Turn missed enquiries into answered questions, bookings and next steps

Customers do not always contact a business during office hours or when staff are free. AI Architecture Lab designs controlled AI systems that can answer approved questions, guide customers, collect enquiries and support the next step when the team cannot respond immediately.

Start your free 30-day experience

No fee. No obligation. Nothing goes near your website or your customers without your written approval.

A controlled response path

Enquiry
Controlled AI capabilityApproved information · defined boundaries
  • Approved answer
  • Collect enquiry
  • Next-step support
  • Human escalation

Problems we help organisations solve

These problems look different from one industry to another, but the commercial pressure is often the same: a customer needs an answer or a next step, and the organisation cannot provide it at that moment.

Customers waiting for answers

People make contact while staff are closed, asleep, serving someone else, driving, in a meeting or busy with operational work. The customer may not wait.

Missed enquiries and bookings

A person who cannot understand the service, request a booking or find out what happens next may contact another business.

The same questions over and over

Frontline people keep answering the same approved product, service, policy and process questions instead of handling work that needs their judgement.

Slow follow-up

An enquiry arrives, but acknowledgement and follow-up depend on someone finding time between other responsibilities.

Staff cannot find the right answer

The right information exists, but people cannot find it quickly or different teams give different answers.

Some questions need a person

Some questions are sensitive, urgent, unusual or outside scope. The system needs clear escalation and refusal rules rather than an answer at any cost.

The customer may not wait

Your customer does not care that the phone rang while you were on another job. They only know that nobody answered.

Consider a customer with a pipe problem at about 5 a.m. They contact several plumbers. One offers a business-hours-only response. Another does not respond. A third immediately explains whether the service appears relevant, collects an appropriate enquiry and supports the next available step. Which business is more likely to keep the enquiry?

5:00 AM Customer enquiry

Business A

No answer

The customer receives no response at that moment.

Business B

Business-hours only

The customer is asked to wait until the business is available.

Business C · controlled path

Approved information

  1. The enquiry arrivesAcknowledge it immediately and explain whether the business handles that type of job, using approved information.
  2. The customer needs a next stepCollect approved basic information and, where an approved integration permits, offer or request an available booking.
  3. The request needs judgementExplain what happens next, escalate an urgent or unsuitable request according to defined rules, and refuse questions outside the approved boundary.

We do not start with the technology

First we understand what is going wrong. Then we decide whether AI is useful, what information it may use, what it may do and when a person takes over.

That is a design problem before it is a technology problem. It means understanding your business, your customers and your constraints first, then deciding whether AI is the right answer at all. Sometimes it is not, and we will say so.

A controlled way to build

Our work runs on ACOS — the AI Consultant Operating System. It is our own methodology, developed in-house, and it structures every engagement we take on. Seven phases:

  1. Discover

    understand the business, the customers, the objectives and the constraints

  2. Assess

    establish what is verified, what is assumed, and what has to be confirmed

  3. Architect

    design the knowledge foundation and the boundaries the system works within

  4. Build

    construct and configure the solution

  5. Deploy

    put it in front of the right people, in a controlled way

  6. Govern

    set the rules, the accountability and the testing that keep it trustworthy

  7. Evolve

    improve it as the organisation changes, because capability that is not maintained decays

ACOS moves from the business problem to a tested and governed AI capability. You do not need to understand the methodology before talking to us.

See how ACOS works

Start small. Go further only if it earns it.

Five ways to work with us. The right package depends on the capability you need and the complexity of the AI environment being supported. All prices are shown in Australian dollars and exclude GST. GST will be applied where applicable.

Free 30-Day AI Assistant Experience

Demonstration only. The Free 30-Day AI Assistant Experience uses appropriate public or supplied business information. Any assumptions are identified, and authoritative knowledge is verified and refined with you.

AI Assistant Essentials

Discover, Assess, Architect and Build. A$5,900 + GST setup and A$695 + GST/month.

ACOS Managed

The full ACOS lifecycle for a managed AI environment. A$12,500 + GST setup and A$1,750 + GST/month.

ACOS Integrated

The full ACOS lifecycle for a more technically complex AI environment. From A$24,500 + GST setup and from A$3,500 + GST/month.

Enterprise AI Architecture

The full ACOS lifecycle plus enterprise requirements. Custom pricing.

Managed and Integrated cover the same ACOS phases. The difference is not how much of the methodology you receive — it is the complexity of the AI environment being supported.

Compare the packages

Specialist experience

Retail & Customer Operations

Customer enquiries, product and policy knowledge, frontline information access, service consistency and operational support. Inventory visibility remains an important secondary capability.

Aged Care & Healthcare

Administrative burden, workforce pressure, documentation, and knowledge that is hard to reach when it is needed. These are specialist experience areas, not limits on the organisations we can work with.

What this could look like in your business

The right capability depends on the problem and approved scope. It may support one or more of these areas; it does not automatically perform all of them.

Customer enquiries

Answer approved questions and explain services.

Booking or appointment support

Guide a request or next step where an approved workflow and integration permit.

Lead or enquiry follow-up

Acknowledge and follow up using approved rules.

Staff knowledge

Retrieve approved operational information while work is happening.

Policies and procedures

Help staff locate current approved guidance.

Human escalation

Recognise when a question needs a person and route it appropriately.

See what one of these actually looks like

How we work

  1. Tell us the problem

    Bring us the problem. We will work out the AI part.

  2. We understand the workflow

    We look at where people wait, where work gets stuck and what information is available.

  3. We decide whether AI makes sense

    If a simpler response is more appropriate, we would rather say that before anything is built.

  4. We build and test within boundaries

    Approved information, refusal rules, human escalation and deliberate testing are part of the design.

  5. You decide what happens next

    Continue, change direction or stop. The evidence should earn the next step.

That is the short version. Underneath it sits ACOS, our seven-phase methodology, which is how the detail is structured, governed and kept honest.

See the full ACOS methodology

How we keep it trustworthy

Controlled information

An initial demonstration may use appropriate public or supplied business information. Assumptions are identified, and production knowledge is verified and approved before it is treated as authoritative.

Defined boundaries

That narrows what it covers, deliberately. When a question falls outside what it has been given, it says so rather than guessing.

Human escalation

Questions requiring judgement or falling outside scope go to an accountable person rather than receiving an answer at any cost.

Testing before deployment

Difficult questions, missing information and boundary attempts are tested before wider use.

Read our approach to responsible AI

You work directly with Dennis

AI Architecture Lab was founded by Dennis Wills, who spent more than twenty years in senior retail management in Australia. That experience taught him that customers experience a business through the answer they receive, the product or service they can access, the person who helps them and how quickly their problem is resolved. He then completed the Professional Certificate in AI and Machine Learning, a Purdue University program delivered by Simplilearn, and now builds AI systems and authored ACOS.

Your first conversation is with the person responsible for understanding and designing the work. Practical retail and frontline experience helps Dennis ask where customers wait, where staff lose time and which decisions must remain human.

More about Dennis

See it working on your own information

30 days. No fee, no obligation, no payment details. We build an assistant from appropriate public or supplied business information, verify and refine it with you, you use it in a private environment with your own team, and you finish with a written record of what was tested and what was not.

Start your free 30-day experience

Would rather talk first? That is often the better place to start.

Request a consultation