Sales
AI for proposals and quotes
Build a small library of your best past proposals, standard service descriptions and terms. Give the AI the client's brief and ask for a first draft that follows your structure. Keep prices out of the AI's hands and take them from your price list or quoting tool. A senior person reviews scope, assumptions and terms before anything reaches the client.
Effort:
Medium
Time saved:
Often roughly 30–50% less first-draft time; measure it in a pilot
The problem today
Each proposal starts from an old file that is copied, renamed and patched. It takes hours, a previous client's name slips through, and the scope drifts away from what you actually deliver.
How it works, in plain words
It is like handing a capable writer your three best past proposals and a short brief, then asking for a first draft in the same shape.
What goes in: the client's brief, your proposal structure, standard service descriptions, terms and a few anonymised past proposals.
What the AI does: it fills each section of your structure in turn, reusing your own wording and adapting it to the client's situation and goals.
What a person checks: a senior person reviews scope, exclusions, assumptions and terms, and adds prices and quantities from your price list or quoting tool.
What comes out: a branded proposal or quote ready to send, built from your library instead of an old file with another client's name in it.
It does not set prices or judge whether a job is profitable. It can also promise more than you deliver if the scope is not checked.
How it could work for businesses like yours
Illustrative examples, not client case studies.
A 6-person events and social media agency in Heraklion
Each proposal for a festival or hotel launch starts from the last one, and old client names slip through. The agency would build a folder with its standard service blocks and ask AI to draft each section from a one-page brief. They would track drafting time per proposal and aim to roughly halve it, measured over the first ten proposals.
An 11-person office-supplies e-shop in Kalamata
Companies email lists of items and ask for a formal quote with delivery terms. Staff would paste the request, let AI draft the cover note and terms from the standard text, and take every price from the shop's system. A realistic first target could be sending most quotes on the same working day.
A 40-person metal fabrication workshop in Sindos
Technical quotes for railings and steel structures take the engineer hours of writing after the costing is done. The workshop would let AI draft the scope, exclusions and assumptions from the engineer's notes, while the costing stays in its own spreadsheet. They would measure the engineer's writing time and how often clients ask to clarify scope.
How to do it, step by step
Choose 3–5 strong past proposals and remove client names and prices. Keep them in one folder with your standard service descriptions, company profile and terms.
Fix the structure: client situation, goals, proposed approach, scope and exclusions, timeline, assumptions, next steps. The AI fills the structure; it does not invent one.
For each new request, write a short brief: who the client is, what they asked for, budget signals, deadlines and what is out of scope. Use the client's own wording where you can.
Ask for the draft section by section, not the whole document at once. Shorter sections are easier to check and stay closer to your examples.
Add prices and quantities yourself from your price list or quoting tool. Never let the AI calculate totals, discounts or VAT.
A second person checks scope, exclusions, dates and terms. Save the final version back to the library so the next draft starts from better material.
Tools to consider
Microsoft 365 Copilot in Word: drafts from your own files in OneDrive or SharePoint, so it can reuse past proposals. Caveat: it only finds what is well organised and shared with you; messy folders give messy drafts.
ChatGPT or Claude with projects: upload your template and examples once and reuse them for every brief. Caveat: use a business plan and share only the client information the draft needs.
Notion AI: useful if your service descriptions and templates already live in Notion. Caveat: weaker for complex formatted documents; export and finish in your usual format.
Canva: turns the final text into a clean, branded layout. Caveat: it handles design only and does not check the content.
Data protection and GDPR
Proposals mix personal data, such as contact names and roles, with confidential commercial information. Use AI tools under a business agreement that does not train on your data, and anonymise past proposals before using them as examples. Do not paste a client's internal documents into a tool they have not agreed to, and check any NDA you signed. Keep margins, pricing logic and supplier costs out of general chatbots. The legal terms in a proposal should come from your approved template, not from AI. If you need new terms, check with your legal advisor.
Mistakes to avoid
Letting the AI produce prices, totals or VAT calculations.
Leaving a previous client's name or details in a reused example.
Promising scope or timelines that the delivery team never agreed to.