Everything that usually decides it, before the detail below.
In detail
Quoting is where margin quietly leaks: the same job priced three ways by three people, discounts given because it was late, and quotes that arrive two days after the customer asked. We encode your price list, volume tiers, discount limits and approval thresholds once, and the agent applies them the same way every time — producing a finished, branded document from an enquiry in minutes.
Parameters and terms
Time to live
4–6 weeks to production
Price
From €1,600
Autonomy
Autonomous within your limits, approval above the threshold
Deployment
Our managed cloud or yours
What the agent does
Quotes generated from your own pricing logic and price list
Volume tiers, discounts, surcharges and margin floors enforced
Branded PDF or document output, ready to send
Pushed into CRM, email or your ERP with the deal record updated
Approval step for anything outside your normal limits
How we build it
1Pricing discovery
2Rules and template
3Parallel running
4Live, within limits
What you get
Quotes out the same day, which is frequently the whole competitive advantage
The same job priced the same way regardless of who handled it
Margin floors and discount limits enforced rather than remembered
Every quote logged against the deal, so you can see what was offered and what won
Who it is for
Manufacturing and fabrication
Wholesale and distribution
Service companies with a rate card
Teams where one person quotes everything
Optional extensions
Win/loss analysis on quoted price against outcome
Automatic follow-up on quotes that have gone quiet
Order this agent
Tell us which process is costing you the most hours. The first consultation and the process audit cost nothing, and you keep the findings either way.
Companies where quoting is frequent, rule-driven, and currently done by whoever has the time.
Manufacturing and fabrication
Material, labour, tooling and volume tiers combined into one number. The rules exist; they are just held in one or two people's heads.
Wholesale and distribution
Customer-specific price lists, volume breaks and framework agreements — high volume and entirely rule-driven, which is the ideal case.
Service companies with a rate card
Agencies, installers, logistics and maintenance, where a quote is an assembly of known rates and the delay is purely administrative.
Teams where one person quotes everything
The classic bottleneck. When that person is on holiday, quoting stops — and the agent removes that dependency entirely.
Before we start
What we need from you
Your pricing logic, written down properly for the first time.
Your current price list and cost basis, in whatever form it lives — ERP, spreadsheet or a document.
The rules behind it: volume tiers, customer categories, surcharges, rounding and the margin floor nobody is allowed to cross.
Discount and approval thresholds — who may approve what, and where a quote must stop for a human.
Twenty to thirty recent quotes, including the awkward ones. The exceptions teach the agent far more than the standard cases do.
Your quote template and terms, and access to the CRM or ERP where the deal record lives.
Setup
How much it decides, and where it runs
Autonomous inside your limits, approval above them, running in our managed cloud or your own tenancy.
How much it decides
Draft-first
The agent prepares everything and a human approves before anything leaves the building. Slower, and the right choice while you are still establishing how often it is correct.
Autonomous
The agent acts on its own within the rules you set, and escalates whatever falls outside them. Almost every client arrives here gradually, one process category at a time, rather than on day one.
Where it runs
Our managed cloud
We host, monitor and maintain the agent, and you pay nothing for infrastructure. The fastest way to production and the usual choice for a first agent.
Your own infrastructure
The agent runs inside your cloud tenancy or on your servers, so your data never leaves an environment you control. Common in banking, healthcare and the public sector.
Hybrid
Processing happens in our environment while the sensitive data stays in yours, connected over a restricted interface. A practical middle ground when policy and speed pull in different directions.
This is the agent where the approval threshold matters most, and we spend real time on it during the design phase. Standard configurations inside your normal margin go out on their own; anything involving a new customer, an unusual volume, a non-standard specification or a discount beyond the limit routes to a named person with the calculation shown. Most clients set that threshold cautiously at first and relax it once they have watched a few hundred quotes come out correct.
Process
How the rollout runs
From pricing that lives in someone's head to a rule set that runs itself.
Weeks 1–2
Pricing discovery
The longest phase, and the one that pays for itself. We reconstruct the rules you actually apply — including the exceptions nobody wrote down — and hand them back as a document. Clients regularly find inconsistencies here before any software is built.
Week 3
Rules and template
The logic is encoded and tested against your historical quotes: the same inputs must produce the same outputs your team produced, or we find out why not. The document template and terms are built in parallel.
Week 4
Parallel running
Real enquiries are quoted by both the agent and your team. Differences are examined one by one — sometimes the agent is wrong, and quite often the rule was.
Week 5 onward
Live, within limits
The agent quotes inside the thresholds you set and escalates above them. We review the escalation rate monthly and tighten or relax the limits against what actually happened.
Good to know
Frequently asked questions
Our pricing has too many exceptions to automate.
That is what almost every client says in the first meeting, and the discovery phase is designed around it. Exceptions are rules with conditions attached — the question is only whether there are twelve of them or two hundred. We find out before you commit, and we will tell you if the answer is that automation is not worth it.
Can it quote something it has not seen before?
No, deliberately. A genuinely novel configuration escalates to a human with everything it could assemble already prepared. Quoting is one place where a confident guess is far worse than a short delay.
Will it undercut our margins?
The margin floor is a hard constraint, not a guideline — the agent cannot produce a quote below it, and a request that would require one becomes an escalation. In practice quoting agents tend to raise average margin slightly, because they never give a discount out of fatigue.
Does it work with our ERP?
If your ERP has an API, yes — we read the price and stock data from it and write the quote back against the deal. Where there is no API, we work from a scheduled export, which is less elegant but perfectly workable.
Who is responsible if a quote is wrong?
You are, commercially — which is exactly why the thresholds and parallel running exist. We do not ask you to trust it; we ask you to watch it match your own team for several weeks first, and keep the limits wherever they make you comfortable.
Keep going
Related AI agents
Companies often pair this agent with one of these.
E-commerce Catalogue Agent
Supplier feeds to live listings, no retyping
From €1,600
For shops that add products faster than a person can describe them. The agent ingests supplier feeds in whatever shape they arrive, writes descriptions that read like a human wrote them, translates them into every market you sell in, and pushes complete, SEO-ready listings into your store.
3–6 weeks to production Autonomous for new products, approval for price and stock rules
Your shared inbox stops being a queue. The agent reads each incoming message, classifies it, pulls the customer's history from your CRM and drafts a reply that matches how your company writes. Routine confirmations go out on their own; anything sensitive waits for one click from you.
3–5 weeks to production Draft-first by default, autonomous per category