Semantic Search and
Intent-Based Systems

Semantic search understands the intent behind a query instead of matching exact words. Text is turned into vectors (embeddings), and the system finds the nearest matches in a meaning space, so a search for "something to sit on" also returns "chair" and "stool". With Meilisearch, written in Rust, hybrid vector search runs in under 50 milliseconds, straight from disk.

For a long time, digital search has been limited to pure text matching. When a user enters a search term, the system has traditionally only looked for identical character strings in a database. This lexical model is fast, but lacks understanding. It falls short when the user and the online store use different words for the same thing (synonymy), or when a word can mean different things depending on context.

The customer gets an empty page even though the product is in stock, because they used the "wrong" word. For online retailers, those zero results mean millions in lost revenue every year.

Semantic search and vector databases move us from matching characters to understanding what the user means. The technology used to be reserved for the giants; today it comes off the shelf and powers everything from precise chatbots to intelligent recommendation engines. In 2026, that is the difference between a passive archive and a helpful sales assistant.

This article is a technical and strategic overview of how modern search technology improves the user experience. We use Meilisearch as an example to show how the architecture behind vector search and hybrid ranking works in practice, and how you can build next-generation search solutions with frameworks like Laravel.

To see what semantic search fixes, start with the limits of the traditional way of searching.

Traditional search engines rely on an inverted index. If you search for "car", the engine scans the database for this exact letter combination. If a document only contains the word "vehicle", it is practically invisible to the search engine. This is called the lexical gap.

In an online store this costs sales. A customer searching for "warm winter jacket" might get no results for a "lined parka", even though that's exactly what they're looking for. Lexical search requires the customer to guess the exact words the product text uses. Semantic search removes this guessing game by understanding the meaning behind the words.

The solution is vectors (embeddings). Machine learning models such as OpenAI's text-embedding-3 convert text, images, or audio into long sequences of numbers, the vectors.

Think of an enormous multi-dimensional map. On it, "cat" and "dog" sit close together because they share conceptual properties (pets, four legs, fur). "Car" ends up in a different "neighborhood" altogether.

When a user searches, the query becomes a coordinate in this space. The system then looks for the nearest points ("Nearest Neighbor Search"). That is how a search for "something to sit on" returns "chair", "sofa", and "stool" without anyone defining those synonyms by hand.

Meilisearch has established itself as a specialized challenger to market leader Elasticsearch. Elasticsearch (based on Java) is the preferred tool for heavy log analysis and big data. Meilisearch (written in Rust) is built for one thing: fast, relevant search results delivered straight to the end user.

Rust and predictable performance: By avoiding the Java Virtual Machine (JVM) and its memory management pauses ("garbage collection"), Meilisearch delivers stable response times, often under 50 milliseconds. That stability is what makes "search-as-you-type" work without visible lag.

Hybrid search in one core: Meilisearch integrates vector search directly into its core, so you don't run two separate systems, one for text and one for vectors. The result is a hybrid model with the precision of traditional keywords and the contextual understanding of AI.

Cost-effective scaling (Disk-based HNSW): Many vector databases require the entire index to be loaded into expensive memory (RAM) to be fast enough. Meilisearch uses a custom implementation of the HNSW algorithm (called Hannoy) built to run straight from disk. That cuts operational costs, since you don't need enormous amounts of RAM for large datasets.

Meilisearch recently completed a full rebuild of its vector engine, moving from a tree-based structure (Arroy) to a graph-based structure (Hannoy). The numbers below illustrate the effect of this modernization:

The reduction in search latency from 227 ms to 13 ms is the difference between an application that feels sluggish and one that feels instantaneous.

The weakness of pure vector search is that it can become too smart. If a customer searches for a specific part number, for example "XJ-900", a pure vector search might return "XJ-800" because the products are semantically similar. In such cases precision matters more than understanding, and that is the job of hybrid search.

Meilisearch combines results from traditional keyword search (BM25) and modern vector search. The challenge is that these systems speak different languages: BM25 might give a relevance score of 45.0, while vector search gives a score between 0 and 1 (e.g., 0.9).

To reconcile these, Meilisearch uses a normalization technique (affine transformation) that makes the scores comparable before merging them. That is a step up from many competitors, who use a simpler method called Reciprocal Rank Fusion (RRF). While RRF only looks at what order the results come in, Meilisearch preserves information about how good the match actually was. This produces a more accurate final list.

A unique advantage of Meilisearch is the semanticRatio parameter. In practice it is a "slider" between 0.0 (pure text search) and 1.0 (pure AI search), so you can tune the strategy per use case:

• For online stores (e.g., 0.3): Here, product names and specifications are often most important. A low ratio ensures we match exactly on search terms, while still catching synonyms and typos via the vectors.
• For support bots (e.g., 0.9): Here, users often ask vague questions ("how do I fix the thingy?"). A high ratio lets the AI take control to understand context and intent, rather than looking for verbatim matches. This is something we like to A/B test to find the optimal setting.

Technology has little value until it solves a problem. Here are three concrete use cases where intent-based search delivers measurable gains in the form of increased sales, lower costs, and better customer loyalty.

Data from Baymard Institute shows that 42% of major online stores don't support thematic searches (e.g., "spring jacket for men" instead of just "jacket"). When we know that about 30% of visitors use the search field, and that these have 2–3 times higher purchase intent than those who just browse, it pays to meet them the right way. A zero-results page is an invitation to visit a competitor.

• Bookshop.org: The online bookstore found that users often searched for concepts rather than titles (e.g., "blue book about a wizard"). Traditional search returned no results. By switching to Meilisearch with semantic search, they increased the conversion rate from search by 43%. The system connected the vague description to Harry Potter through context, not just keywords.
• Etsy: Etsy implemented their own semantic search engine to handle aesthetic searches like "boho chic decor". The result was a significant increase in conversion and sales, generating hundreds of millions of dollars in additional annual revenue. Customers simply found products they didn't know the name of.
• Walmart: When an item is out of stock, Walmart uses vector search to understand what the product is, so they can suggest a relevant replacement. Instead of losing the sale, they show another item that's close in "semantic space". This has drastically reduced lost revenue during stock shortages.

Retrieval-Augmented Generation (RAG) is changing customer service. By letting a language model (ChatGPT, Gemini, Claude) search the company's own data before responding, you eliminate "hallucinations" and get precise answers. That moves AI from a simple "chatbot" to a team member.

• Klarna: Klarna is perhaps the most well-known example of RAG in 2024. Their AI assistant handled 2.3 million conversations (two-thirds of all inquiries) in the first month. It now does the work equivalent to 700 full-time employees, resolves cases in 2 minutes (down from 11 minutes), and is expected to improve the bottom line by $40 million annually.
• Morningstar: The financial services company Morningstar developed "Mo", an internal assistant for analysts. Instead of spending hours searching through thousands of PDF reports, analysts can ask "What is the risk analysis for company X?". The system retrieves relevant sections from internal databases and summarizes the answer. That frees hours for actual analysis rather than information retrieval.

Traditional recommendation engines ("Customers who bought this also bought...") depend on history. They work poorly on new products, a problem known as "Cold Start". Vector search solves this by analyzing the product's content (image, text, attributes) rather than user behavior.

• Spotify: Although Spotify uses a lot of user data, their recommendations are also based on semantic understanding of audio files and lyrics. This allows them to recommend a brand new song from an unknown artist to you, simply because the song's profile (tempo, mood, genre) resembles what you like. This keeps users in the app longer.
• IKEA: IKEA uses visual AI and vector search to recommend products that match stylistically. If you're looking at a sofa, the system can find pillows and rugs that match the sofa's color and texture. Such recommendations often increase average order value by 10–30% because customers are inspired to buy a complete style.
• Meilisearch: Technically, this is solved via the /similar endpoint. An online store can immediately show "Similar products" for an item that was added 5 minutes ago. This ensures new products get exposure from day one, without waiting weeks for click data.

PXL knows Laravel well, and the integration there is straightforward. Through the Laravel Scout package we control the search logic in code, so we can tailor a hybrid search to your needs down to the smallest detail.

Example: Enabling Hybrid Search

Setting up hybrid search in Laravel is quick, but you have to go a step beyond the standard setup to reach the vector side. Here we tell the engine to weight keywords and semantic understanding equally (0.5):

For online stores on WooCommerce, the job is to get search out of the database. Standard WordPress search loads the SQL server heavily and often returns irrelevant results.

Instead of the built-in search, we run Meilisearch as a dedicated search engine alongside the online store and talk to its API directly, which takes the load off the server. That gives an immediate speed boost plus spelling correction ("did you mean...?") and semantic understanding, functionality that usually needs much heavier enterprise platforms.

In larger architectures, we often run Meilisearch as an isolated service. This means we can have a mobile app (iOS/Android) and a website talking to the same search engine. Since Meilisearch is language-agnostic, we can build fast frontends in JavaScript (Vue, React, Next.js) while the backend handles the data flow.

Regardless of platform, data security is critical, especially in B2B solutions (SaaS) where Customer A must never be able to search Customer B's data.

Meilisearch solves this with Tenant Tokens. Instead of giving the frontend application full access, the backend generates a signed key containing hardcoded rules (e.g., tenant_id = 1). Meilisearch then guarantees, at the engine level, that the search only occurs within that customer's data. This eliminates the risk of data leakage through the search field.

The search field is no longer limited to text. Meilisearch supports multimodal search, which blurs the line between text and images.

This solves a classic problem in e-commerce: How do you describe a style or aesthetic with words?

• How it works: A user uploads an image of a chair they like, but which might be too expensive or the wrong color.
• Vectorization: The system uses an AI model (like CLIP) to "see" the image and convert the visual expression into a vector.
• Result: Meilisearch immediately finds products in your database that have the same shape, style, or expression as the image, regardless of whether the product name is similar.

This opens up new user experiences in fashion, interior design, and real estate, where the visual often outweighs the technical.

Semantic search has gone from a technological curiosity to a requirement for modern digital solutions. Understanding what the user actually means, across typos, synonyms, and languages, is often what converts a visitor into a customer.

With tools like Meilisearch, a business can offer search on par with the international giants at a fraction of the price and complexity. Smarter online stores, precise chatbots, or automated recommendations: intent-based search turns user needs into business value.

PXL helps you get there. We combine long experience in system architecture and modern web development with AI and search technology. We adapt the tools to your needs, whatever platform you run today, and deliver solutions that work technically and understand your customers.