We built an app inside Shopify that fills in the data the ERP doesn't sync and reviews the catalog every 30 minutes.
Result: over 20,000 complete products in every import and a catalog with no incorrect products published.
case study
Client
Sector
about the case study
Fesmés wanted to use artificial intelligence to scale its e-commerce business, improve catalog quality and cut the manual work needed to manage thousands of products.
We defined a strategy that connects the information available in SAP with AI-powered content generation systems, adds assistants to guide users through their purchase and applies AI tools to the technical development and evolution of the online store.
In a DIY online store, the catalog is one of the biggest drivers of conversion.
Fesmés works with a large volume of products whose data often starts out as structured information and bare-bones descriptions coming from its SAP ERP. That information is enough for internal management, but not always for building a product page that is ready to sell online.
Manually turning that data into titles, descriptions, features, benefits and content tailored to the digital channel takes time and limits how quickly the team can expand or improve the catalog.
The challenge was to find a way to make use of the information already in the company’s systems and automatically turn it into useful content for the online store, while keeping shared standards of quality and consistency.
At the same time, the project set out to apply AI at other points in the customer journey: improving search, guiding users through their purchase and speeding up certain programming tasks and the evolution of the platform.
The project began with a consulting phase to analyze the existing process for creating and maintaining product pages.
We looked at what information was available in SAP, how it was carried over to the online store and which data could serve as a basis for generating content with artificial intelligence.
Based on this analysis, we defined a system in which structured ERP information can be turned into more complete product pages using ChatGPT.
AI makes it possible to transform technical data and short descriptions into content suited to the digital environment, following predefined guidelines on structure, tone, commercial information and consistency across products.
The system is designed to reduce the team’s repetitive work and make it easier to add new products to the catalog without writing every page from scratch.
The project also brings tools such as Doofinder, Netrivals, Commerzia and PrestaShop into the store’s technology ecosystem, along with AI agents designed to guide users through their purchase and help them find the products that best fit their needs.
We also introduced AI into certain coding and programming processes to speed up the development, testing and evolution of features on fesmes.com.
The key is not adding a single AI tool, but integrating it at different points in how the online store works: catalog, search, shopping experience and technical development.
−70%
less time to create and update each product page
+3.000
product pages generated from SAP data
x3
products added to the catalog each month
The rollout reduces the time the team spends creating and completing product pages, making it easier to manage a larger catalog without increasing the operational workload at the same rate.
Automation also improves the consistency of published information. Product pages can follow a common structure and give users more information, drawing on data already available in the company’s internal systems.
This allows Fesmés to expand and improve its catalog faster and get more products ready to sell online.
The agents and tools added to the purchase process also help users find products, answer their questions and move forward with less friction in the online store.
Internally, using AI for content management and programming tasks reduces repetitive work and frees the team to spend more time on commercial decisions, optimization and the evolution of the digital channel.
The result is an online store with a structure that is better prepared to scale: more products available, less manual work, better information for users and a greater ability to turn catalog growth into revenue growth.
Let's talk
Tell us how you attract customers, how you sell and what you want to improve. We'll assess whether you need more lead generation, AI content, CRO, an assistant, smart search or a combined solution.