Do I need to know what should be automated?

No. Most people arrive with a feeling rather than a specification — something takes too long, or only one person can do it, or it gets dropped whenever the week is busy. That is enough to start from. If you can point at one job you would gladly stop doing by hand, we can work with that. If you can’t point at one yet, start with the free diagnostic instead: it finds the constraint first, and the workflow follows from it.

What happens with the first workflow?

We build it. You describe the job the way you would explain it to somebody starting on Monday — what comes in, what you would check, what should go out, and what you would never let leave without reading it. That becomes a working workflow, connected to the tools you already use, running on your account. You watch it run on real work before you rely on it. One workflow, start to finish, so what you are judging is a thing that works rather than a proposal for one.

What happens after that?

It runs. Keeping it running is the recurring part — hosting, the models it calls, monitoring, and the fixes for when something upstream changes its mind about a field name. Most people then add a second workflow, and it is usually one they spotted themselves while watching the first one work. There is no bundle to commit to up front. The next workflow is a decision you make when you want it, and not before.

Can my team eventually build workflows themselves?

That is the intended end state, and the training exists for it. Once your team has watched a few workflows get built, the skill turns out to be the describing — knowing what the work actually involves and where the judgment sits — which is the part your team already has. Some businesses would rather we kept building, and that is a legitimate choice rather than a failure. Most want at least one person inside who can change something without picking up the phone.

Do I become dependent on hmn.plus?

It is the right question to ask any vendor, and the honest answer is that you can be for exactly as long as it is useful. Three things are deliberate: the workflows are described in plain language rather than locked in a format only we can read, they run on infrastructure that can sit in your own name, and we teach the building instead of keeping it. Being the only people who can change your systems would be a business model. It is not ours.

Can we eventually own the system?

Yes, and it is an option rather than a destination — plenty of businesses would rather we kept running it, and that is a reasonable thing to want. When you do want it, the workflows, their definitions and the infrastructure transfer to you, and your team keeps the training that makes them usable. We would rather you stayed because staying is worth it.

The first automation is a fixed price, so the first decision is a small one: NZ$500 + GST to map it and get it built. After that, the platform it runs on is NZ$150 + GST a month for each person who builds workflows — everyone else who uses it is free — and that covers hosting, running your workflows, monitoring and support. Workflows you build yourself cost nothing extra; the ones we build are quoted one at a time, on what each actually involves. No annual lock-in, and nothing larger to sign before you have seen something of your own working.

Why not use ChatGPT, Claude, Zapier or n8n?

They answer a different question. Chat tools put something capable on the other end of a conversation, and then nothing happens unless a person types — every session starts again, and what the business knows ends up living in whoever writes the best prompt. Zapier and n8n wire apps to each other; the wiring holds until a field name changes, and it never makes a judgment call. Both leave the work depending on a particular person. What we build has a fixed shape: the same steps in the same order, your rules written in, a record of every action and the reason for it, and a person standing where a person should stand. The thinking inside a step can be as sharp as it needs to be. The path through it does not change because somebody phrased something differently. That is the difference between a tool one person is good at and a system a business runs on.

Where should AI actually be used?

On the judgment inside a step, not on the whole step. Reading an enquiry and working out what kind it is: yes. Drafting a reply in your voice, from your own past replies: yes. Moving money, or telling a client something binding, with nobody looking: no. What we build to is that anything with consequences waits for a person, and everything in front of it does not have to. A faster version of a bad process is still a bad process, which is why deciding where AI belongs happens before anything gets built, and is most of the value.

ChatGPT for business: uses, risks, workflows

ChatGPT for business is best used as an assistant for people, not as the operating system for the company. Use it for drafting, summarising, research support, ideation, and decision preparation. Move repeated, customer-facing, or risky work into owned workflows with rules, records, and human approval where needed.

That distinction matters because ChatGPT can feel like a business process before it actually is one. A useful prompt becomes a weekly habit. A weekly habit becomes part of sales, delivery, hiring, or customer service. Then the business depends on a private chat session that only one person knows how to run.

We are not anti-ChatGPT. We built hmn.plus because AI assistants clearly make capable people faster. The problem starts when a founder mistakes speed for control.

What ChatGPT is good for in business

ChatGPT is strongest when the task is human-led and the final judgement stays with a person. You bring the context. It gives you a faster first pass.

Good business uses include:

Drafting: emails, proposals, policies, job ads, internal updates, and customer replies.
Summarising: meeting notes, long documents, customer feedback, transcripts, and research notes.
Thinking: options, angles, objections, scenarios, and rough plans.
Research support: starting points, questions to ask, and summaries you still verify.
Decision preparation: trade-offs, risks, checklists, and comparison tables.
Editing: clearer wording, stronger structure, shorter copy, and tone adjustments.

This is where chat works well. You ask. It answers. You correct it. It improves the draft. The loop is fast and easy for non-technical people to understand.

At hmn.plus, we separate this from operations. Chat is a conversation. It creates an answer. It does not create a dependable business process unless something outside the chat owns the rules, records, approvals, and handoffs.

Where ChatGPT breaks for business operations

The risk is not that ChatGPT gives bad answers all the time. The risk is that it gives useful answers often enough that people start building work around it without noticing.

That is fine for experiments. It is weak for operations.

It depends on one person doing the ritual

A chat-based workflow usually lives in someone’s head. They know which prompt to use, what to paste in, what to ignore, and where to copy the output.

If that person is busy, away, or leaves, the process becomes fragile. The company does not own the workflow. The employee does.

For a one-off draft, that is acceptable. For follow-ups, billing checks, service delivery, hiring steps, or customer support, it is not.

It hides the rules

Prompts often contain business rules, but they rarely manage them well. A prompt may say which customers need a different response, which issues need review, or which tone is acceptable. Over time, someone edits the prompt. Someone else uses an older version. A third person adds context from memory.

Now the rule exists, but the business cannot clearly say where it lives.

Governance you write down but do not enforce is theatre. It looks safe. It is not. A business process needs rules that run the same way every time, not instructions scattered through chats and saved notes.

It gives weak audit trails

When something goes wrong, you need to know what happened. What data went in? What instruction was used? Who checked the output? Was anything changed before it reached the customer? Did the AI make a judgement it should not have made?

Chat sessions are poor places to manage that kind of operational record. They are useful working spaces. They are not reliable control systems.

This matters most when the output affects money, customer trust, legal risk, staff decisions, or delivery quality.

It spends AI effort on work rules should handle

Many businesses use ChatGPT to do tasks that should not need AI at all. Routing, formatting, required-field checks, naming conventions, simple classification, and basic handoffs can often be handled by clear rules.

If you ask AI to think through every small step, cost and inconsistency rise together. The better approach is simple: use rules for predictable work and AI only where language, judgement, or synthesis is genuinely useful.

It blurs accountability

AI can make responsibility feel slippery. The employee says the assistant wrote it. The owner says the employee should have checked it. The customer only sees the mistake.

Your business needs clear boundaries. AI can draft, summarise, suggest, and prepare. It should not quietly decide things that require business accountability unless the workflow has explicit rules and approval points.

The better model: assistants for people, workflows for the business

The simplest rule is this:

Use ChatGPT to make people faster. Use owned workflows to make the business repeatable.

An owned workflow is not just a better prompt. It is a defined process with inputs, rules, outputs, approvals, and records. It can still use AI. It just does not depend on a person remembering the right chat ritual.

A dependable AI workflow should:

Run without one person needing to be logged in.
Use documented rules instead of remembered prompts.
Check required information before work continues.
Escalate when a decision is sensitive or uncertain.
Record important actions and approvals.
Survive staff changes.
Use AI only where AI adds value.
Avoid being trapped inside one subscription or assistant.

We built hmn.plus around this distinction because most automation fails at the boundary between helpful output and repeatable work. The issue is rarely the model alone. The issue is the missing operating layer around it.

A practical example: sales follow-up

Take a common business task: following up after a sales call.

The chat version looks productive:

A team member pastes call notes into ChatGPT.
They ask for a summary and follow-up email.
They edit the draft.
They copy it into email.
They update the customer record if they remember.

The output may be fine. The process is still fragile. No one can easily see whether the right notes were used, whether important risks were flagged, whether the customer record was updated, or whether the follow-up matched the company’s rules.

The workflow version is different:

The call notes enter a defined process.
The system checks for required details.
AI summarises the parts where language judgement helps.
Rules decide whether the message is standard, sensitive, or needs review.
A draft is prepared in the right structure.
A human approves it where approval is required.
The action is recorded.

The customer may see a similar email. The business risk is completely different. In the first version, the company depends on discipline. In the second, it has a process.

What should stay in ChatGPT

Not every AI use needs a workflow. Overbuilding is its own problem.

Keep work in ChatGPT when:

It is occasional.
The stakes are low.
The user owns the final judgement.
The input is not sensitive.
The output does not need a formal record.
Variation is acceptable.

Examples include rewriting a paragraph, brainstorming a campaign angle, preparing questions for a meeting, summarising a public article, or exploring a rough idea.

These are good uses of chat. Treat the output as a draft, not as an operational fact.

What should become an owned workflow

A task should move out of ChatGPT when it becomes repeated, relied upon, or risky.

Use this checklist:

Frequency: Does this happen every week or every day?
Dependency: Would work stop if one person was unavailable?
Risk: Could a wrong output harm a customer, payment, delivery, or decision?
Consistency: Do the same rules need to apply every time?
Visibility: Would you need to prove what happened later?
Cost: Are you using AI for repetitive steps rules could handle?
Scale: Would the process break if volume doubled?

If several answers are yes, chat is the wrong home. The work needs structure.

How to use ChatGPT safely in a business

You do not need a huge AI policy to start well. You need clear boundaries your team can follow.

1. Define allowed and restricted use

Tell people what ChatGPT is good for in your company. Also tell them what must not go into a chat window. Sensitive customer information, private staff details, financial records, and contractual material need stricter handling.

If your rule is vague, people will guess. Busy people guess differently.

2. Require human approval for customer-facing work

AI can prepare a customer reply. A person should own the final message. This is especially important for complaints, pricing, commitments, technical advice, and anything that could create a promise the business must keep.

3. Store the final output where the business works

If a ChatGPT answer matters, it should not live only in the chat history. Put the final version in the right customer record, project note, document, or workflow log.

The business should be able to find the decision later.

4. Review repeated prompts

Ask your team which prompts they use again and again. Repetition is the signal. A repeated prompt may be a business process trying to happen.

Do not start by polishing the prompt. Start by asking what rule it is trying to apply.

5. Move the risky parts first

You do not need to rebuild everything. Start with the places where a bad output would hurt: customer trust, money, compliance, delivery quality, or staff decisions.

That gives you the most protection without slowing useful experimentation.

Is ChatGPT good for business?

Yes, when you use it in the right layer.

ChatGPT is good for business when it helps capable people think, write, summarise, and prepare faster. It is not good as the hidden backbone of repeatable operations.

The mistake is treating a chat window like a process engine. It feels efficient at first because the work gets done. Then the business grows, more people copy the habit, and no one can clearly see the rules, records, approvals, or handoffs.

If ChatGPT is already helping your team, keep using it. Just sort the work. Keep low-risk, human-led tasks in chat. Move repeated and risky work into owned workflows.

That is how AI becomes useful without making the business fragile.