A self-hosted AI quoting agent is a workflow, built in a tool such as n8n on your own server, that reads a quote request, looks up prices in your own systems, drafts a priced quote using your margin rules and sends it to a person for approval. The AI model runs in an Australian cloud region under your account or on hardware you own, so pricing data stays onshore.
If you run sales, operations or estimating at a multi-branch distributor or service organisation, you know the queue. Requests for quote land in a shared inbox. Someone opens the PDF, finds each item in the price book, applies the customer's terms and keys it all in. Meanwhile the customer has emailed two other suppliers.
This guide covers how the agent works, the two ways to host the AI behind it, how to choose between them, and when self-hosting is the wrong call.
Key takeaways
- The quoting agent is mostly lookups and rules your team already follows. Only one or two steps use AI.
- n8n Community edition can be self-hosted for your own internal business use, and approval steps are built in.
- You can run the AI in an Australian cloud region under your own account, or run an open-weight model on your own server.
- Clean item codes and a written-down price book matter more than which model you pick.
- A person approves every quote before it reaches a customer.
Why does quote speed matter so much?
Research published in Harvard Business Review in 2011 ("The Short Life of Online Sales Leads", Oldroyd, McElheran and Elkington) found that firms contacting a lead within an hour were nearly seven times as likely to qualify it as firms that waited longer.
A quote request is a lead where the customer has already told you what they want and is waiting for a number. Most of the delay sits in the reading and the looking up, which a machine can draft. The judgement call, whether this customer gets the sharper price, stays with your people.
| What slows a quote down | Who it waits on | Can an agent draft it? |
|---|---|---|
| Reading the RFQ, PDF or spec | Whoever opens the inbox | Yes |
| Finding each item in the price book | The person who knows the codes | Yes, if item data is clean |
| Applying customer terms and margin rules | Estimator or sales lead | Yes, as fixed rules |
| Deciding on a special price | Sales manager | No, a person decides |
| Checking and sending | Estimator | A person approves; the send is automatic |
Why won't IT let you use a public AI chatbot for quoting?
This is a common starting point. Someone pastes a customer request and a page of the price list into a public chatbot and gets a decent draft. Then IT or the finance director asks, fairly, where that data went.
Your price book, margins and customer list are the business. Under Australian Privacy Principle 8, an organisation that discloses personal information overseas must take reasonable steps to ensure the recipient does not breach the APPs, and under section 16C of the Privacy Act it generally stays accountable for what that recipient does, as set out in the OAIC's APP 8 guidelines.
Customer contacts on a quote are personal information. Your pricing is commercially sensitive whether or not the Privacy Act applies to it. Self-hosting moves the AI to somewhere your IT team already accepts. For the wider picture, see our guide to AI data sovereignty and onshore deployment.
How does a self-hosted AI quoting agent turn an RFQ into a draft quote?
The agent is a chain of ordinary steps. Here is the flow for a typical wholesale or trade supply quote:
- Request arrivesRFQ email, web form or PDF spec lands in the quotes inbox
- Request readAI pulls out items, quantities, site and delivery date
- Prices looked upItems matched to your price book, ERP, Xero or MYOB
- Margin rules appliedCustomer terms, freight and minimum margin calculated
- Person approvesEstimator checks, edits or rejects the draft
- Sent and loggedQuote emailed and recorded against the customer in your CRM
The AI reads the request and matches messy descriptions ("20 x 90mm PVC elbows, DWV") to your item codes. The prices and margins come from your system and your rules.
A person approves the draft before anything reaches a customer. In n8n this is built in: the human-in-the-loop documentation shows how to require approval before an AI agent takes an action, and the Slack node can send and wait for an approve or decline.
For quotes built from photos and site walks rather than written requests, the flow changes. Our post on AI field quoting for service operators covers that case.
Where does no-code n8n fit in?
No code means you build the workflow by dragging steps onto a canvas. Your operations team can read it, and a change to a margin rule is a change to one box.
n8n suits this job because you can host it yourself. Its Community edition is free to self-host and includes most of the feature set. It leaves out enterprise features such as single sign-on, environments, shared credentials and Git version control.
Two licence points matter before you commit:
- n8n's Sustainable Use License allows use "only for your own internal business purposes or for non-commercial or personal use". Running your own quoting fits that. Reselling n8n as a hosted service does not.
- For self-hosted instances, n8n's privacy documentation states that workflow data stays on your infrastructure. Telemetry is on by default, and your administrator can turn it off with environment variables.
n8n also ships model connectors for Azure OpenAI, AWS Bedrock, Google Vertex and Ollama. Switching the AI behind the agent later is a settings change. If you are still choosing a workflow tool, our Zapier vs Make vs n8n comparison covers fit.
Option 1: AI in an Australian cloud region, under your own account
The first way to host the model is a managed service from Microsoft, Amazon or Google, set to an Australian region in your own cloud account. Nobody on your team looks after GPUs, and you pay for what you use.
The catch is that each provider offers a different set of models in Australia, and the deployment type decides where data is processed. The notes below reflect official documentation as at October 2026. Check each page again before you commit, because availability changes often.
Azure OpenAI (Australia East)
For Standard (regional) deployments, Microsoft states that it processes prompts and responses in the region associated with the deployment. Its regional availability page lists a subset of models for Standard deployment in Australia East. Newer models often appear first under Global or Data Zone deployment types, and Microsoft's Asia Pacific data zone covers processing anywhere in that zone, which is wider than Australia.
Amazon Bedrock (Sydney)
The Bedrock regional availability page lists Amazon Nova models in-region in Sydney, with some partner models reached through cross-region inference. AWS documents an Australia-only inference profile alongside wider Asia Pacific and global profiles, so the profile you pick matters.
Google Vertex AI (Sydney)
Google's locations page lists Gemini and some partner models at the Sydney endpoint, and warns that an endpoint location does not on its own guarantee in-region processing. Check the data residency page for the specific model.
What this means for a quoting agent
Across all three, in-region options lag the newest global models. For a quoting agent it rarely matters, because reading an RFQ and matching item codes does not need the newest model. For a deeper comparison, see AWS vs Azure vs Google Cloud for AI in Australia.
Option 2: Your own model on your own server
The second way is to run an open-weight model yourself, using a tool such as Ollama or vLLM, on a server in your building or your own Australian virtual machine. Nothing leaves your network.
Ollama's FAQ states that when it runs locally, Ollama does not see your prompts or data. Ollama also offers cloud-hosted models, and its FAQ explains how to switch cloud features off for local-only use. Whoever sets it up should do that.
Running your own model means accepting a few limits:
- Models that fit on a single server are smaller than the cloud flagships. They handle structured reading and matching well and are weaker on long, messy specs.
- You own the upkeep: hardware, patching, model updates and monitoring. If the server stops, quoting falls back to manual.
- The cost is mostly upfront hardware rather than a monthly usage bill.
Our OpenAI vs Claude vs Ollama comparison goes further into model quality.
Which hosting option suits your organisation?
Both options keep your price book out of overseas consumer AI tools. The difference is who looks after it and how much the model can do.
| Australian cloud region (Azure, AWS, Google) | Your own model (Ollama or vLLM) | |
|---|---|---|
| Where the data sits | Your cloud account, Australian region, if the deployment type is regional | Your server, your building or your own Australian VM |
| Cost shape | Pay per use, no hardware | Hardware upfront, low running cost |
| Upkeep | Provider runs the model | Your IT team or a partner runs it |
| Model quality | Larger models, though the newest may be global only | Smaller open-weight models |
| Fits your IT team if | You already run Azure, AWS or Google | You run on-prem servers and want no external calls |
| Main risk | Choosing a global or data zone deployment by mistake | Nobody owning the server after go-live |
What matters most to your IT and finance teams?
- We already run Azure, AWS or Google Cloud and want no servers to manageAustralian cloud region in your own account
- Nothing may leave our network, and we have someone to run a serverYour own model on your own server
- We quote only a handful of jobs a weekDo not build an agent yet; tidy the price book and templates first
- Nobody internally can own a workflow after go-liveManaged support, or wait until someone can
When should you not self-host a quoting agent?
- Low volume. If you quote a few jobs a week, a good template and a clean price list will save more time than any agent.
- No owner. A self-hosted workflow needs someone for updates, backups and broken connections. Without that person, expect it to break and stay broken.
- Prices live in someone's head. The agent can only look up prices that exist in a system. If your best estimator prices from memory, write that down first.
- Bespoke work. Design-and-fabricate jobs need an estimator's judgement on every line. AI can still read the drawing pack, but the draft saves less time.
Across our consultants' careers, including operational data work in BHP and Rio Tinto environments, the same lesson came up repeatedly: master data quality decides the result far more than the tool on top. Clean item codes do more for you than a bigger model.
How long does it take to get the first draft quotes?
A sensible build starts with one quote type, measures today's turnaround, and runs the agent alongside your team before any customer sees its work.
- Weeks 1 to 3Baseline and designMeasure today's turnaround, pick one quote type, map price book and margin rules, choose cloud region or own server
- Weeks 4 to 9Build and run alongside your teamWorkflow built in n8n; the agent drafts quotes in parallel while your team quotes as normal; drafts compared weekly
- Weeks 10 to 14Person approves every quoteOnce drafts meet the accuracy level you set, estimators approve each one before it is sent; handover to your IT
While it runs alongside your team, the agent drafts quotes nobody sends. You compare its drafts to the real ones each week and set the accuracy it must reach before it moves to approval mode. Each phase ends with a decision to proceed, pause or stop, and you keep everything built so far.
What could a quoting agent free up?
Consider a hypothetical trade wholesaler with several branches. The figures are assumptions for illustration, not results, so swap in your own.
- Quotes per week (assumption)40
- Minutes spent reading and keying each quote (assumption)30
- Hours per week on reading and keying20
- Hours still spent if approving a draft takes 10 minutes (assumption)6.7
- Estimator hours freed per week in this example13.3
Measure your own minutes per quote before anything is built. That baseline is how you will know whether the agent earned its place.
| Manual quoting | With a self-hosted quoting agent | |
|---|---|---|
| First look at the request | When someone gets to the inbox | A draft is waiting for review |
| Item lookup | Searched by hand in the price book | Matched to item codes, flagged where unsure |
| Margin rules | Applied from memory, varies by person | Applied the same way on every draft |
| Final decision | Estimator | Estimator, unchanged |
| Where pricing data goes | Stays in your systems | Stays in your systems and your Australian region or server |
How Solve8 can help
Our process automation and private AI infrastructure teams scope and build quoting agents that run on your infrastructure in Australia, with a person approving before anything reaches a customer. Our own product RootCauseAI ships as a self-hosted install for the same data reasons. Talk to us about whether a quoting agent suits your quote volume.
Sources: Microsoft Learn, region availability for Foundry Models (accessed October 2026); AWS, Amazon Bedrock regional availability and cross-Region inference documentation (accessed October 2026); Google Cloud, Vertex AI locations and data residency pages (accessed October 2026); n8n documentation and Sustainable Use License (accessed October 2026); Ollama FAQ (accessed October 2026); OAIC, APP Guidelines Chapter 8; Oldroyd, McElheran and Elkington, "The Short Life of Online Sales Leads", Harvard Business Review, 2011.
