Straight answers about working with Korbai.

Where do I even start with AI?

Start with the workflow that eats the most time. Ignore the flashy tools for now. The first step is a free discovery call, to work out whether AI would genuinely pay for you at all. If it makes sense, the Audit is generally next: a fixed-fee, fixed-window look at how your business runs, so you leave knowing where AI would pay first, with the number proven on it. If you already know exactly what you want built, in some cases we'll just scope that one build. No big overhaul, one thing at a time.

How we work

We feel like we're falling behind on AI. Are we?

Probably not as far as it feels. Most businesses are in the same spot: a few subscriptions, some experimenting, nothing measured. The gap that matters isn't who started first, it's who turns AI into a real return. That's a decision you can make today.

We've already got scattered AI tools and experiments. Is that a problem?

No, it's raw material. Those experiments show what your team already reaches for and where the interest is. The Audit sorts what's worth scaling, what stays, and what to stop paying for. You don't need to undo anything first.

Do we need an AI strategy before we automate anything?

No. Prove one workflow first, then plan the rest around a real, measured win. A roadmap built on that is something you can fund and staff. A roadmap built on a whiteboard is a slide deck. A real roadmap starts with the Audit.

AI strategy & roadmaps

What does working with Korbai cost?

Every engagement is priced before you commit. The Audit is a fixed fee for a fixed window, and builds, training and AI visibility work are quoted up front, so you know the cost before you decide. No lock-in.

How do I know if AI is even worth it for a business like mine?

You find out on a free discovery call. We'll look at how your business runs and tell you honestly whether AI would pay right now. If it wouldn't, we'll say so, and it hasn't cost you anything to find out.

Are we too small for you?

Unlikely. If a job eats hours every week, there's usually something worth doing, and smaller businesses often see it fastest because the owner feels every hour saved. If it genuinely isn't worth it at your size, that's a five-minute conversation, not a sales process.

Will I be locked in, or can my team own it in-house?

You own it. We build so you can carry it forward, document it in plain English, and train your people to run it. If you keep us on, it's because you want to, not because you're stuck.

Is my business and client data safe, and does it get used to train the AI?

We set things up so your data stays yours and isn't fed into public models for training. We use business-grade tools and proper access, we never handle your passwords, and your client relationships and IP stay with you. Safe use is built in from the start.

Doesn't AI just make things up? How do you stop wrong answers?

It can, and pretending otherwise would be dishonest. We design workflows so a human stays in the loop on anything that matters, ground the AI in your own information instead of the open internet, and build checks in. The goal is reliable help that you can always check.

Do I stay in control, or does AI start making decisions?

You stay in control. This isn't all-or-nothing. We automate the repetitive, low-judgment parts and leave the decisions with your people. AI drafts, sorts and prepares; a human still approves. You decide how much rope to give it, and we build to that.

Will AI replace my staff?

No. We use AI to take the repetitive, time-eating jobs off your team, so the hours go back into the work only people can do. The businesses that win with AI redeploy their people, they don't shrink them. If the goal is headcount cuts, we're probably not the right fit.

Will my team actually use it, or will it end up gathering dust?

They will if someone brings them along. Gathering dust is the real risk, and it's why we treat AI as mostly a people job. We train your people in their own work and build around how they operate, so it's still in use months later.

AI training & enablement

How to get a team using AI consistently

Isn't the hard part the people, not the technology?

Yes, and it's the part we're built for. Roughly 30% of the work is the tools and 70% is the people side (BCG, 2024): habits, confidence and change. That's what twenty years in commercial technology has taught us to handle.

How does a Korbai engagement run?

It starts with a discovery call to see if we're a fit. If it makes sense, the Audit finds where AI pays first and proves it. If you already know exactly what you want built, in some cases we'll just scope that one build. Build puts that one workflow live, with your people. Train gets your team running it on their own work. Partner is ongoing support as things change. One proven step at a time.

How we work

Do we need to be technical?

No technical background needed. We meet your team where they are and build the confidence as we go, in their own day-to-day work. The aim is simply people who use AI well in their own work. Nobody needs to know what's under the hood.

Do we have to rip out the tools we already use?

We'll always look to work within the systems you already run and avoid rip-and-replace where we can. But if a tool is causing real friction or holding you back, we'll tell you and suggest a better fit. The goal is to get more from what you're already paying for.

AI automation

How fast will we see a return?

Measured on one workflow, inside your first 90 days. That’s deliberate: we pick one job that’s costing you time, put a number on what it costs today, and measure what changes. Often you’ll see movement in weeks. And if the return isn’t showing up, we’ll say so plainly rather than stretch the engagement.

Will you tell me honestly if AI isn't the answer?

Yes, and we often do. Part of every Audit is the "not worth it" list: the things we'd skip for now, and why. Knowing what to ignore is half the value, and it's why it ends in a decision.

Do buyers really ask ChatGPT for recommendations?

Yes, more every month. Unlike Google searches, you can't see the ones you're missing. When someone asks an AI assistant who to use, the answer depends on what the AI can read, trust and verify about you. AI visibility is making sure yours is the business it recommends.

Getting found by AI

What is "AI visibility", and what do SEO, GEO and AEO mean?

AI visibility is being found, cited and trusted when people, and increasingly their AI assistants, go looking for what you do. It's the outcome. Three mechanics sit behind it: SEO (search engine optimisation) gets you the click, AEO (answer engine optimisation) gets you the answer, and GEO (generative engine optimisation) gets you cited in the AI's answer. We handle the technical side and keep you in the loop on what we're doing and why.

Is this just SEO rebranded?

No, but good SEO is still the baseline, and anyone selling you one without the other is selling half the job. AI visibility layers new signals on top: being cited, trusted and consistent wherever an AI assembles its answer. We come at it from twenty years in search, so we build on the foundations instead of ignoring them.

Can you guarantee AI will recommend us?

No. Nobody controls what an AI says about a business, and anyone selling guaranteed AI rankings is selling smoke. What we can promise is transparency: you'll see what we changed and what moved after it.

Isn't this just consulting slideware?

No. We lead with a working thing you can see running. The Audit proves a number, the Build ships a workflow that runs in your business, and you leave with decisions you can act on.

How do you measure the return and put a number on it?

We start from the hours a job takes today, then measure what it takes once AI is doing the heavy lifting. The difference is the return, and we count it. We only claim numbers we can stand behind.

Does this actually work, and where's the proof?

It does when it's aimed at the right workflow, which is the whole point of the Audit. The proof is in what clients say. Kirsty Giordani, Global Managing Director of IAA: "A report that used to take a week now takes hours."

Client stories

What is Korbai?

Korbai is a Sydney-based AI consultancy that helps Sydney construction, property and professional services firms with 5 to 100 staff get more profit from the same team, starting with the one job where AI pays. We put AI to work in your everyday workflows, get you found by AI, train your team, and set the roadmap. Founder-led, and happy to work in the room with you.

Who is Paul Korber?

Paul Korber is the founder of Korbai. He's spent twenty years in commercial technology, including eight at Microsoft, the last two inside Microsoft AI as technical sales lead for Asia Pacific, plus years building service-based businesses of his own. Most AI consultants have the corporate credentials or the founder scars. Paul has both.

About Paul Korber

Where are you based, and do you work in person?

Sydney. We work with clients across Australia, and being local is an edge we use: where it helps, we'll be in the room with your team.

Who do you work with?

Sydney construction, property and professional services firms with 5 to 100 staff: builders, trades and property managers, and accounting, legal and advisory firms. The owners, principals and COOs we work with are capable and time-poor, and know there's something in AI but haven't seen it pay yet.

AI in construction: where it pays, and where it doesn't

AI for accountants: where it saves time, and where it doesn't

What makes you different from other AI consultants?

Most are one of two things: a pure technologist who can't sell the change, or a corporate escapee who's never run anything. We're built on having done both, which shapes what we build, how it lands with your team, and what we'll tell you isn't worth doing. And we're honest about where AI doesn't belong.

What does the name Korbai mean?

Korb is German for basket. The name nods to the basket-maker’s trade: taking loose strands and shaping them into something that holds weight. That’s the job, and it hasn’t changed.