As of 13 August 2026, AI can only partly calculate a quote for a customer.
This still needs a person who signs their name to it.
Can you do it?
2 minutesto a draft.
15 minutesto something you’d act on.
Cost, all in£0
Skill neededchat-fluent
Who has to check ityou
What the alternative costsQuoting software with rule engines built in, such as Salesforce CPQ, costs from £50 to £200 per user per month.
If this goes wrong: you issue a quote at the wrong margin, or apply a discount the customer was not entitled to, and the mistake spreads across repeat orders before you catch it.
What to actually do
Hand it to a person
The route this page recommends
A person who owns the outcome does this end to end, worth it when the failure is dear.
Use a tool built for this
Second choiceDo it yourself
The distant thirdA chat interface, chat-fluent skill, and roughly 15 minutes until you can act on the result.
How to actually do it
- 1. Open ChatGPT, Claude or Gemini in a web browser.
- 2. Copy your current price list (product codes, names, unit prices) into a text document and have it ready to paste.
- 3. Write down your discount rules in plain English: for example, '10% off orders over £5,000', '5% loyalty discount for repeat customers', 'no discount on items marked final sale'.
- 4. Gather the customer details: product codes they want, quantities, whether they qualify for any discount, any special terms they have negotiated.
- 5. Paste the prompt above into the chat, and replace the bracketed placeholders [pricing table], [discount rules], and [customer request] with your actual data.
- 6. Copy the generated quote into your quoting system or email template.
- 7. Before sending: check the maths line by line against your price list, and verify that every discount applied matches your rules and the customer's entitlement.
- 8. Send the quote to the customer.
Prompt
You are a pricing assistant. I will give you our pricing table, discount rules, and a customer request. Generate a quote with line items, totals, and applied discounts. Show your working for every calculation. Do not invent discounts or prices. If a product code or customer tier is unclear, ask rather than guess. Pricing table: [paste your current price list] Discount rules: [paste your rules: e.g. 10% off orders over £5,000; 15% for repeat customers] Customer request: [product code, quantity, customer type, any special terms] Generate the quote. Line by line, show: product, unit price, quantity, line total, discount applied, subtotal. Final total with VAT at 20%.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
What it gets wrong
- Applies discounts mechanically without understanding your margin targets, so a quote can lock in a price that undercuts your profitability if your rules are vague or out of date.
- Cannot flag unusual requests: if a customer asks for a discount they have never received before, the model will calculate it if the request is plausible-sounding, leaving you to catch the error.
- Requires you to transcribe your pricing and rules as text; if your system is in a spreadsheet or bespoke software, you must extract and paste manually, which is the slowest step.
- Drifts on regeneration: if you ask it to recalculate or adjust the quote, the discount application or VAT treatment can shift without warning.
- Cannot integrate with your sales or billing systems, so the quote sits in email or a PDF; someone still enters it into your CRM by hand.
What caps this at PARTLY: stakes of error, verification cost and context depth.
How we scored this
Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 1 |
| Verification | 1 |
| Liability | 1 |
| Effort delta | 2 |
| Total | 7 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can AI calculate a quote faster than I can?
- Yes, if you have your pricing data ready. A quote that would take you ten minutes by hand takes the model two minutes, but you still need five minutes to check it before sending. The real time saving is in repetitive quoting: if you generate five quotes a day, you save thirty minutes daily.
- What goes wrong if I use AI for quoting?
- The model applies discounts mechanically without understanding your margin targets, so a quote can lock you into a price that loses money if your rules are out of date or the customer's entitlement is unclear. It also cannot flag unusual requests, so you must check the discount logic against your records before sending.
- Do I need to use a special tool, or can I just ask ChatGPT?
- ChatGPT, Claude or Gemini work fine for one-off quotes once you have your data ready. For frequent quoting, a spreadsheet with AI analysis (Rows) or a purpose-built quoting tool with rule logic will save more time and avoid the manual transcription step. This is NOT professional advice; for complex pricing strategies, a pricing consultant can review your margin assumptions.
- What if the customer disputes the quote later?
- You are responsible for the quote regardless of how it was generated, so you must be able to justify every line and discount against your current rules. Keep a record of the discount rules and customer entitlements you gave the model, and check the output before sending.
Nearby answers
Assessed by claude-haiku-4-5 (claude-haiku-4-5-20251001) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.
The newsletter
AI news, new answers and product picks, straight to your inbox.