Category: Non classé

Trust: a vital issue for marketplaces

The success of marketplaces is based on a simple principle: connecting sellers and buyers seamlessly. But behind this promise lies a persistent challenge: trust. Counterfeit products, fake customer reviews, uncertainty about provenance and late payments still too often undermine the experience.
In this context, blockchain appears to be a technology capable of enhancing transparency and restoring trust in the long term. Far from being reserved for cryptocurrencies, it offers concrete and powerful applications for e-commerce.

Blockchain, in simple terms

Often perceived as complex, blockchain can be summarised as follows: a digital ledger shared between several parties, which is tamper-proof and accessible with complete transparency. Each new piece of data entered (transaction, certification, review, etc.) is time-stamped and cannot be modified retrospectively.
Its main characteristics – security, traceability, immutability – make it a relevant lever for environments where trust is central… such as marketplaces.

Practical uses for marketplaces

1. Ensuring product traceability

In sectors such as agri-food, fashion and cosmetics, consumers demand to know where products come from and how they were manufactured. Thanks to blockchain, each stage – from production to delivery – can be recorded and consulted. This gives buyers access to a ‘digital passport’ for their product.

2. Combating counterfeiting

Marketplaces are regularly confronted with the sale of counterfeit products. By associating a unique blockchain certificate with each item, it becomes possible to instantly verify its authenticity. For brands, this is a way to protect their image; for customers, it is a guarantee of trust.

3. Making customer reviews tamper-proof

Online reviews strongly influence purchasing decisions… but their reliability is often questioned. By recording reviews on a blockchain, we can ensure that they come from real buyers and have not been manipulated.

4. Securing payments with smart contracts

Smart contracts automate certain critical steps. For example, payment can be blocked until the customer confirms receipt of the product. The result: fewer disputes and greater peace of mind for all parties.

Benefits for all parties involved

• For buyers: transparency regarding product origin, verified reviews, secure transactions.
• For sellers: promotion of quality, reduction in disputes, improved customer relations.
• For marketplace operators: a unique market positioning, an image of reliability and increased appeal to reputable sellers.

Challenges to overcome

The promise is appealing, but it is not all plain sailing. Setting up a blockchain infrastructure represents a significant investment, both financially and technically. Some public blockchains still suffer from slowness and scalability issues.

Added to this is the need for education: many customers do not yet have a clear understanding of blockchain, which means that efforts must be made to explain it in layman’s terms in order to build trust. Finally, regulatory aspects, particularly in Europe with the GDPR and the future MiCA regulation, will have to be incorporated into any initiative.

Towards next-generation marketplaces

If these challenges are met, marketplaces could transform into truly trustworthy ecosystems. Each product will have its own unique digital identity, each transaction will be secure, and each review will be verifiable.

Blockchain does not replace marketplaces: it strengthens them. It paves the way for a new generation of platforms, where transparency and reliability become major competitive advantages.

Blockchain has the potential to revolutionise the way marketplaces operate. By providing traceability, security and transparency, it directly meets consumer expectations and seller needs.

More than a technological innovation, it is a strategic tool for building trust – that rare resource that drives growth and loyalty in e-commerce.

Tomorrow, in the era of Web3, marketplaces that know how to take advantage of blockchain could well become the leaders of a new standard: that of total transparency.

Have you heard of generative AI? For the past few months, this branch of artificial intelligence has known growing success brought on by the advent of such tools as ChatGPT, Dall-E, and Midjourney.

These tools seem to breed innovation as much as controversy and generative AI is well on its way to disrupting long-established industries.

The e-commerce sector, in particular, will experience profound changes due to this new form of artificial intelligence. In this article, we’ll touch upon the concept of generative AI and the functionalities that could help online retailers boost their sales.

What is Generative AI ?

Generative AI – also known as GenAI – is an artificial intelligence system capable of creating content from data models. This content can take many forms, including text, images, video, or any other media.

In recent months, several startups have garnered attention, sometimes for their innovative products and other times for the stratospheric amounts they have raised or because of the controversy they have stirred up. Some of these startups include:

• OpenAI, the entity behind ChatGPT and Dall-E, which raised €175 million last May
• MidJourney, creator of the eponymous tool MidJourney
• Google – bolstered by the phenomenal success of ChatGPT – which very recently created Bard

How does Generative AI actually work ? 

So, how does generative AI operate in practice? How can you put it to work in your company or use it for a given project?

It all starts with a query – also known as a prompt – formulated by the user in the form of a question or command for the generative AI. The AI generator has undergone training and has been exposed to extensive amounts of data.

It then uses these data sets and its artificial neural networks to provide an original response.

Record-Breaking Mass Adoption

The massive adoption of AI is puzzling many experts around the world. ChatGPT, for example, reached the 100 million user mark in just 2 months. Comparatively, it took 9 months for TikTok to achieve similar numbers and 24 months for Instagram.

What’s clear is that the more consumers choose to embrace these tools, the more companies will offer generative AI-based services.

Where does e-commerce come in ?

We’re getting there! Generative AI models have disrupted many sectors and hold the promise of tremendous technological advances. E-commerce is, without a doubt, one of the sectors that will be closely affected by these innovations. Here are 5 examples of how generative AI could be relevant to e-commerce.

1. AI Generators for products recommendations 

Among the functionalities an AI generator can provide to address the challenges of e-commerce and marketplaces, product recommendation generation turns out to be the most promising.

This type of technology makes it possible to create a personalized browsing path for each customer or prospect based on their preferences previously identified by the algorithms. Generative AI can be a differentiating asset in the context of loyalty strategies as it provides a deeper understanding of consumer habits.

Customer data is at the heart of strategic stakes for most online merchants. Unfortunately, many don’t know how to take full advantage of it or don’t have the resources to do so. AI generators, on the other hand, are capable of processing a huge amount of data in very little time. This allows them to identify product trends and showcase certain items accordingly.

That said, utilizing generative AI shouldn’t make us lose sight of legal regulations concerning the use of personal data. In the European Union – and some other regions – users give their consent to have their behavior analyzed by accepting the presence of cookies on websites. In principle, no AI-based marketplace is exempt.

2. AI-Generated product descriptions

Product descriptions are a form of text-based content designed to enhance the user experience (UX) and improve the product page’s search engine optimization (SEO), ensuring its visibility in search engine results.

When managing a very large number of items, this kind of task can quickly become tedious.

The natural language processing abilities of generative AI make it possible to produce natural language content by analyzing product data. AI generators are especially well-suited to this type of task since product descriptions always tend to follow the same structure:

• General characteristics of the product. 
• Its benefits. 
• Its specificities. 

Thus, depending on the type of products the online retailer offers, the components of a textual product description can be essentially the same. For instance, a store that sells furniture and home decor will always include information about the dimensions of the products, which AI is fully capable of understanding through data analysis.

For example, Phrasee is a platform dedicated to product description generation for e-commerce players. It relies on generative AI to come up with original product descriptions.

3. AI Image Generators

One of the downsides of e-commerce is the inability to interact with the products physically. This means product images are crucial for triggering purchases. For online shop owners who need to process many items, taking pictures and editing images can be time-consuming and costly. Using pictures created by an AI image generator thus becomes a compelling alternative to help renew the presentation of products on a website.

Based on generative adversarial networks (GAN), product image generation uses a technology that employs pictures of existing products, which the algorithm analyzes repeatedly until it becomes capable of developing substantially similar content.

Among the tools that can be utilized to generate product images, DALL-E 2 – mentioned earlier – deserves the spotlight. This easy-to-use generative AI has already conquered the field of advertising and is now making progress in the e-commerce sector.

With that said, despite the performance of generative AI platforms dedicated to product images, this application isn’t yet suited to all types of retailers. For the sake of the brands’ credibility in the eyes of their customers, the authenticity of the images remains crucial.

4. Building close relationships with customers

Chatbots – or conversational agents – can provide real-time assistance for the users by answering generic questions. This does away with the last remaining barriers to purchase and is an effective way to improve the conversion rate of e-commerce websites.

5. Fine-tuning internal search engines 

Users rely on marketplace search engines extensively. This means they expect clear, swift responses and a selection of products that match their original search intent. 

Thanks to generative AI, search engines will continue to become increasingly powerful as time goes on. Retail giants such as Amazon are fully aware of the potential and seek to incorporate these improvements right into their applications. In the United States, 56% (1) of product searches begin on the American marketplace.

However, Generative AI raises many controversial issues

We can’t mention generative Artificial Intelligence without addressing a few areas of concern. As innovative as AI generators may be, it should be noted that they will not replace the cognitive abilities of human beings, at least not in the short term.

Instances of inaccurate information

Generative AIs such as ChatGPT sometimes offer erroneous information, and generative AI models are still in their learning phase. We believe generative AI can be a complementary tool to human intelligence, but it is still far from being able to replace it altogether (thankfully!). If you want high-quality content, be circumspect about the raw information these tools provide. We recommend you always double-check it thoroughly.

What about our personal data ? 

When we converse with a chatbot online, we may be in a position to give out some of our personal information. This data is collected and stored by the conversational agent. That’s when it can get difficult to get it back if we no longer want it to be used or if we are worried about it being sold to third parties.

As you may have gathered, generative Artificial Intelligence is opening up a whole world of possibilities in various sectors as consumers continue to embrace it.

E-commerce is one of the areas where its hold is already tangible and where it will keep on increasing in the years to come.

These AIs can be tremendous business drivers and offer significant time savings. However, the limitations of AI systems – still in their learning phase – and their propensity to sometimes deliver inaccurate or incomplete information should not be ignored.

---

Sources

(1) “Every single” Amazon team is working on generative AI, The Verge, 2023

1. Stocks and logistics

Indeed, in order for an e-merchant to generate significant revenues, he must have a large depth of catalog: vast references, regularly supplied stocks… This requires a significant investment.
In addition, the manager of a merchant site must also be able to manage logistics and the shipment of products directly to the end customer.
On the contrary, the Marketplace operator does not have an inventory: his role is to put buyers in contact with third party sellers. It is the latter who directly manage the inventory and ship the products.

2. Professions

Although some functions are found in both activities, such as management or marketing, the core business of the Marketplace operator and the e-merchant differ.
Indeed, the job of a Marketplace operator is above all to find and integrate sellers on its platform, as well as to manage the acquisition of traffic on the site.
The manager of a merchant site must focus on the purchase of merchandise, customer support (the part delegated to the sellers in the “Marketplace” model), manage its stocks soundly and take charge of the shipment of its products.

3. Business model

In a classic e-commerce scheme, the manager buys products, which he then resells on his site at a higher price, in order to generate a margin for his business. A good budget management is thus induced by a good sourcing of suppliers.
The Marketplace operator, for its part, is generally remunerated on a commission taken from each sale. His commercial approach is therefore based on business volume: the higher the sales, the more his income will increase.

4. Catalog

One of the great strengths of the Marketplace is the decentralization of inventory to its vendors. This makes it possible to aggregate an almost infinite number of references, to put sellers in competition with each other on certain products, thus making it possible to offer attractive prices (and thus increase sales), and to push merchants to display impeccable customer service (thanks to the instructions left by buyers).
An e-merchant’s catalog, on the other hand, depends on its storage and logistics capacities, and is therefore very limited.
For example, on the Amazon site, king of e-commerce and logistics, there are 2.8 million “Amazon” products, the rest, more than 285 million, are sold by Marketplace sellers.

5. Legal status

From a legal point of view, the Marketplace operator acts as a broker, providing a secure platform for the transaction. The legal responsibility for the products is therefore imputed to the seller.
Conversely, the manager of an e-commerce site is legally responsible for the products sold.
And if a buyer decides to take legal action following damage caused by a product/service sold on the Marketplace, the operator is covered: it is the seller who must be heard.
More information on legal issues can be found on the website of Haas Attorneys at Law, which specialises in the Marketplace.

6. Technology

While building an e-commerce site is now simple, thanks to numerous SaaS or open-source technologies (from Shopify to SAP Hybris, via Magento), few platforms allow you to create a marketplace. E-commerce CMSs are in fact not natively thought of as “multi-seller” (and all the attributes that come with it: payment, invoicing, synchronisation of flows with external IS, commission models, logistics management etc.).
IZBERG, the most complete Product and Services Marketplace solution on the market, meets precisely this need, and now equips many players, such as Vente-Privée, Europ Assistance, Suez and Gifi. In addition, IZBERG can be integrated in less than 4 months with any type of front-end technology, from e-commerce CMS to a custom framework.
Thus, for a multi-brand pure-player, the Marketplace is much less risky than a classic merchant site. And this disintermediation mechanism will see two strong trends in the years to come: the B2B sector, which is still not very digitalized, and Services platforms, making it possible to reinvent sometimes dusty business models.