Lucidworks adds AI-driven data enrichment to tackle weak search performance

Lucidworks has introduced a new Data Enrichment feature designed to automatically improve product data using multimodal generative artificial intelligence, as retailers and e-commerce groups look to fix persistent problems with on-site search.

The company said incomplete or inconsistent product information is one of the main causes of poor search performance.

According to Lucidworks, as many as 30% of searches fail because key attributes are missing, keywords are weak or products sit in unclear or overlapping categories.

When search systems cannot clearly understand what a product is, they struggle to return relevant results, even if the item exists in the catalogue.

The new feature is intended to address this by analysing both images and text associated with a product, then generating richer and more consistent data at scale. Lucidworks said the system produces clearer product categories, higher-quality keywords, relevant synonyms and more detailed descriptions, all designed to make products easier for search engines to understand and retrieve.

For non-technical users, multimodal AI refers to models that can process more than one type of input at the same time. In this case, the system looks at product images as well as written descriptions, allowing it to infer meaning from visual cues, such as shape, colour or style, alongside language. That combined understanding is used to resolve ambiguity and standardise how products are described.

Lucidworks said the feature generates “search-ready” data automatically, without the need for manual tagging or major changes to existing systems. This means companies do not need to move platforms or invest in large data-labelling projects, which are often costly and slow to scale.

“Search and AI are only as good as the data behind them,” said Rishi Setia, senior product manager of AI at Lucidworks. He said the goal is to improve the underlying quality of product information so search and recommendation systems can perform more effectively.

The company said early client results show a measurable business impact. Reported outcomes include three times more useful, searchable data, significant improvements in recall, which measures how many relevant products are returned in search results, and an 8.66% lift in conversion rates. For one global retailer, Lucidworks said this translated into more than $25m in annualised revenue impact.

Related reading

• Automation Anywhere unveils agentic solutions with OpenAI
• OpenAI: The capability overhang and AI for self-empowerment
• BingX signs multi-year partnership with Scuderia Ferrari HP

Technically, the enrichment process runs offline and at scale, meaning it can process large catalogues without affecting live site performance. The output is delivered ready to be indexed and works natively with Lucidworks’ Neural Hybrid Search and Commerce Studio, feeding improved data into ranking and relevance models.

Lucidworks said the Data Enrichment feature is available immediately for existing customers, positioning it as a practical upgrade for companies seeking to improve search and discovery without overhauling their technology stack.

The Recap

• Lucidworks introduces Data Enrichment for eCommerce product data.
• Addresses missing attributes that cause up to 30% search failures.
• Available immediately for existing Lucidworks customers to deploy.