building an app, a customer-facing chatbot, or backend integration — the workshop flags that as out of scope and redirects to a simpler task, rather than trying to force it through.
Q: What if a task turns out to be too complex to finish in the workshop?
The facilitator redirects to a different, more tractable task from the participant’s list rather than letting the session stall on one problem. A handful of task types are deliberately kept out of the workshop’s scope because they eat the whole session on their own — building full applications, deploying autonomous unsupervised agents, and backend or database integration work. Those are better handled as a separate, scoped engagement once the basics from the workshop are in place.
Q: How much does the workshop cost?
Open-enrollment seats for a shared half-day cohort run roughly SGD 480–680 per participant. A dedicated in-house session for your own team, up to 20 participants, runs roughly SGD 3,800–6,800 flat. Both are benchmarked against comparable training in the Singapore market; full pricing detail is in the accompanying programme outline.
Q: Is this SkillsFuture-subsidised or claimable?
Not directly — SWDA course registration is a separate, multi-month accreditation process that a premium, custom half-day format doesn’t go through. That said, Absentee Payroll and Enterprise Innovation Scheme tax treatment may still apply to company-sponsored attendance depending on how the training spend is structured, and that’s worth confirming with SWDA/IMDA before budgeting.
Q: What’s the realistic ROI on a half-day investment?
The clearest ROI comes from the 2–3 real workflows each participant leaves with already running, multiplied by how often that task recurs — a weekly report or reconciliation task that used to take two hours and now takes twenty minutes pays back a participant’s seat within a few weeks. The workshop is structured around your team’s actual recurring tasks specifically so this calculation is concrete, not hypothetical.
Q: Are there ongoing subscription costs beyond the workshop fee?
Yes — participants need an active Claude plan to use Cowork after the workshop, separate from the training fee itself. This is standard across every Claude-specific training provider in the market, not specific to this workshop, and is worth budgeting into the total cost of adoption up front.
Q: What happens after the workshop — is there follow-up support?
Every participant leaves with a personal 30-day action plan and department playbooks to keep applying what they learned independently. For teams that want more, a full-day Power User session and an ongoing adoption-coaching retainer are available as separate add-ons — the half-day workshop is the on-ramp, not a one-off event with nothing after it.
Q: Is it safe to connect Claude to our company’s tools and files?
Claude Cowork is built with enterprise controls for this specifically: role-based access, per-connector permissions (read-only versus write access can be restricted per tool), and a visible approval step before consequential actions. The governance module in the workshop walks through exactly what to configure before a wider rollout.
Q: Who controls what Claude is allowed to access or do?
Your admin does. Claude Cowork’s enterprise controls let an administrator assign users to groups, set per-connector permissions, cap usage budgets by team, and pull audit logs of every tool call and file operation. Nothing is granted by default beyond what’s explicitly connected and approved.
Q: What about data privacy and confidentiality for our clients’ information?
Data handling, retention and confidentiality are governed by your organisation’s Claude plan agreement with Anthropic, not by the workshop itself. The governance module covers what to check with your IT/compliance function before connecting systems that hold client or personal data, so that conversation happens before rollout, not after.
Q: How is this different from generic AI/ChatGPT courses available in Singapore?
Most AI courses in the Singapore market are multi-day, multi-tool and generic — built to cover ChatGPT, Copilot and Gemini broadly rather than go deep on any one of them. This workshop is narrowly built around Claude Cowork specifically, runs in half a day instead of two or three, and is structured around your team’s actual recurring tasks rather than a fixed generic curriculum — with live proof drawn from real operating businesses, not borrowed case studies.
Q: We've already sent staff for a generic AI course before and it didn't stick — how is this different?
The usual failure mode is a course that teaches prompting in the abstract, with no connection to what the participant actually does at their desk on Monday morning — so the skill fades within weeks. This workshop is structured to avoid that specific failure: every hands-on hour is spent on the participant’s own real tasks, using their own files and tools, so what they leave with is a working setup already solving a problem they had before lunch, not a slide deck of prompting tips.
Q: How do we get started?
Start with a short discovery call to walk through your team’s structure, the tasks you want to target, and whether the open-enrollment or in-house format fits better. From there, workshop scheduling and any tool-access setup needed ahead of the session can be arranged within a couple of weeks.
The companies getting the most out of AI right now aren’t the ones with the biggest budgets — they’re the ones who got a structured half-day into every department’s calendar before their competitors did. If your team is spending hours a week on reporting, reconciliation, drafting or admin that follows the same pattern every time, that’s exactly what this workshop is built to fix. Book a discovery call to scope your team’s session.

Copyright © 2019 eDragon - All Rights Reserved
Powered by Edragon
We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.