Below is a concise, practical guide plus ready-to-use prompt templates you can paste into ChatGPT (or another LLM) to generate product, category, and CMS content for PrestaShop. It covers structure, required inputs, examples, output formats (JSON / CSV), SEO fields, multilingual needs, and iteration tips.
How to think about a “perfect” prompt
- Be explicit: tell the model the task, format, and constraints (length, tone, fields).
- Give examples and a desired output format (JSON, CSV, plain text with headings).
- Provide product data (name, features, specs, audience, keywords). The better the input, the better the output.
- Ask for multiple variations (A/B) and short + long versions.
- Ask the model to produce import-friendly output (CSV/JSON with exact field names).
- Request checks for accuracy, claims, and compliance.
Essential product data to include in the prompt
- Product name, SKU
- Short summary / USP
- Key features & specs (materials, dimensions, weight)
- Target audience / use cases
- Brand voice / tone (e.g., friendly, technical)
- Primary and secondary keywords for SEO
- Any forbidden claims (e.g., “medical”, “guarantees”) or legal notes
- Required fields for your PrestaShop import (short description, long description, meta title, meta description, tags, categories, image filenames)
General prompt template (structure)
- System role (optional): define persona (copywriter, SEO specialist)
- Task: produce specific fields
- Input: include the product data
- Output: format and constraints (JSON/CSV, character limits)
- Extra: ask for variations, alt text, FAQs, cross-sell suggestions
Ready-to-use ChatGPT prompt templates
1) Product (single) — full output (JSON)
Use when you want structured output to import or paste into PrestaShop.
System message suggestion:
You are an expert e-commerce copywriter and SEO specialist for PrestaShop stores. Produce concise, accurate product content ready for import.
User prompt:
Create product content for PrestaShop. Output must be valid JSON with these keys:
{ "sku", "name", "short_description", "long_description", "meta_title", "meta_description", "tags", "category", "features", "specs", "image_alt_text", "additional_information", "faq", "cross_sell_suggestions" }
Product data:
- sku: PSJKT-001
- name: "AeroDry Men's Waterproof Hiking Jacket"
- USP/summary: Lightweight, fully waterproof membrane, breathable, packable
- features: 2-layer waterproof membrane, DWR finish, taped seams, adjustable hood, pit zips, chest pocket
- specs: weight 320g (M), materials: 100% polyester + PU membrane, color: slate grey, sizes: S–XXL
- target audience: active hikers, weekend adventurers, commuters in rainy climates
- primary keywords: "waterproof hiking jacket", "lightweight hiking jacket"
- secondary keywords: "packable rain jacket", "men's hiking jacket"
- brand voice: helpful, crisp, outdoorsy, trustworthy
- character constraints: short_description ≤ 160 chars; meta_title ≤ 60 chars; meta_description ≤ 160 chars
- forbidden: no medical claims, no absolute lifetime guarantees
Produce:
1) short_description (≤160 chars)
2) long_description (HTML-friendly, up to 600 words; include 2 short bullet lists: key benefits and technical specs)
3) meta_title and meta_description within constraints including primary keyword
4) tags (comma list)
5) image_alt_text for main image (1-2 short phrases)
6) 3 suggested cross-sell products (names)
7) 3 FAQs with short answers (max 30 words each)
Return only JSON.
(Example expected output format)
{ "sku":"...", "name":"...", ... }
2) Product short + long variation prompt (for copy drafts)
You might want multiple tones and lengths.
Prompt:
Write 3 variations of a short product blurb (max 160 chars each) and 3 variations of a long product description (200–400 words each) for the product below. Provide a heading for each variant like "Short A", "Short B", "Long A", etc. Maintain brand voice: outdoorsy, helpful. Focus keyword: "waterproof hiking jacket". Product data: [paste same data as above]. Also provide a 10–12 word meta description candidate for each long variant.
3) Category description (SEO + shopper-focused)
Prompt:
Write an SEO-friendly category page description for the category "Men's Waterproof Jackets". Provide:
- A short intro (1–2 sentences)
- A longer category description (150–350 words) that includes primary keyword "men's waterproof jackets" and natural related phrases
- A 50–80 words meta description
- 5 category-level selling bullet points (1 line each)
- Suggested internal links anchor text (3) to other site pages (e.g., "breathable hiking trousers", "waterproof backpack covers", "size guide")
4) CMS page (About / Brand story)
Prompt:
Create an "About Us" page (350–500 words) for an outdoor apparel brand named "AeroTrail" that emphasizes sustainability, durable gear, and community. Tone: genuine, slightly informal. Include:
- A headline
- 3 short sections with headings: Our Story, Our Materials, Community & Care
- Call to action (join mailing list)
- 1–2 short SEO meta suggestions (title + meta description)
5) Bulk generation — CSV-ready products
When generating many products, ask for CSV output with exact field headers matching your PrestaShop import. Example request:
Prompt:
I need a CSV for bulk import. Create 10 product rows for the following items (provide list of product names + short features for each). Output CSV with headers: "sku,name,price,quantity,short_description,long_description,category,meta_title,meta_description,tags,images". Ensure no commas break fields or wrap fields in double quotes. Provide realistic but unique SKUs and short descriptions (≤160 chars), and meta descriptions (≤160 chars). Use the following product inputs: [list]. Return only the CSV content.
6) Image alt text and filenames
Prompt:
For each product below, generate 3 image alt text options (concise, keyword-rich) and suggested image filenames (lowercase, hyphen-separated, .jpg). Product: [paste]. Output as JSON or CSV.
7) Attributes & combinations (sizes, colors)
Prompt:
Given product SKU PSJKT-001 with sizes S, M, L, XL and colors: slate-grey, navy, olive — generate PrestaShop-ready combinations with unique reference codes, e.g., PSJKT-001-GR-M, stock qty per combination (suggest logical distribution), and a table-like JSON output: { "sku","combination_reference","size","color","qty","ean13_or_upc_if_any" }.
8) Translation / localization
Prompt:
Translate the following product short_description and meta_description to French and Spanish. Keep length similar and keep primary keyword translated and natural. Provide output blocks labeled EN/FR/ES with both short_description and meta_description.
9) FAQ + schema (structured data)
Prompt:
Create 5 FAQs for the product, each with a short question and answer. Also produce JSON-LD schema for product + FAQs (basic fields: name, image, description, sku, brand, offers.price, offers.currency, aggregateRating optional). Make sure the description uses the product short description.
Practical examples (short)
Example 1 — Single-product prompt you can paste:
You are an e-commerce SEO copywriter. Create product content for PrestaShop in JSON. Keys: sku,name,short_description(≤160 chars),long_description(HTML),meta_title(≤60 chars),meta_description(≤160 chars),tags,image_alt_text. Product details: sku PSJKT-001, name "AeroDry Men's Waterproof Hiking Jacket", features: 2-layer waterproof, DWR, taped seams, adjustable hood, pit zips; specs weight 320g (M), materials polyester+PU membrane; target: hikers; keywords: "waterproof hiking jacket","lightweight hiking jacket". Tone: outdoorsy, helpful. Output only JSON.
Example 2 — Category prompt:
Write a 200-word SEO-friendly description for category "Camping & Hiking Jackets", include keyword "camping jackets", 50–70 char meta title, 150 char meta description, and 5 bullet selling points. Tone: inspiring, practical.
Tips for best results / troubleshooting
- Provide sample content: include one or two example product descriptions you like as guides.
- Use the system message to fix voice/tone across many requests (ChatGPT + system role).
- Ask for multiple variants and rank them, then ask the model to combine the best parts.
- Use strict format constraints when you plan to automate import — ask for JSON/CSV only.
- Validate generated specs and claims (weights, waterproof ratings). AI can hallucinate numbers — don’t publish without checking.
- For multilingual stores, generate content in the exact language and ask for local keyword variants.
- If results are too generic, add competitive context: “Write copy that beats competitors A, B — emphasize lighter weight and better breathability.”
- For SEO, supply target keywords and ask the model to include them naturally (avoid keyword stuffing).
What to avoid
- Don’t leave the model to invent specs or certifications (e.g., IP ratings, guaranteed lifetimes).
- Don’t rely solely on AI for legal or safety claims.
- Avoid prompts that are too vague (e.g., “Write product text” without details).
Quick “starter” prompt you can paste and reuse
You are an expert e-commerce copywriter and SEO specialist. For each product I give you, produce: (1) a short_description ≤160 characters, (2) a long_description (HTML-ready, 150–400 words with 2 bullet lists: benefits and specs), (3) meta_title ≤60 chars with the primary keyword, (4) meta_description ≤160 chars. Keep tone [brand voice]. Do not invent specs. Use these inputs: [paste product name, features, specs, keywords, target audience]. Return only JSON with keys: sku,name,short_description,long_description,meta_title,meta_description,tags,image_alt_text.
Final note
If you want, paste one product (or a small CSV) here and I’ll create the exact prompt + example outputs (short description, long description, meta tags, alt text) tailored to your store and brand voice.
