How to use AI to write product descriptions without sounding generic

26/05/2026
  • Learn how to make the most of AI to write product descriptions without sounding generic, with useful prompts.

  • If you have ever used ChatGPT, Claude, Gemini… or any other AI tool to write a description in your product page, and the result has sounded like a hypermarket catalogue or a stale shop from the 90s, rest assured: you are not alone.

    “High quality”, “elegant design”, “perfect for any occasion”… are just empty phrases that say nothing to anyone and that, on top of everything, could describe anything: from trainers to an office chair.

    So, AI is not useful for creating product pages? Of course it is, you just have to learn how to ask it for things.

    In this post, we are going to explain how to make real use of these tools, so that your product pages truly sell, are unique, sound like your brand and do not seem generated by popping out like an assembly line.

    Notice: this post contains prompts that you should save.
  • Do you know that generic descriptions cost you sales?

  • A product description is not only an informative text, and it is not an excuse to stuff keywords for SEO either. The product page text is, in fact, the salesperson who works 24 hours for you (and without charging commissions). If that salesperson limits themselves to vaguely describing your product and, what is worse, is the same employee your competition has hired… we are not adding anything to our eCommerce.

    Because when the customer does not feel that they are being given special treatment, when they do not have the feeling of speaking with an expert who advises them, nor is there someone working on persuasiveness on the other side… they do not buy. Or worse: they buy elsewhere where someone has taken the trouble to treat them as expected.

    To that, add the impact on SEO. Google does not reward content that sounds the same as that of a hundred competitors. If your product pages look as though they came out of the same mould as everyone else’s, your visibility in search engines is going to suffer, and a lot.

    And then there is GEO and LLM engines for AI. Bear in mind that if you have the same content as other eCommerce sites with more authority, the chances of your product pages being cited fall sharply.

    So this is not a trivial matter.
  • The most common mistake: treating AI as a text machine

  • Most online store owners who use AI for their descriptions do the same thing: they open the chat, write “describe this product for me” with four basic details and expect magic.

    The result, you already know it: correct, odourless, colourless and tasteless text. With the same stylistic resources and the same structure.

    AI does not know who your customer is. It does not know whether you sell to parents looking for quick solutions, to sports enthusiasts who want technical specifications, or to design lovers who buy with their eyes. If you do not tell it, it will write for everyone. And writing for everyone is, in practice, writing for absolutely no one.
  • How to build a good prompt for product descriptions?

  • Think of the prompt as the detailed instructions you would give to a copywriter.

    The more context you give it, the better it will work.

    These are the elements that should not be missing:

    • Define your ideal customer. Not “people interested in this product”. Be specific. For example: “women aged 30-45 who are looking for comfortable clothes to work from home, while maintaining a certain style”. The more specific, the more relevant the text will be; do not be afraid of detail.
    • Specify your brand tone. Are you approachable and casual? Technical and rigorous? Aspirational? Define how you speak and ask the AI to maintain that register at all times. It is no bad thing to give it some examples of texts written by you; it can extract many nuances from there.
    • Focus on the benefit, not only on the features. Of course, a coat is “80% wool with double-breasted fastening”. But, above all, it is “the coat that gets you to work without being cold and without looking like you are wearing a sack”. Technical features inform, but if you want to convince, use benefits as a differential advantage.
    • Add style restrictions. Explicitly ask it to avoid certain clichés. Something as simple as “do not use words such as “premium quality”, “exclusive” or “perfect for…””, improves the result immediately.

    PROMPT: “write a 100-word description for [product] aimed at [ideal customer], with a [adjective] tone, highlighting the benefit of [specific benefit]. Avoid generic adjectives such as 'elegant', 'versatile' or 'high quality'”.

    The difference from the undeveloped approach is enormous.
  • Practical examples: from the prompt to the description

  • Without examples, this would be very poor, would it not? So that you can see better how the instructions we are giving to the AI are not a mould and adapt very well, we are going to do a simple exercise, comparing a prompt for a product and the difference between a generic response and a well-oriented one.
  • #1— Work backpack

  • Basic prompt: “Write a description for a black work backpack with a laptop compartment.”

    RESULT: High-quality work backpack in black. Includes padded compartment for laptop up to 15". Elegant and versatile design, perfect for any occasion. Made with resistant and durable materials.

    Optimised prompt: “Write a 90-word description for a black work backpack with a 15" laptop compartment, aimed at urban professionals aged 28-40 who travel by public transport and need to have everything organised without looking as if they are carrying a travel suitcase. Direct and practical tone. Avoid ‘high quality’, ‘elegant’ and ‘versatile’.”

    RESULT: Everything you need for the day, while avoiding chaos. The backpack has a padded compartment for laptops up to 15", organiser pockets so that the charger, keys and headphones do not all end up together in the same black hole, and a slim profile that does not turn you into an obstacle on the metro. Black, discreet, functional. For going from home to the office without thinking about what you are carrying.
  • #2 — Natural facial cream

  • Basic prompt: “Create the text for the product page of this moisturising facial cream with natural ingredients.”

    RESULT: moisturising facial cream made with high-quality natural ingredients. Its exclusive formula nourishes and moisturises your skin deeply, leaving it soft and luminous. Suitable for all skin types. Free from parabens and artificial colourants.

    Optimised prompt: “Write an 85-word description for a moisturising facial cream with rosehip oil and hyaluronic acid, aimed at women aged 35-50 with dry or combination skin who have been trying products for some time without seeing results and distrust empty promises. Honest tone and without exaggerations. No “exclusive formula”, “luminous” or “revolutionary””.

    RESULT: “We do not promise you miracles, only science. What we can tell you is that rosehip oil has been used for decades to regenerate damaged skin, and that hyaluronic acid retains moisture in a proven way. This cream combines the two in a light texture that is absorbed without leaving a greasy feeling. Apply it in the morning, before sunscreen. Nothing more, nothing less.”
  • Never publish the first version

  • There is no magic prompt for creating product pages in eCommerce; that does not exist. AI is a tool to speed up the process, not to automate it 100%.

    The first draft is exactly that: a draft. What works is using it as a starting point and then introducing variables: ask it to make it shorter, more direct, with a stronger hook in the first line, with a call to action at the end... modify what does not fit and polish it until it truly convinces you.

    You can also ask it for several different versions and combine the best of each one. In ten minutes you get three different approaches to the same product and can thus choose the one that works best for your audience.

    That, before these tools, took hours.
  • The 20% that makes the difference

  • No matter how well you explain the context to AI, there are things that only you know: why you chose that supplier, what the first customers who bought it told you, what problem you had to solve in your own life.

    That type of detail (a brief anecdote, a phrase that came from a real review, a piece of information that only you know… storytelling) is what, in the end, ends up turning a correct description into a memorable description.

    Use AI to do 80% of the work. The remaining 20% is yours. And that 20% is exactly what no one else can replicate and what makes the difference.

    As a reflection, keep in mind that AI should not write for you, but it can help you write better, faster and with more consistency if you learn to direct it.

    Start with one of your best-selling products, apply these principles and compare the result with what you had before. I think you will be surprised by the differences.
  • Is it clear to you how to use AI to write product descriptions without sounding generic? Would you like us to write another post on prompting for eCommerce? You only have to ask!

Miguel Nicolás


Miguel Nicolás O'Shea is a life-long copywriter (more than 20 years working in agencies) and a specialist in Search Marketing (SEO and PPC). From now on, he will contribute with his online marketing experience to Oleoshop, publishing regularly.

search posts

Last posts

This website stores data as cookies to enable the necessary functionality of the site, including analytics and personalization. You can change your settings at any time or accept the default settings.

cookies policy

Essentials

Necessary cookies help make a web page usable by activating basic functions such as page navigation and access to secure areas of the web page. The website cannot function properly without these cookies.


Personalization

Personalization cookies allow the website to remember information that changes the way the page behaves or the way it looks, such as your preferred language or the region in which you are located.


Analysis

Statistical cookies help web page owners understand how visitors interact with web pages by collecting and providing information anonymously.


Marketing

Marketing cookies are used to track visitors on web pages. The intention is to show ads relevant and attractive to the individual user, and therefore more valuable to publishers and third-party advertisers.