Can technical attributes be the key to success in e-commerce?

In today's e-commerce landscape, precise product data is the foundation of success. This is particularly evident in stores with broad, diverse product ranges, where proper management of technical attributes can determine competitive advantage. The challenge lies in the fact that most suppliers and manufacturers don't provide these attributes in a structured format.

This is exactly why we created getName.ai - a tool that automatically analyzes product descriptions and names, identifying key parameters required by sales platforms and industry standards. Regardless of the number of suppliers or description languages, getName.ai effectively extracts valuable technical attributes.

1. Precise product filtering

Today's customer expects quick access to information. Well-defined technical attributes allow them to instantly narrow down their choice to products meeting specific requirements - whether budget-related, technical, or tied to detailed specifications. In practice, this translates to higher conversion rates and greater customer satisfaction.

2. Effective product comparison

Standardized attributes enable customers to quickly compare selected products. Instead of analyzing lengthy descriptions, they receive a clear table with key parameters. This not only speeds up the purchasing decision but also reduces returns resulting from mismatched purchases.

3. SEO optimization

Search engines love structured data. Precise technical attributes increase product visibility in search results, especially for long-tail queries related to specific parameters.

4. Automatic product descriptions

Having a complete set of technical attributes, you can automatically generate consistent and SEO-optimized product descriptions. This saves time and eliminates copywriter errors in technical specifications.

5. Advertising feeds

Google Shopping or Facebook Ads require precise product data. Complete technical attributes allow for automatic creation and updating of advertising feeds, saving dozens of work hours monthly.

6. Cross-selling in Practice

A system using technical attributes can automatically detect compatible or complementary products. Example? Showing appropriate accessories for a camera based on its model and lens mount.

7. Data-Driven Upselling

Knowing exact product parameters, the system can suggest more expensive alternatives differing in specific features important to the customer. This is more effective than general recommendations of pricier products.

8. AI Assistant Integration

Chatbots and virtual assistants can provide precise answers to technical questions when they have access to structured attributes. This increases customer trust and reduces support department workload.

9. Personalized remarketing

Technical attributes enable creating personalized remarketing campaigns. Example? Ads targeted at people who viewed printers with specific print speed or resolution.

10. Intelligent recommendations

The system can build advanced user preference profiles based on viewed technical parameters. This allows for delivering more accurate recommendations, increasing conversion rates.

The method of acquiring product attributes primarily depends on the business model and the company's position in the supply chain. Let's examine four common scenarios.

Manufacturers selling their own products

Theoretically, this is the most advantageous situation - the company has full control over product data. However, practice shows that manufacturers face two significant challenges. The first is internal - teams focus on product development, treating data management as a necessity. The second, often more serious, involves external requirements from trading partners.

Each significant recipient expects their own product data format. This results in the need to maintain multiple versions of the same information. Industry standards like ETIM can offer a solution, though not all sectors have developed such standards yet.

Distributor working with brands

Distributors receive marketing materials and data directly from manufacturers. Sounds good? The problem is that each manufacturer delivers them in a different format - Excel, PDF, XML, or even Word. Add to this differences in attribute naming between manufacturers. Result? The product team spends more time mapping or copying data than on actual offer development.

Selling on Marketplaces

Amazon or Allegro have strict requirements regarding product data quality. Sellers must adapt their attributes to each platform's standards. This means double or triple work - the same information must be entered in different formats for each marketplace. Additionally, each platform regularly updates its requirements, forcing continuous data adaptation.

Online store with wide assortment

This is the most challenging situation. The store buys from many distributors and wholesalers, receiving only basic data - often in the form of a simple product description. There's no direct access to manufacturers or their official materials. Result? The necessity of manually extracting attributes from descriptions, which with thousands of products monthly can generate huge costs and slows down offer development.

Each product has its unique set of important technical characteristics. Different parameters will be crucial for a drill (power, chuck, power supply type) versus paint (color, capacity, surface type). This is precisely why sales platforms and industry standards introduce the concept of product classifications - precisely defined categories, where each has its own set of dedicated attributes. This isn't about simple product categorization (like "Electronics" or "Home and Garden"), but very detailed classifications like "Drill-Driver", "Exterior Paint" or "Installation Cable".

Each classification defines not only the list of required attributes but also their exact type:

• Text values (e.g., manufacturer name)
• Numbers (e.g., number of LED lights)
• Numbers with units (e.g., power in watts)
• Numerical ranges (e.g., allowable operating temperature)
• Boolean values (e.g., whether the product is waterproof)
• Dictionaries - predefined lists of allowed values

Dictionaries are simultaneously the most valuable and most problematic. On one hand, they form the basis for building advanced filters in online stores and creating standardized product names. On the other hand, they are the source of the biggest problems when integrating supplier data.

Let's imagine an example of the "Power Supply Type" attribute in power tool classification. A marketplace might define dictionary values as: ["Mains Powered", "Battery Powered"], while the manufacturer uses different terms: "230V", "battery operated", "mains powered", "cordless". The traditional approach requires tedious mapping of each possible supplier value to the values required by the sales platform. With thousands of attributes and dozens of suppliers, this creates a complicated matrix of connections that needs constant updating.

getName.ai was created to radically solve this problem. Instead of creating another value mapping tool, the system uses large language models (ChatGPT, Claude) to understand text meaning. Regardless of how a supplier describes the power supply type, getName.ai can match this information to the appropriate dictionary value required by the sales platform.

getName.ai is a SaaS solution that provides an API for automatic product attribute recognition. The system operates based on three key pieces of information:

• The classification or standard according to which attributes should be recognized
• The product category/class of the analyzed item
• Input data in the form of a product description or URL

This simplicity allows getName.ai to function like a virtual product specialist - always available, precise, and reliable. The system continuously analyzes received data, extracting valuable technical information.

Supported product classifications

Currently, getName.ai supports attributes according to Allegro and Ceneo platform requirements, as well as the international ETIM standard in versions 7, 8, and 9, which is widely used in the electrotechnical industry. Importantly, getName.ai can also work with clients' own product classifications - those created and developed for specific company or industry needs. Simply provide the data model of such classification, and the system will learn to recognize attributes according to its requirements.

The list of supported standards will be continuously expanded based on market needs and client suggestions. If your company uses other product classifications or industry standards, we encourage you to contact us - getName.ai was designed with flexibility and quick adaptation to new requirements in mind.

Input data types

The getName.ai system was designed with simplicity of use in mind. Currently, it offers two methods of providing product data:

Text description

You can provide any text describing the product to the system, regardless of the language it's written in. The application can handle various forms of description which may include:

• Complete product name
• Product description
• Specification attributes in tabular form

The application output will be a JSON with the following data fragment:

Product URL

The second method is even simpler - you just need to provide the product subpage URL. The system will automatically:

• Download content from the specified page
• Extract product information
• Send the obtained description to AI for extracting attributes and their values

The application returns results in JSON format containing attribute identifiers and values consistent with the chosen ETIM classification. The system deliberately doesn't return attribute labels (names) and dictionary values for two reasons:

• Technical - labels aren't needed in the integration process because each target system (marketplace, PIM, or own store) already has its own official translations for used identifiers
• Legal - label translations in the ETIM standard are protected by copyright. Therefore, getName.ai operates solely on identifiers, leaving the matter of displaying labels to systems that have appropriate licenses for their use.

It's important to clearly define the system's current limitations:

Doesn't Suggest New Attributes

getName.ai operates strictly within defined product classifications. The system doesn't propose new attributes or additional dictionary values, even if it detects such information in the description. This is a conscious limitation, aligned with the idea of product data standardization.

Doesn't Automatically Classify Products

Precise management of technical attributes is becoming an increasingly crucial element of success in online retail. getName.ai was created to help meet these requirements by automating one of the most time-consuming processes in e-commerce.

Contact us to learn more about how getName.ai can be utilized in your business.