The AI Revolution in MDM, PIM, and DAM – DATASOLUTION and MaPS System Are Here to Support You

Rapid technological advancements have paved the way for an unprecedented digital revolution. Today, artificial intelligence (AI) is emerging and establishing itself as a driving and innovative force in many fields. Among these constantly evolving fields is enterprise product information management plays a central role. 

In the age of big data and interconnectivity, companies face a major challenge: how can they effectively manage and leverage the vast amounts of information at their disposal? 

Faced with the growing complexity of product information management, companies must equip themselves with the right tools and strategies. In fact, AI-powered PIM is becoming an essential foundation for your marketing and sales teams. 

Joint Interview with Two Experts: Kévin Corte, Sales Director at MaPS System, and Arnaud Bernard, PIM expert at DATASOLUTION.

How important is AI in the field of PIM?

Kévin Corte – MaPS System:  

It’s a revolution in the world of product information management (PIM) and master data management (MDM). Artificial Intelligence can be very powerful, if used correctly. 

Before implementing a PIM, we noticed that our prospects typically work with Excel spreadsheets or their website’s back office. They perform repetitive tasks to keep the product information. By implementing a PIM, we offer them the ability to streamline these tasks using a single data model. Then, during the data entry, it is necessary to clean the data—a task that was previously done manually. Today, thanks to artificial intelligence, we are able to to streamline this process and save time in setting up and using the PIM. We saw this in action with our Boticinal project in collaboration with YZR.  

Then, during the PIM implementation phase, to generate descriptions tailored for each persona, and to crop photos automatically instead of going through tedious manual work. When used properly, it’s the equivalent to having an assistant for a product manager—that’s the AI built into the PIM today.   

Overall, it helps to reduce manual labor by automating up to 90% of tasks.  

Arnaud Bernard – DATASOLUTION:  

Artificial intelligence can be very beneficial if used appropriately and if you know how to leverage it. Currently, Hugging Face is the repository that publishes the most usable models, with 106,416 different models that process data in various ways. Some models are specific to sectors such as insurance, retail, or manufacturing. We train these models so that artificial intelligence becomes more effective and can deliver results based on how it has been trained. For me, integrating AI into PIM (Product Information Management) is a catalyst that helps marketing teams correct and improve data. However, the tool cannot completely replace human input, as rigorous preparation for AI training is essential to obtain relevant results. It’s the combination of existing marketing data and technical expertise in AI that makes it possible to create product descriptions tailored for Facebook, Instagram, or different customer segments.  

In what area will AI be most useful in the future for marketers in retail or for brands? 

Kévin Corte – MaPS System:  

First of all, when you load data into an MDM, a PIM or a DAM such as the MaPS System, it is necessary to classify the products. Today AI helps with this categorization, which saves a considerable amount of time when implementing a PIM project.  

The second point concerns the writing descriptions for product pages. In fact,, it’s important to tailor the description to each customer, based on their age, socio-professional category, etc. For example, the description of a car offered to an experienced driver will differ from the one suggested to a young driver. 

The third point concerns the processing and standardization of images, specifically image cropping. The use of AI makes it possible to standardize and ensure consistency in the presentation of each product on a multi-brand e-commerce platform, where numerous photo editing tasks—often performed manually—need to be carried out. 

AI is therefore a valuable ally in addressing the need to optimize online product sales for product teams and category managers, especially since we know that, statistically speaking, a greater number of images is required to optimize online product sales. 

The main benefit is improved sales performance, since 84% of abandoned carts are due to poor product descriptions. 

Arnaud Bernard – DATASOLUTION:  

The benefits of AI will continue to become increasingly evident in the coming months and years; it is worth noting that even today, practical applications are already delivering significant benefits, particularly in the area of product description management. Marketers will shift from spending a great deal of time creating as many product descriptions as required—for example, for different websites or variations of those descriptions (by region, country, short or long versions, etc.)—to generating as many as needed, while being able to specify details such as: “SEO-optimized copy,” “a short version based on this long version,” or “adapted for a Facebook post,” etc. 

Speaking of SEO: Our clients have hundreds of thousands of SKUs (unique product codes) organized into different categories, which requires a considerable amount of work—especially if the content needs to be adapted for different channels (such as e-commerce websites, social media, or marketplaces). AI is a tool we can rely on to generate content that isn’t duplicated.  

For all aspects of retargeting, AI can help generate specific information tailored to the profile of the person browsing the site. The product catalog can thus be dynamic and adapt to any user. The user is identified through cookies or authentication.  

Today, similar products that are recommended are often the result of a very basic algorithm, but AI goes a step further in terms of personalization and dynamic display. 

In fact, the user journey is no longer static; it is no longer limited to simply navigating through a directory structure but is evolving toward a dynamic display based on the user’s profile. 

It’s important to note that, depending on the country, the number of images needed on product pages to optimize the purchase funnel may vary. Decathlon is a perfect example: depending on the market, consumers view product pages differently. In China, for example, it’s essential to include many more images on a product page to help consumers make a purchase decision, whereas in other countries, two photos are sufficient. And that’s where automatic image cropping comes in and makes sense. Because the more images there are, the more time-consuming the cropping process becomes. 

Automated image processing is crucial. The combination of generative AI capabilities and automated processes (cropping, color correction, resampling, etc.) makes it possible to meet the demands of increased image and video production. This accelerates time-to-market and boosts business performance. 

What are some specific examples of AI applications that you have already implemented? 

Arnaud Bernard – DATASOLUTION:  

Some of our DAM solutions already incorporate AI. The results are sometimes still imperfect, but this allows the marketing team to collaborate with graphic designers by using it as a creative brief. 

However, some of the results are already very promising, so this bodes well for the future.  

  Wedia DAM already incorporates AI as a built-in feature within its DAM, particularly for the automatic cropping of images. Today, generative AI is capable of pairing the selected product with various backgrounds and showcasing it in several different contexts.  

In the field of PIM and MDM, there is also the automatic generation of text based on product properties. Today, it is possible to envision workflows in which product descriptions are generated entirely automatically, freeing up time for the marketing team so that its role is limited to review, approval, or editing.  

This is particularly interesting in the context of translation, which can also be integrated into a workflow.   

 Generating short or long text allows you to adapt to different contexts and different marketing channels, such as Facebook or Instagram, for example. 

And let’s not forget the new ChatGPT platform, which is the perfect tool to address this issue.  

While ChatGPT has gained widespread popularity among the general public, we are exploring the diversity of AI applications, whether in text, images, or specialized business processes.  

Kévin Corte – MaPS System:  

One of the major challenges for a company like Boticinal—which, as we’ve noted, operates about 50 pharmacies in its network—is to centralize and standardize all of its resources within a single database. This has been achieved with MaPS System and YZR. In fact, as part of this project, we have enabled product grouping, automatic product categorization, and data reconciliation based on files received from all the pharmacies to populate MaPS System with a single file. 

As part of our R&D efforts, as Arnaud mentioned earlier, we have also been working on prototypes based on ChatGPT, for generating descriptions, etc. This feature could be useful for companies that don’t currently have the necessary resources in-house specialists in writing product descriptions. Here, there are real savings in terms of time and time-to-market.

What are the most significant benefits that AI has provided to the client? 

Kévin Corte – MaPS System:  

Artificial intelligence can offer many benefits to customers, including reducing the time spent on manual tasks and by providing assistance with various stages of creating product listings. The creation a product listing involves at least 10 steps , and thanks to AI, that number can be reduced to 3 or 4. PIM and AI enable the reduction of these steps in addition to the standard features of a PIM system such as MaPS System. This reduction in manual work translates to a faster time-to-market, which can significantly improve business performance. For example, by increasing the conversion rate on a website thanks to the quality of product pages, filters, and access to products via search tools, customers will be able to quickly find the products that meet their needs. 

​​In conclusion, I would say that this helps reduce the risk of project failure. Since half of all tool migration projects (PIM, MDM, and marketplace) fail due to poor data quality.  

Arnaud Bernard – DATASOLUTION: 

Thanks to the advent of AI, our clients can expect a significant improvement in the management of their PIM projects, particularly on e-commerce sites. In fact, AI is capable of generating or correcting high-quality text, which helps reduce project timelines and mitigate risks related to data verification.  

For customers who manage hundreds of thousands of SKUs, this daily task can be very challenging, which is why AI is so valuable in streamlining this process. Thanks to AI, virtually all products can be published with reliable information, and discontinued products can be suspended and removed from catalogs and distribution channels (marketplaces, resellers, web and print catalogs, etc.). We can therefore conclude that AI significantly improves data quality and streamlines the day-to-day management of PIM projects. 

In short, the benefit is an increase in the speed of product catalog management, as we have seen a 75% reduction in the time it takes to bring new sellers to market.

Need more information?

Contact our experts in the MaPS System division to get answers to all your questions about this technology.