The quote is not hard. It is the fourth copy of a quote that went out last month, with two lines changed and the same three optional extras someone retypes every time. It goes out as an attachment, comes back as an email saying "yes, and let's add the second one", and somewhere between that reply and the order being raised, a person reads one document and types into another. The order is now a claim about what was agreed. The evidence is a file.
The quote you build is the order you confirm, there is no second document to create and nothing to re-key. In Odoo a quotation is a sales order read at an earlier point, moving from Quotation to Quotation Sent to Sales Order; confirming it changes a state, not the record.
Templates that are configuration, not a starting point to edit. A quotation template carries its own lines, its optional lines, a validity in days, the journal it will invoice through, its mail template and the PDF pages appended to it, so calling one up produces a complete quotation rather than an empty shell.
Optional extras that do not inflate the number. Optional products sit apart from the priced order with their own price and discount, and only become order lines and only start counting toward the total, once the customer accepts them. Bundles have their own product type, a base price plus a per-item extra, so a bundle is defined once rather than re-priced on every quote.
The signature and the deposit in the same step. Requiring a signature and requiring a deposit are both settings on the order; the signature is stored with the name of the signer and the moment it was signed, and the deposit is a percentage of the order total rather than an amount someone keys in.
All of it is in the base system, nothing to add, no upgrade tier to reach, no extra fee per user, and it runs on infrastructure you own. That includes the branded proposal builder: the capability a buyer assumes is a paid upsell ships in the base product, and that it is already included is the point.
A sales quotation is a priced, time-bounded offer to a customer: the items and quantities, the price and any discount, the terms, and a date after which the offer no longer stands. It becomes a sales order at the moment the customer accepts it and the seller commits, after which the same document is a promise to deliver rather than an offer to. In some systems those are two separate records, and acceptance means creating the second from the first. In others, including Odoo, they are one record read at different points in a sequence of states, and acceptance changes the state rather than producing a second document. The difference is not cosmetic. Where they are two records, the question "what exactly did the customer agree to" is answered by comparing them; where it is one, that question does not arise, and neither does the class of error where the order says something the quote did not.
Because the repeatable part of quoting is rarely the wording. It is the set of lines that always go together, the optional extras that are always offered, how long the price stands, which journal the eventual invoice belongs to, and what gets attached to the front of the PDF. A template that holds only text leaves all of that to be reassembled by whoever is quoting, which is why the fourth copy of a quote drifts from the first. A template that holds the structure, the lines, the optional lines, the validity period, the invoicing journal, the covering documents, turns the repeatable part into configuration, and the quote into the part that is actually specific to this customer. It is also the difference between a quoting process a new salesperson can join in a week and one where the accurate quotes are the ones built by the person who has been there longest.
Yes, where the quotation itself carries both as requirements. In Odoo Community, requiring a signature and requiring payment are flags on the order record: the signature is stored on the order along with the name of the signer and the timestamp, and the deposit is expressed as a percentage of the order total rather than as a figure someone types, with the system rejecting any percentage outside zero to one hundred. Both default from a company-level setting and both can be overridden per quotation or set on a quotation template, so "everything over this size needs a signature and half up front" becomes a property of the template rather than a rule people remember. The practical benefit is not that the signature is electronic, most tools do that but that the signed artefact and the priced document are the same record, so there is no step where a signature in one system has to be matched back to an order in another.
They solve opposite problems and are often confused because both change what the customer eventually pays. A discount adjusts the price of something already being sold; it belongs on the order and it belongs in the total. An optional product is something being offered but not yet sold, the upsell, the extended warranty, the second unit, and putting it on the order as a line means the quoted total is a number the customer did not ask for and may reject, which is the fastest way to make a quote look expensive. The clean implementation keeps optional products on a separate model with their own price and their own discount, contributing nothing to the total until the customer accepts one, at which point it becomes an ordinary order line. Worth checking in any system that markets upsell on quotations: ask whether the optional item is a line with a flag on it, or a genuinely separate record, because the first will eventually appear in a total that someone has to explain.