AI Readiness Audit

Assess the whole business before you invest in AI

We sit with every department and go through the work task by task: what AI can take over, what it pays back, what it risks. You get a work map and a roadmap with the hours and the baht attached — enough to decide, without buying a system first.

A free 30-minute call See a sample output

Work in

AI Readiness Audit+ human approval

To the customer

A person approves before it ever reaches the customer· drag to rotate

Timeline

Usually 1–2 weeks. A small business of a few people can be done in 5 working days; an organisation with many departments and branches can run to 3 weeks.

Your time

About 4–8 hours from your team in total: 30–60 minutes each for interviews, some sample documents (purchase orders, customer chats, the reports you run regularly) and one wrap-up meeting. No slides to prepare, no AI knowledge needed.

Deliverables

6 deliverables

Starts with

A free 30-minute call

Work arrivesNo idea where to start, so nothing startsYour problem

The AI agent does itAI Readiness AuditA work map of the whole company (Task Map)

A person decidesYou review and approveEvery recommendation names where a person approves

Result3–5 workflowsA clear place to start

Example: Assess the whole business before you invest in AI

LIVE DEMO

1 · Work arrives

Everyone says you need AI. Nobody says which job in your company should go first — so the decision sits there for a year.

2 · The AI agent does it

A work map of the whole company (Task Map)
A payback-against-risk score for every task
ROI estimated in baht and hours per month
A data risk and PDPA report
A 3–6 month roadmap: start small, then widen

3 · You just approve

The work map says plainly which tasks AI can finish and which have to pass a human first, especially anything touching money or customers. We never recommend leaving it all on automatic.

Approve in one click

Result

3–5 workflows

A clear place to start

Who it is for

Is this service right for you?

A fit if you…

Owners who know AI belongs in the business, but not where to start so it pays off without breaking anything
Companies that have tried several AI tools and still have none of them inside the real work
Executives who need real numbers for a partner or a board before the budget is signed off
Organisations holding sensitive customer data who need to know where AI is allowed under PDPA

Not yet, if…

If you already know which workflow you want and your existing systems are clear, skip straight to Agentic Workflow Design
If you only want staff to get good with ChatGPT and nothing built yet, AI Skills Training is the better fit

What it fixes

What this service unlocks

Everyone says you need AI. Nobody says which job in your company should go first — so the decision sits there for a year.

You have heard about firms that bought an expensive system nobody used, so signing off a budget without numbers in hand feels reckless.

ID numbers, balances, customer histories everywhere — and nobody can say which of it may legally go into AI under PDPA.

The owner wants it, department heads fear more work, staff fear losing their jobs. The meeting ends with no conclusion because there are no shared facts.

What you get

Real deliverables you can hold

Every task in every department: who does it, how many hours a month it takes, what data it uses, and what share of it AI can take over.

Each task scored on two axes — hours saved, and the damage if AI gets it wrong — then ranked into the order you should start in.

What the first three to five points should return, worked out from your own wage costs and your own volumes, not an average borrowed from somewhere else.

Which data sets are sensitive, which may go to an outside AI, which must stay inside the company, and where a person has to approve every time.

The order to build workflows in, the existing systems to connect, a budget band per phase, and the point where you should stop if it is not paying.

We walk the owner and the department heads through the findings together and answer every concern, so the whole team decides from the same picture.

Steps

How it runs step by step

Usually 1–2 weeks. A small business of a few people can be done in 5 working days; an organisation with many departments and branches can run to 3 weeks. in total

Step 1 / 5 · A free 30-minute call to set the scope

Day one · 30 minutes

1Day one · 30 minutes
A free 30-minute call to set the scope
You tell us about the business, we ask where it hurts and what systems you run, then we agree whether we look at every department or only some.
22–4 days · 30–60 minutes from each of your people
Interviews and watching the real work
At your office or online with the people who actually do the work, watching one job travel through however many hands, systems and files it takes.
33–5 days · you send us sample documents
Building the work map and scoring it
We go through every task we found — hours a month, risk, data used, systems to connect — then run your real sample documents through AI to see whether it can actually read them.
42–3 days
Roadmap and ROI numbers
We calculate baht and hours from real wage costs, write the PDPA risk report, and rank which point should be the first workflow.
5Final day · 1–2 hours
Presenting the findings and deciding together
A 1–2 hour meeting with your leadership team, every question answered. Then you choose: carry on with us, build it in-house, or stop here.

Sample of the output

An extract from the work map you receive

Illustrative numbers from a 10-person wholesaler, to show the shape. Yours will be your own tasks and your own numbers, in the order you should start.

Answering price and stock questions in chat

Hours/month AI can take over≈ 60 of 80 hrs

Risk · orderLow · start here, 1

Drafting quotes from chat

Hours/month AI can take over≈ 25 of 30 hrs

Risk · orderMedium (a person approves before sending) · 2

Keying customer POs into the system

Hours/month AI can take over≈ 35 of 40 hrs

Risk · orderMedium (totals checked before posting) · 3

Daily sales report for the owner

Hours/month AI can take over≈ 20 of 20 hrs

Risk · orderLow · 4

Approving credit for new customers

Hours/month AI can take over≈ 5 of 15 hrs

Risk · orderHigh · stays with a person

Task	Hours/month AI can take over	Risk · order
Answering price and stock questions in chat	≈ 60 of 80 hrs	Low · start here, 1
Drafting quotes from chat	≈ 25 of 30 hrs	Medium (a person approves before sending) · 2
Keying customer POs into the system	≈ 35 of 40 hrs	Medium (totals checked before posting) · 3
Daily sales report for the owner	≈ 20 of 20 hrs	Low · 4
Approving credit for new customers	≈ 5 of 15 hrs	High · stays with a person

Results to expect

Measured in numbers — not feelings

3–5 workflows

A clear place to start

Ranked by payback and risk, naming the one that should be your first workflow in the next 30 days

20–40% of back-office work

Hours you can see

From the cases we have run, paperwork, chat replies and routine reports usually land in this range that AI can take over while people still approve

Baht / month

Numbers to take to your partners

An estimate of the wage cost you get back and the budget each phase needs — enough to approve or reject

Budget

How we charge — straight up

A one-off project fee, based on how many departments and how many people we interview. No ongoing monthly cost, and if you carry on with us into design or build, part of the audit fee comes off the next project.

Solo / small shop (1–5 people)

A short-form audit covering the whole shop in 2–3 days, around the cost of one part-time employee for a month

SME (5–50 people)

Every department covered in 1–2 weeks, about one full-time salary, paid once

Mid-sized company (50–300 people)

Several departments, several systems, sometimes several branches — about two to three full-time salaries, paid once

Enterprise / listed (300+ people)

Quoted to scope, usually split into phases by business line, and including governance and PDPA at company level

Included in the price

Interviews and on-site observation across every department in the agreed scope
The work map, with a payback and risk score on every task
ROI estimated in baht and hours, plus the PDPA data risk report
A 3–6 month roadmap and one presentation to your leadership team
AI tested on 2–3 kinds of your own sample documents, so you see it read them first
30 days of follow-up questions after the report is delivered

Get a quote / book a free call

Guarantees

Safe — you can always back out

The work map says plainly which tasks AI can finish and which have to pass a human first, especially anything touching money or customers. We never recommend leaving it all on automatic.

We sign an NDA before we start. Sample documents are used for testing only and deleted once the report is delivered, and sensitive data can come to us with names removed.

The report is yours. Build it in-house, hire someone else, or stop for now. The audit is meant to be worth it on its own.

FAQ

Before you decide — ask anything

How is this different from sitting down with a normal consultant?

We do not hand over opinions. We sit with the real work, count the real hours, and run your own documents through AI so you see the result first. What you get is a table of tasks with numbers, not a deck of concepts.

We are tiny — an owner and two staff. Do we need an audit?

A short-form version takes 2–3 days and usually pays back fast because the work is simple. But if the free 30-minute call shows there is only one real pain point, we will tell you to skip ahead and build that first workflow instead, to save the budget.

What do we need to prepare? Does the team have to stop working?

Nothing stops. We interview people for 30–60 minutes whenever suits them and ask for samples of routine paperwork such as purchase orders, customer chats and daily reports. The whole project costs your team about 4–8 hours.

Our customer data has ID numbers and balances. Can you look at it?

Yes, with names removed or the sensitive fields stripped first. We sign an NDA, and part of the report tells you exactly which data sets must stay inside the company and which may go to an outside AI under PDPA guidance.

How much can we trust the ROI numbers in the report?

They are estimates from your own wage costs and volumes, with every assumption written down. The real numbers appear once the first workflow runs and is measured every month, and they generally come within 20–30% of the estimate.

What happens after the audit?

Most clients pick one point from the roadmap, take it into Agentic Workflow Design, and have it running for real within 30 days. You can also pause, or build it in-house. The report is yours.

Related cases

See the real thing we already built

Wholesale & e-commerce · SME · 8 people

Services that usually come next

Design work that flows on its own Build it, then hand it over

Start with a free 30-minute call. Decide nothing yet.

Tell us about your day-to-day work and we will say straight out whether your business needs a full audit or just one workflow to begin with. No obligation.

Book a free 30-minute call See all services

Assess the whole business before you invest in AI

Is this service right for you?

A fit if you…

Not yet, if…

What this service unlocks

Real deliverables you can hold

How it runs step by step

A free 30-minute call to set the scope

Interviews and watching the real work

Building the work map and scoring it

Roadmap and ROI numbers

Presenting the findings and deciding together