A supplier sends a spreadsheet of products. You need them in Odoo.<\/p>\n
What is in the file:<\/strong> product names, maybe a code, maybe dimensions, a price.<\/p>\n What Odoo needs:<\/strong> a name, an internal reference, a category, a unit of measure, a cost, a sales price \u2014 and for anything customer-facing, a description worth reading.<\/p>\n The gap is the problem.<\/strong> A spreadsheet with 400 rows becomes 400 records, each needing fields the file does not contain and a description somebody has to write.<\/p>\n Done by hand, that is weeks.<\/strong><\/p>\n Be specific, because this is what decides whether the workflow is safe.<\/p>\n AI helps with:<\/strong><\/p>\n Descriptions.<\/strong> Turning attributes into readable text, in a consistent tone, at volume.<\/p>\n Category suggestion.<\/strong> Reading a product name and proposing which of your categories it belongs in.<\/p>\n Reference generation.<\/strong> Producing a consistent internal reference from the name and category.<\/p>\n AI must not decide:<\/strong><\/p>\n Prices.<\/strong> Ever. Cost and sales price come from the file or from your pricing rules.<\/p>\n Units of measure.<\/strong> A guess here produces wrong stock and wrong costs, silently, forever.<\/p>\n Specifications.<\/strong> If the file does not say the bracket is stainless steel, the description must not say so either.<\/p>\n That last one is the risk that matters.<\/strong> A model given gaps fills them, plausibly and confidently. Invented specifications reach customers<\/strong>, and in a product catalogue that is a real problem.<\/p>\n FIGURE 1: WHAT AI DOES AND WHAT IT MUST NOT<\/p>\n Safe for AI<\/p>\n Never AI<\/p>\n Read the file.<\/strong> Excel or CSV into structured rows.<\/p>\n Validate each row.<\/strong> Does it have a name? A price? Anything required?<\/p>\n Check whether the product already exists.<\/strong> By supplier reference or SKU. Skip or update \u2014 do not create a second one.<\/strong><\/p>\n Generate the missing fields.<\/strong> Description, category suggestion, reference.<\/p>\n Create the product<\/strong> \u2014 as a draft or unpublished, not live.<\/p>\n Report what happened.<\/strong> Created, skipped, failed, and why.<\/p>\n FIGURE 2: THE CHAIN<\/p>\n Read and validate<\/p>\n Check existing<\/p>\n Generate what is missing<\/p>\n Create as draft<\/p>\n The single biggest factor in output quality.<\/strong><\/p>\n Supply the actual attributes<\/strong> from the row \u2014 dimensions, material, colour, size, whatever the file contains.<\/p>\n Supply your category list.<\/strong> A model choosing from categories you gave it produces usable values. A model inventing categories makes a mess of your product configuration that somebody has to unpick.<\/p>\n Supply examples.<\/strong> Three descriptions you like produces far better output than a paragraph describing what you want. This is the highest-return technique available.<\/p>\n Say what it must not claim.<\/strong> Certifications, compliance, compatibility, materials not stated in the file. An explicit instruction works<\/strong>, and it prevents the most damaging category of error.<\/p>\n Not live. Not published. Not sellable yet.<\/strong><\/p>\n Why:<\/strong> somebody has to check the descriptions before customers see them, and the volume that makes this workflow worthwhile is the same volume that makes one systematic error expensive.<\/p>\n A workable arrangement:<\/strong><\/p>\n The workflow creates products unpublished, in a state that means “generated, not reviewed”.<\/p>\n Somebody reviews a batch \u2014 reading for truth, not for grammar.<\/p>\n Products are published in groups as they are checked.<\/p>\n The saving is still large.<\/strong> Reviewing 400 descriptions is a day. Writing them is weeks.<\/p>\n Worth stating plainly, because it changes who should do the reviewing.<\/p>\n The grammar will be fine.<\/strong> Generated text is fluent.<\/p>\n The question is whether the claims are correct<\/strong> \u2014 and answering that needs somebody who knows the products, not somebody who can read.<\/p>\n What to check specifically:<\/strong><\/p>\n The habit that prevents most expensive mistakes.<\/strong><\/p>\n Run the workflow on ten rows. Review those ten properly. Find the systematic error<\/strong> \u2014 and there usually is one, in the first attempt.<\/p>\n Adjust the instructions. Run ten more.<\/p>\n Then scale.<\/strong><\/p>\n Generating everything at once means discovering the systematic error after it is everywhere<\/strong>, in four hundred products, some of which are already published.<\/p>\n FIGURE 3: WHAT KEEPS THIS SAFE<\/p>\n Give it real attributes<\/p>\n Supply your category list<\/p>\n Create as draft<\/p>\n Generate ten first<\/p>\n Same rule as every import.<\/strong><\/p>\n Check before creating<\/strong>, on a stable identifier \u2014 the supplier’s product code, a SKU, a manufacturer reference.<\/p>\n Never on the product name.<\/strong> Supplier names get reformatted between files, and matching on them creates duplicates every time the format changes.<\/p>\n And decide:<\/strong> if the product exists, update it or skip it? Updating keeps prices current. Skipping avoids overwriting descriptions somebody has already reviewed.<\/p>\n Usually: update the price, leave the description alone.<\/strong><\/p>\n Worth its own warning.<\/p>\n Odoo blocks changing a product’s unit of measure once it has stock moves or order lines.<\/strong><\/p>\n Which means a wrong unit at creation is effectively permanent.<\/strong> Fixing it means creating a new product and archiving the old one.<\/p>\n Never let a model guess this.<\/strong> Take it from the file, or set a default and have it reviewed. If the file does not specify, that is a question for the supplier, not for a model.<\/p>\n AI turns a supplier spreadsheet into a usable catalogue by writing descriptions and suggesting categories<\/strong> \u2014 the parts that would otherwise take weeks.<\/p>\n It must not touch prices, units of measure, or specifications the file does not contain.<\/strong><\/p>\n Give it real attributes, your category list, and examples.<\/strong> Create as draft. Review for truth, with somebody who knows the products.<\/p>\n And generate ten before four hundred.<\/strong> That one habit catches the systematic error while it is still cheap.<\/p>\n A supplier catalogue with no descriptions?<\/p>\n Get in touch. We build product import workflows that generate the writing and leave prices, units and specifications where they belong.<\/p>\n<\/div>","protected":false},"excerpt":{"rendered":" The problem A supplier sends a spreadsheet of products. You need them in Odoo. What is in the file: product names, maybe a code, maybe dimensions, a price. What Odoo needs: a name, an internal reference, a category, a unit of measure, a cost, a sales price \u2014 and for anything customer-facing, a description worth […]<\/p>\n","protected":false},"author":3,"featured_media":9292,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[28],"tags":[],"class_list":["post-9291","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-n8n"],"yoast_head":"\nWhere AI helps and where it does not<\/h2>\n
\n
\n
The workflow<\/h2>\n
\n
\n
\n
\n
Give the model your real data<\/h2>\n
Create as draft<\/h2>\n
Review for truth, not style<\/h2>\n
\n
Generate ten before you generate four hundred<\/h2>\n
\n
\n
\n
\n
Duplicates<\/h2>\n
Units of measure<\/h2>\n
The short version<\/h2>\n