For owner-operated service businesses in Queensland

Don’t make the right person do the wrong job

Most growing service businesses answer the same workload with another administrator. The hire is usually right. The job they are hired into is not: it still carries the renaming, the repetitive copy-pasting, constant re-prompting, and chasing follow-ups that grew along with the business. So the new person spends a third of their week on work that they shouldn’t have to do in the first place.

Find out how you can trim the “task fat”: the messy, boring admin that robs people of the focus needed to concentrate on work that directly brings in revenue for the business.

No cost, no obligation, held over Google Meet to see how we can help.

A$2,500 Fixed price for the scopeOne role, quoted upfront, not an hourly estimate.
3–5 days From start to a written scopeThe task-by-task breakdown based on samples of the final product the work should produce, and the role restated as it should stand once that work is removed.
100% Refunded if it does not deliver thatWe suggest and design the scope where AI can be implemented safely.
01

Whether I can help you, and where I cannot

Short enough to check yourself, right here.

Likely a good fit if these are true

  • You run the business and can approve a change to how work is done
  • Someone capable is spending a large part of their week on work that is not why you have them: the same details keyed between a CRM and an inbox, the same follow-ups chased, the same documents put together
  • Something has just changed the load: a role advertised, an administrator leaving, a virtual assistant added, a system swapped, a second office
  • You have a steady stream of leads from marketing that converts into sales, and you serve at least several clients
  • Turnover is reasonably high, without wild fluctuations
  • You are independently run, with one location or a few
  • Your industry has rules and a governing body, but AI can still be implemented in your operations so long as it is compliant

Probably not (yet) if

  • Their week already goes on the judgement, the negotiation and the relationships, which is the role doing what it is there for
  • The volume is annoying rather than costly. Hire a virtual assistant
  • Nobody is available to explain how the process actually works
  • You are an enterprise, a franchisee, or an authorised representative of another business, with several layers of approval such as procurement or operations
  • There is no direct and reliable access to the people who decide, or replies take weeks
  • The business is operationally immature, with inconsistent enquiry-to-sale volumes
  • The main problem is a marketing one, meaning lead generation: SEO, UGC, content

Most of the businesses this fits are independent service businesses with just some regulatory hurdles: clients served include those from real estate, property management, NDIS, trades, and small-office agencies. The industry matters less than the shape of the work.

02

I was hired to add renewal capacity. Renewals were still going out late after I arrived

I worked as a broker assistant for an insurance brokerage in Toowong, Brisbane, and my job included renewals, insurer portals, quote comparisons, documents, follow-ups and deadlines. Much of it was the repetitive work that kept a book of clients intact.

Renewals were usually late, and the reason was not capacity

In this industry, a client is meant to be notified with enough time to read a renewal and decide on it. Renewals there were regularly being finalised inside the last 14 days before expiry. I had been hired as a helping hand to expedite that exact process, but even after I arrived, renewals were being processed at the same speed per person. Adding another person was never going to improve the underlying process that caused the slowness and the chaos in the first place.

That is not an internal inconvenience. A client given a few days to consider a renewal is a client more likely to let it lapse, or to take a call from somebody else. Late renewals become cancellations, cancellations become churn, and churn arrives two quarters later looking like a marketing problem.

Part of the operation ran on multiple systems that hardly talked to each other: Outlook, WinBEAT, Sunrise, and a shared file store whose folders had been arranged differently depending on who had made them. Hours went on hunting for emails whose names were formatted inconsistently. Renewal calculations were done by hand in an Excel spreadsheet, after copy-pasting the figures out of legacy insurance software.

The other part of it ran on paper, including a bound book of hand-ruled tables with client names written in by hand. The rest ran on what individual people happened to remember, which made it reliable exactly as long as they were at their desk. None of that was anybody’s fault. It accumulated the way it does, because daily routines and deadlines mean nobody has the time to clean up operational mess.

Every answer on the table added load to the same process

Instead of fixing the underlying issue, the director considered band-aid fixes: another broker assistant, a Salesforce licence, and a cold-caller to generate leads and replace the clients who had gone. Every one of those seems like a reasonable thing to buy.

The software was the clearest case. You can buy the CRM, but if the work still moves on printed paper and in someone’s head, you have bought a digital layer over a paper workflow and kept that workflow. Salesforce was also too complicated for a smaller brokerage that needed something easy to use. It created more steps than it removed.

More leads had the same shape. A business that struggles to serve its existing clients does not improve when you add clients to it: the new work enters the same sequence, service slips further, and more clients leave. Before asking how to win more customers, it is worth asking whether the business can absorb the ones it would win.

Frustratingly, I was not in a position to change any of this. I was close enough to see the work up close, and when I asked why it was run this way the usual answer was that this is how it has always been done. It was not long before I realised I could provide more value by addressing workflow inefficiencies than by inheriting them. So I left my job to do that part.

You cannot automate everything, and you do not need to

Before I left, a colleague raised the obvious objection: you cannot automate everything. That is true, and it is not the question. You do not have to remove all of a job for the economics of that job to change. If a role is mostly repetitive coordination, moving information between systems, chasing, preparing standard documents, updating records, watching dates and triaging routine requests, and most of that comes out, what is left may no longer justify a whole additional person.

Say 10-20% of it genuinely needs judgement, a relationship, or somebody to carry responsibility for an exception. Fair enough. The question then becomes whether that remainder warrants a hire, or whether the existing team absorbs it once the same drag has been taken out of their own weeks. The human work does not disappear. It moves to the people whose judgement and relationships are the reason you are paying them.

Every response to it looks responsible

None of that was specific to insurance. It is what happens to any business whose process grew by accretion while the client list grew.

The same shape turns up wherever the work recurs and the client list is the asset. In real estate it is a contact list living in a spreadsheet, property images cropped by screenshot, copy drafted in a chat window and pasted into Gmail by hand, one contact at a time, a few thousand times over. In trades and field service it is job sheets, variations and invoices going out late because what is needed to raise them sits with whoever did the work. The vocabulary changes. The sequence, the retrieval by hand and the reliance on somebody remembering do not.

Even the way most people use AI creates more work than it removes. A model drafting a custom message does not replace the manual work of copy-pasting that copy, addressing the recipient, uploading the attachments and recording somewhere that it all went out. Multiply that by 2,500 recipients and you have a time sink that puts a hole in the week.

All of this falls under the radar because the existing solutions hide the inefficiency. A CRM or an admin hire works well on good foundations and poorly on bad ones. Solutions get looked at before problems, and that ordering is backwards.

However, businesses usually do not have the time or the energy to analyse their own workflows, because the day-to-day comes first. That is exactly where I come in. I free up capable people to spend more of the week on the work that directly generates revenue.

A job, at its core, is just a bundle of tasks. What becomes redundant is the task, not the person. A role with its repetitive half removed is the same person doing more of what you hired them for, which is the point of having hired them.

Hiring for the role as written

Ramp-up
Weeks to months learning the role, with your team absorbing the training load.
Mistakes while learning
Costly twice: the redone work, and the client whose first impression was a slower month.
Ongoing cost
Salary, super, leave and management time, whether or not there is enough work to fill the role.
Who carries the risk
Mostly you. You find out in month three, with someone already in the job.

Scoping it first

Ramp-up
Tested against your actual workflow before go-live. The learning curve happens once, not again with every hire.
Mistakes while learning
Caught against a defined standard before anything runs unsupervised.
Ongoing cost
A fixed fee for the work actually required, and a lower ongoing cost once built.
Who carries the risk
Guaranteed at every stage. You get the fee back if the scope does not hold up.
03

As the old saying goes: Time is Money.

Put a figure on the role before you judge anything else on this page

What follows is that saying with your own numbers in it. Three you already have, and every judgement below is easier to make against a figure than against an impression.

hours
people
A$an hour

An hour is worth something whether or not anyone is paid for it. Your own time has a value. An hour belonging to someone who bills for their time is worth what they bill. An administrator on A$65,000 is near A$45 an hour once super and overhead are counted.

At those figures, that time is worth about A$35,100 a year: 5 hours a week each × 3 people × A$45 an hour × 52 weeks Said as a role, that is 39% of a full-time administrator 15 hours a week across the team, against a 38-hour full-time week

That is what the way you work now costs. It is not the cost of a problem. Some of that work is why the role exists and always will be. The rest is time the role was never created for, and the scope is what tells you which is which.

The percentage is the figure worth keeping. It comes from the hours alone, against the 38-hour week the Fair Work standard sets, so it holds whatever you think the rate should be. Every result For The East reports is expressed the same way: the share of a role returned, against the role’s own cost.

04

What the whole path costs, before you read the parts

Priced up front so the total is known on day one, not discovered on the second invoice.

Founder pricing. The founder rates below are locked in for the first 3 clients For The East takes on. Once that group is filled they move to standard pricing: the crossed-out figures shown against them. If you start before then, your rates are fixed for your engagement and do not move with them.

Step one: Narrow the Scope of Human LabourWhat the role should contain, and what should leave it. Was , now A$2,500
Step two: Build & RolloutOnly if the scope says it is worth building. Was , now A$2,500
Both stages, start to live Was , now A$5,000

The ongoing retainer, which covers maintenance, upgrades and continued strategic work, is separate and optional, though strongly recommended. It is charged monthly rather than as part of this, at A$1,000. The scope is half of that total. The path above runs in two stages, each with its own set timeline. Step one settles what is feasible, realistic and safe to build, and scopes what stays human and what does not. If it concludes the role should be filled as written, the engagement ends there and the second line never happens. Step two is the build, which runs in four phases across ten business days and begins once that design is settled in full. This is the whole of it: there is no third invoice between the scope and a working system.

The fee you agree is fixed for your engagement. A later change to published pricing does not reach back into it. Work outside what the scope covered is quoted separately. It starts only once you have approved it in writing, so nothing is folded into a stage you have already agreed.

All figures are in Australian dollars. No GST is charged.

05

The scope answers three questions, and the third is what makes the build safe to commission

A breakdown of the role on its own would still leave the build to be worked out on the way. The same week of work settles three things: what the role should contain, what to build, and how to switch it on. That is what makes the second stage a defined piece of work rather than an open-ended one.

What the role should contain

A task-by-task breakdown of the role: which parts need a person’s judgement, which are repetitive information processing, and what share of the week each accounts for. The output is the role restated as it should stand once the automatable work is removed, with the hours that frees stated against the role’s cost. Usable to brief whoever holds it now, or to advertise it if you are hiring.

What should be built, if anything

The removed work written down as it actually runs: the decisions inside it, the exceptions that break it, and the accuracy standard it has to meet before it goes near a client. A specification a developer can build from, not a brief they have to interpret. It also names every point where the work touches an obligation you have to meet, and records your answer against it. You know your industry’s rules. What goes wrong is that nobody asks you at the moment the automation is designed, so an assumption gets built in and surfaces later.

How it gets switched on

Which of your systems connect, what access each connection needs and who inside your business grants it, and the order the rollout runs in. Staged so you can stop it, and worked out on paper before anything is asked for.

The scope is useful whichever way it lands. A role that should be filled as written still leaves with a better job description than it arrived with.

06

The risk sits with For The East in writing

For The East is taking on its first 3 clients at founder pricing. Every stage carries a written guarantee rather than a testimonial. These are the terms, not a summary of them.

Narrow the Scope of Human Labour

If the scope does not produce a task-by-task breakdown of the role, the share of it that is automatable, the role restated as it should stand once that work is removed, and the workflows inside it that carry that work, the fee is refunded in full. It guarantees you a scope, not a particular answer.

Build & Rollout

We iterate to the accuracy threshold agreed in writing before the build starts. If it is not met within 30 days of the build starting, the fee is refunded in full. Ten business days from start to live, and only one build runs at a time.

The Standing Operations Partner

Each expansion is guaranteed to reach the same result the original build achieved, or that unit is not charged. Available as a performance-linked arrangement against an avoided-hire baseline agreed in advance, if you would prefer that structure.

Theo Jeong
Theo Jeong Founder, For The East

I run the scope and write it myself. The guarantees above are mine to honour, whoever does the building. The ABN in the footer is there so you can check who you are dealing with before any of it starts.

07

A virtual assistant is the cheaper answer while the work is small, and the more expensive one once it is not

Two alternatives come up on nearly every call. A virtual assistant is often the right decision, and this page says so where it is. An IT provider is not an alternative at all, because the job is a different one. It is worth being exact about where each stops.

Hire the VA if the work is genuinely small

A few hours a week, irregular enough that it benefits from a person deciding what to do next, and a VA is simpler and cheaper than building anything. At low volume that is the commercially correct answer, and the honest recommendation is to stop reading and go and hire one.

The wage stays cheap. The operating model does not

When the same repetitive work grows with volume, the VA model scales one way: more hours, then more people, more handoffs, more supervision, more waiting on a time zone. None of that appears on the invoice. It appears in the weeks of the people you already pay the most, which is the most expensive place for it to sit.

A person is doing the work, and by the sixth hour it shows

The work owners hand over first is simple, repetitive, rules-heavy and digital, which is exactly the work that wears a person down fastest. Accuracy falls as the hours accumulate, so someone has to check it. Anything off the predictable path returns as a question rather than a decision.

The cost nobody counts is the decision itself. Each time volume steps up, the same question returns: who to hire next, at what rate, who trains them, who checks their work. That is your own time, and it goes on staffing instead of on revenue. A system is specified once and changed deliberately after that. It does not resign, and extending it does not begin with a search.

Your IT provider is not the alternative here

What an IT provider owns

The foundation
Network, devices, accounts, security and backups: the ground the business stands on.
The individual systems
Each platform configured, licensed, patched and kept running.
Where it stops
They can reasonably expect the business to tell them what the workflow across those systems is supposed to achieve.

What this work owns

The layer above
The workflow running across those systems: what moves, when, under which rules, and what happens at an exception.
The translation
Your process written down as system logic, so the technical decisions are made once, on paper, rather than learned by you.
Where it stops
Anything infrastructural, security-related or inside a single system stays with your IT provider, which is the right place for it.

They lay the foundations and supply the bricks. This is the work of putting the bricks together into something that runs. A good IT provider makes that easier rather than unnecessary, and the two rarely touch the same task.

08

What happens after you book

01

Pick a time and answer a few questions about the workload, so the call starts with the context already in hand.

02

30 minutes over Google Meet, working out whether there is enough repeating work here to be worth fixing.

03

A recommendation: the paid scope, direct implementation where an existing system already fits, or that the role should be filled as written.

04

You decide. Nothing beyond the stage you agree to is ever assumed.

09

Bring the role. 30 minutes settles what it should contain

You do not need to know what should be automated. Bring the job description, or the workload, and that is the part being worked out.

Not your decision alone? Write to info@fortheeast.com.au with the workload in a couple of sentences, and I will send you the full overview to forward to whoever else has to agree.