Here's a recap of everything that happened this month!

## 🔔 Reminder to upgrade

[![meilisearch-1.14.png](https://unable-actionable-car.media.strapiapp.com/meilisearch_1_14_166a7ff64e.png)](https://www.meilisearch.com/blog/meilisearch-1-14)

Don't forget to upgrade to Meilisearch 1.14 to benefit from improved AI search with composite embedders and embedding cache. v1.15 is coming soon!

[Read the release notes](https://www.meilisearch.com/blog/meilisearch-1-14)

## 📰 In the news

### 💬 Talk to Meilisearch with Claude desktop

Connect your LLM to Meilisearch using our MCP server. 

[Read the guide](https://www.meilisearch.com/docs/guides/ai/mcp)

### 🚀 How we're making AI work at Meilisearch

How we turned scattered AI usage into systematic success for our teams.

[Read the blog post](https://www.meilisearch.com/blog/making-ai-work)

### 🌱 Six years working with the open-source community

Our Head of Engineering Clémentine reflects on our experience working in OSS.

[Read the blog post](https://blog.curqui.com/six-years-working-with-the-open-source-community)

### 📣 How are CTOs spending their AI budgets?

You can help us understand you better by taking this survey!

[Take the survey](https://s.surveyplanet.com/befuwdt7)

### 📣 How long does it take to integrate search?

You can help us understand you better by taking this survey!

[Take the survey](https://s.surveyplanet.com/sjyv6prl)

## 🤗 Made with Meilisearch

HuggingFace's [Daily Papers](https://huggingface.co/papers) uses Meilisearch hybrid search to search across hand-picked machine learning papers published over the last two years.

[![huggingface-papers.gif](https://unable-actionable-car.media.strapiapp.com/huggingface_papers_fceeaf233b.gif)](https://huggingface.co/papers)

_We’re always looking to promote new projects from the community. Post yours in the [#built-with-meilisearch](https://dub.sh/meili-discord?utm_campaign=newsletter&utm_source=blog&utm_medium=may-25-updates) Discord channel!_

---

Receive this updates in your inbox by subscribing to our [monthly newsletter](https://dub.sh/meili-newsletter?utm_campaign=newsletter&utm_source=blog&utm_medium=may-25-updates). To learn more about Meilisearch's future and help shape it, take a look at our [roadmap](https://roadmap.meilisearch.com/?utm_campaign=newsletter&utm_source=blog&utm_medium=may-25-updates) and participate in our [Product Discussions](https://github.com/orgs/meilisearch/discussions).

For anything else, join our developers community on [Discord](https://dub.sh/meili-discord?utm_campaign=newsletter&utm_source=blog&utm_medium=may-25-updates).

Meilisearch May updates

To build a fast and reliable JavaScript search engine for your website from scratch, you’ll need to

* Set up your environment
* Prepare your data
* Create an index
* Implement the search functionality

Less complicated ways to handle the task are also available. For example, Meilisearch, a lightweight, open-source, and full-text search engine solution, can provide users with immediate search results, typo tolerance, and easy installation.

Whether you build a custom solution or use solutions like Meilisearch will depend on your specific use case and unique needs.  Either way, this guide will walk you through the steps to create a functional JavaScript search engine exactly how you want it.

## 1. Set up the development environment

The search engine development process starts with creating a development environment where your JS code will run. To do this, you must establish a Node.js environment and install the necessary dependencies for the search engine logic.

The JS runtime is essential for executing code outside the web browser, such as server-side operations like indexing data or querying a search engine.

Furthermore, it is important to have a package manager as a medium for library installations; a typical example is npm. For a basic setup, Node.js should be present to start a new project. Then it's just typing a command to the terminal and done! This creates a package.json file to manage dependencies.

```
mkdir search-engine
cd search-engine
npm init -y
```



This sets the stage for later adding libraries or tools, such as Meilisearch. A proper environment ensures your code runs smoothly across the development and production phases.

Next, let’s prepare the data for searching.

## 2. Prepare and structure the data

Structuring data is the first item on your to-do list before you can search for anything. For this, you need to collect and format the content to make it searchable. This could include blog posts, FAQs, knowledge bases, product descriptions, or user profiles.

Data should be in a consistent format, typically JSON, to make indexing straightforward. Imagine that you are creating a search engine for a blog. Your data may contain titles, content, and meta tags that must be in the same format. 

Clean and normalize the data to remove inconsistencies, like different date formats or missing fields, which can affect search accuracy. This step must be done properly because well-structured data directly impacts the quality of search results — as the saying goes, 'garbage in, garbage out.'


```javascript
const sampleData = [
  { id: 1, title: "JavaScript Basics", content: "Learn the fundamentals...", tags: ["JS", "coding"] },
  { id: 2, title: "Advanced Node.js", content: "Deep dive into Node...", tags: ["Node", "JS"] }
];
```



With your data structured and ready, it’s time to create an index.


## 3. Create the index

Simply put, indexing is the process of arranging your data to make it searchable. Although the data we have at this point is structured, we need to store it properly for quick retrieval based on search queries. That’s where an index comes in: it is like a database optimized for speedy lookups. 

In a custom JavaScript search engine, you might build a simple inverted index, mapping words 1:1 to their locations in your dataset. This would entail tokenizing text (splitting it into individual words), removing stop words (like "the" or "and"), and storing the results obtained.

Even though this might be effective for small datasets, it can get tricky when the complexity of the dataset grows, resulting in issues with performance and scalability.


```javascript
function createIndex(data) {
  const index = {};
  data.forEach((doc, docId) => {
    const words = doc.content.toLowerCase().split(/\W+/);
    words.forEach(word => {
      if (!index[word]) index[word] = [];
      index[word].push(docId);
    });
  });
  return index;
}
```


Now, let’s implement the search functionality, which we’re here for.


## 4. Implement the search functionality

The final step in setting up a JavaScript search engine brings it to life by enabling users to query your search index and retrieve relevant results. The search function will analyze the user’s search term, parse the index we generated in the previous step for matches, and rank search results based on relevance and accuracy.

For a better user experience, consider including features such as partial matching or [ranking by relevance and precision](https://www.meilisearch.com/blog/filter-scoring-guide). However, this can be a drain on computational power, so algorithm optimization may help you increase performance.

```javascript
function search(query, index, data) {
  const queryWords = query.toLowerCase().split(/\W+/);
  const results = new Set();
  queryWords.forEach(word => {
    if (index[word]) {
      index[word].forEach(docId => results.add(data[docId]));
    }
  });
  return Array.from(results);
}
```



Now that we know how to create a JS search engine from scratch, let’s explore how Meilisearch simplifies this process.

## Building a JavaScript search engine for your website with Meilisearch

Meilisearch is an open-source, lightning-fast search solution designed for error-free [integration with JavaScript applications](https://www.meilisearch.com/with/javascript). It’s both developer—and user-friendly, with features like [typo tolerance](https://www.meilisearch.com/blog/typo-tolerance), [faceted search](https://www.meilisearch.com/blog/nested-hierarchical-facets-guide), and [federated search](https://www.meilisearch.com/blog/what-is-federated-search) for searching across multiple indices. These features, and more, make it ideal for building optimal search experiences for web pages, apps, or e-commerce platforms.

Unlike building from scratch, though, Meilisearch handles the heavy lifting of indexing and searching for you. Significant benefits of using Meilisearch include:

* **Speed:** Delivers almost instant search results, even with large datasets.
* **Typo tolerance:** Handles misspellings and typos gracefully with [fuzzy search](https://www.meilisearch.com/blog/fuzzy-search), letting users get to their intended results faster.
* **Easy integration:** Works seamlessly with JavaScript via its SDK.
* **Customizable:** Allows fine-tuning of search relevance and filters.

<Callout intent="tip">

#### Improve user experience with the power of Meilisearch + JS

Our tool is built for speed, simplicity, and performance. Pair that with the power of JavaScript, and you have a search engine that outperforms any other tool-language combination.

[Try Meilisearch today!](https://www.meilisearch.com/cloud)

</Callout>

Here’s how you can use Meilisearch to build a JavaScript search engine for your website.

### Step 1: Install Meilisearch and the JavaScript SDK

Start by setting up a Meilisearch instance locally (e.g., at http://127.0.0.1:7700) or via a cloud provider. Then, install the core Meilisearch JavaScript SDK for server-side operations and, optionally, the instant search package for front-end integration.

```
npm i meilisearch
npm install @meilisearch/instant-meilisearch
```



The `meilisearch `package provides the core functionality for indexing and querying, while `@meilisearch/instant-meilisearch` simplifies front-end search interfaces for instant search capabilities.


### Step 2: Initialize the client and index your data

Create a MeiliSearch client to connect to your instance and index your data. As mentioned, an index is a container for your searchable documents, and the client facilitates all interactions with the Meilisearch server. Use the SDK to send your JSON-formatted data to the Meilisearch server.


```javascript
import { MeiliSearch } from "meilisearch";

const client = new MeiliSearch({
  host: "http://127.0.0.1:7700",
  apiKey: "masterKey",
});

const data = [
  { id: 1, title: "JavaScript Basics", content: "Learn the fundamentals..." },
  { id: 2, title: "Advanced Node.js", content: "Deep dive into Node..." }
];

async function addDocuments() {
  try {
    const index = client.index("documents");
    await index.addDocuments(data);
    console.log("Documents added successfully");
  } catch (error) {
    console.error("Error adding documents:", error);
  }
}
addDocuments();
```



This code initializes the client with your Meilisearch instance’s host and API key, then adds JSON documents to an index named `documents`. It ensures your data is indexed and ready for searching. For this to work properly, ensure your Meilisearch server is running and accessible at the specified host.


### Step 3: Implement client-side search

Integrate Meilisearch with React using `@meilisearch/instant-meilisearch` and `react-instantsearch-dom` to create an interactive search interface. This package integrates with your front-end, providing a pre-built search bar and results display while connecting to the same Meilisearch instance for real-time search.


```javascript
import { instantMeiliSearch } from "@meilisearch/instant-meilisearch";
import { SearchBox, Hits } from "react-instantsearch-dom";

const searchClient = instantMeiliSearch("http://127.0.0.1:7700", "masterKey");

function App() {
  return (
    <div>
      <SearchBox />
      <Hits />
    </div>
  );
}
```



Meilisearch’s speed, simplicity of setup, and robust performance make it a powerful choice for adding search functionality to your website. 

## What are some examples of JavaScript search engine uses?

Search engines power various applications, empowering users to find the knowledge they seek by making content more discoverable. Meilisearch, in particular, excels in several use cases due to its flexibility and performance.

**App search**: Mobile, desktop, or web applications, such as note-taking or task management tools, utilize search functionalities to help users quickly find specific entries. Meilisearch’s typo tolerance ensures users can find accurate results even with minor misspellings.

**Site search**: Websites like blogs or documentation portals rely on search to navigate vast content libraries. Meilisearch powers fast, relevant site search for platforms like intra-team technical docs or wikis.

**eCommerce search**: Online stores and storefronts use search functionalities to help customers find products by name, category, or description. With capabilities like filtering and faceting, Meilisearch allows users to further refine searches by the object’s type, price, brand, or customer ratings.

**Customer support portals**: Corporations integrate search functionality into customer help centers to increase consumer independence, allowing users to find articles or FAQs independently. Meilisearch’s speed and [autocomplete functionality](https://github.com/meilisearch/meilisearch-js-plugins/) ensure a quick and easy problem-solving path.

These use cases illustrate the versatility and applicability of JavaScript search engines, especially when powered by tools like Meilisearch.

### Which languages besides JS can be used to build a search engine?

Aside from JavaScript, other common programming languages you can use to build a search engine are:

* **PHP**: Used for interactive and dynamic web development, PHP can integrate seamlessly with search tools like Meiliseach. Learn more about building a [search engine in PHP](https://www.meilisearch.com/blog/php-search-engine) through tutorials on integrating such tools.
* **Python**: Offering simplicity and adaptable libraries, Python is a top choice for building a custom or tool-powered search engine. Explore scalable search solutions by learning how to build a [search engine in Python](https://www.meilisearch.com/blog/python-search-engine).
* **Golang**: With unmatched performance and concurrency, Go is ideal for high-speed search engines, particularly for backend optimization. Check out how to build a [search engine using Golang for efficient and lean systems](https://www.meilisearch.com/blog/golang-search-engine).

Others like HTML, CSS, and SQL might also work for you. Each language brings specific strengths that are better suited for specific use cases. Thus, choose your language based on your project’s needs.

### Can I build a JavaScript search engine for free?

Yes, you can build a JavaScript search engine for free, that is, without licensing costs. Open-source tools like Meilisearch have free, self-hosted options, and you can use libraries like the [Meilisearch JavaScript SDK](https://meilisearch.github.io/meilisearch-js/) at no cost.

However, you may incur costs using cloud-hosted Meilisearch plans or making extensive API calls in production. Furthermore, advanced features like AI analytics also come at a price. Hence, to avoid surprises, keep your options sorted by budgeting for cloud services and advanced features.

### Are there GitHub repositories for JavaScript search engines?

Several repositories exist for JavaScript search engines, including the Meilisearch JavaScript SDK: [/meilisearch-js](https://github.com/meilisearch/meilisearch-js).

The Meilisearch repo provides everything you need to integrate Meilisearch into JavaScript projects, including client libraries and code examples. It’s actively maintained and well-documented, so you and developers can use it for quick setup.

### Can you make a search engine in React?

Yes, you can build a search engine in React. We recommend that you do.

React’s component-based architecture pairs well with Meilisearch’s Instant Meilisearch package, which offers pre-built modular components like search bars and hit lists. This setup enables you to create a responsive and interactive [search experience in React](https://www.meilisearch.com/blog/react-search-engine) with minimal effort.


Building a JavaScript Search Engine: Tutorial, Examples & More

You can make a search engine in Python using a combination of data structures, algorithms, and libraries to index, rank, and retrieve information based on your search query input.

A Python search engine works based on these key building steps:

1. Collect data & preprocessing
2. Document creation and indexation
3. Add a search system
4. Rank results.

A search engine built with Python can be used by small-scale and large-scale enterprises that opt for open-source solutions that offer scalability and flexibility.

Due to its easy customization, these search engines can be applied in many applications, such as e-commerce, research, marketplaces, enterprise search, and more.

Let’s examine in more detail the different steps to building a simple search engine from scratch with Python.

## 1. Collect data & preprocessing

Data collection can be done in several ways. You may need to web scrape content from HTML webpages with packages such as `beautifulsoup`, or directly connect your script to Google Sheets using `gspread`.

Perhaps you already have a database and only need to use a PostgreSQL Python connector to access it and query the data.

The Python code for data collection can vary a lot depending on your needs, but here is an example of how to use Beautiful Soup for web scraping. First, you need to install the package:

```
pip install beautifulsoup4
```

Here’s the code example:

```pyhton
import requests
from bs4 import BeautifulSoup

url = "https://example.com"
response = requests.get(url)
soup = BeautifulSoup(response.text, 'html.parser')

# Extract data
for item in soup.select('.item', limit=5): 
    title = item.find('h2').text
    link = item.find('a')['href']
    print(f"{title}: {link}")
```


Once the data is collected, it needs to be preprocessed. This step can occur before and after document indexation. Several preprocessing trials may be needed to ensure the dataset is correctly indexed and ready for optimal retrieval.

For instance, text sources that contain emojis, emails, and source links can be cleaned beforehand to avoid adding unnecessary information to the system. 

In paragraphs, punctuation and stop words can be removed, and the sentences can be converted to lowercase.

With Python, several packages can be utilised for data parsing, cleaning, and preprocessing. Let’s take a look at the NLTK library and use it to remove emojis, emails, and punctuation:

First, install the package:

```
pip install nltk
```


Now you can try the following script:


```python
import re
import string
from nltk.tokenize import word_tokenize

# Sample text with punctuation, emojis, and emails
text = """
Hello! 😊 This is a test email@example.com. 
Can you remove this? 👍 Also, check info@domain.org! 
"""

def clean_text(text):
    # Step 1: Remove emails
    text = re.sub(r'\S+@\S+', '', text)
    
    # Step 2: Remove emojis and symbols
    emoji_pattern = re.compile(
        "["
        "\U0001F600-\U0001F64F"  # Emoticons
        "\U0001F300-\U0001F5FF"  # Symbols & pictographs
        "\U0001F680-\U0001F6FF"  # Transport & map symbols
        "\U0001F700-\U0001F77F"  # Alchemical symbols
        "\U0001F780-\U0001F7FF"  # Geometric Shapes Extended
        "\U0001F800-\U0001F8FF"  # Supplemental Arrows-C
        "\U0001F900-\U0001F9FF"  # Supplemental Symbols and Pictographs
        "\U0001FA00-\U0001FA6F"  # Chess Symbols
        "\U0001FA70-\U0001FAFF"  # Symbols and Pictographs Extended-A
        "\U00002702-\U000027B0"  # Dingbats
        "\U000024C2-\U0001F251" 
        "]+", 
        flags=re.UNICODE
    )
    text = emoji_pattern.sub('', text)
    
    # Step 3: Remove punctuation
    text = text.translate(str.maketrans('', '', string.punctuation))
    
    # Step 4: Tokenize and rejoin (optional, removes extra whitespace)
    tokens = word_tokenize(text)
    cleaned_text = ' '.join(tokens)
    
    return cleaned_text

# Clean the text and make it lowercase
cleaned_text = clean_text(text).lower()
print("Original Text:\n", text)
print("\nCleaned Text:\n", cleaned_text)
```


In some cases, Natural Language Processing (NLP) could be required. Take, for instance, a list of companies like “Impossible Foods” and Impossible Foods Co.” Both are the same companies, so you can use NLP to vectorize the words and unify the results in one single term based on their cosine similarity.

Sometimes, excessive preprocessing can be a problem, resulting in information loss. Therefore, the best approach is to start with simple steps. 

Once the documents are indexed, they can be updated with new preprocessed information.


## 2. Document creation and indexation

Documents are units of information (e.g., text, JSON, images, or structured/unstructured data) processed and stored on an index.

This step involves gathering the data sources you want to index in your vector database and converting them to documents.

For instance, if you use a JSON format as input, you can use LangChain Python’s framework to convert it to a list of documents directly.

First, you need to install the LangChain package in your system:


```
pip install langchain 
```


Secondly, you need to import the JSONLoader class and apply the following script:


```python
from langchain_community.document_loaders import JSONLoader

loader = JSONLoader(
    file_path='./my_data.json',
    jq_schema='.messages[].content',
    text_content=False)

data = loader.load()
```


The output should be a list of documents that look like this:


```python
[Document(page_content='Bye!', 'seq_num': 1}), Document(page_content='Hello', 'seq_num': 2}),
Document(page_content='See you later', 'seq_num': 3})]
```


Once the documents are created, they can be added to a vector database such as  [Chroma db](https://docs.trychroma.com/docs/overview/getting-started). 
 
Meilisearch’s [Python SDK](https://github.com/meilisearch/meilisearch-python) simplifies the process by eliminating the need to convert your source data into documents or search for a database solution. You can directly add the JSON or CSV files to an index on Meilisearch’s vector database.

Like LangChain, you first need to install the Meilisearch package on your machine:


```
pip install meilisearch
```


Then, you need to create an index with this simple command:


```python
client.create_index('books', {'primaryKey': 'id'})
```


To add the documents to the index, you can use the JSON format like this:

```json
client.index('movies').add_documents([{
  'id': 287947,
  'title': 'Super Gut',
  'author': 'Dr. William Davis',
}])
```


With the same package, you can also update the documents, apply filters, and delete them by simply changing the function:


```json
# apply filter
client.index('books').update_filterable_attributes(['author'])

# update documents
client.index('books').update_documents(<list_of_documents>)

# delete documents
client.index('books').delete_all_documents()
```


With Meilisearch's Python SDK, multiple indexes can be added, and all functions are easy to follow and implement.

Some examples are available on our [GitHub repository;](https://github.com/meilisearch/meilisearch-python/tree/main) alternatively, you can also refer to the [API documentation](https://www.meilisearch.com/docs/reference/api/overview) for more information.


## 3. Add a search system

If you’re using the LangChain approach with a custom vector database, you must embed your documents with Deep Learning (DL) algorithms. This creates a vector representation of your data, allowing for vector search, hybrid search, semantic search, and more.

Several embedding models are available on Hugging Face and through the OpenAI API.

Let’s, for instance, use the OpenAI embedding model with LangChain and Chroma as a vector database. You first need to install these packages:


```
pip install langchain-chroma
pip install langchain-openai
```


Export your OpenAI key and add the following:


```python
from langchain_chroma import Chroma
from langchain_openai import OpenAIEmbeddings

db = Chroma.from_documents(documents, OpenAIEmbeddings())
```


The command above embeds the documents with the OpenAIEmbeddings class and creates an index in the Chroma database. Now you can query the `db` instance:


```python
query = "Find a book about Nutrition"
docs = db.similarity_search(query)
print(docs[0].page_content)
```


All the above steps can be converted into one with [Meilisearch’s Python SDK](https://www.meilisearch.com/docs/reference/api/search). 

There’s no need to add embeddings or to find a package for your vector database. All you need is to search directly on the index previously created with the following function:


```
client.index('books').search('Find a book about Nutrition')
```


But it does not stop here, you can add filters like this:


```python
client.index('books').search('nutrition', {
  'filter': ['author = "Dr. William Davis"']
})
```


Or create a [faceted search](https://www.meilisearch.com/docs/reference/api/settings#faceting):


```python
client.index('movie_ratings').search('nutrition', {
  'facets': ['authors', 'rating']})
```


You can experiment with the API using other search options, such as specifying the number of documents to retrieve, [querying by locale](https://www.meilisearch.com/docs/reference/api/search#hybrid-search), or implementing hybrid search. 


## 4. Rank results

Ranking results can take complex Machine Learning (ML) algorithms, which are usually integrated in the Python libraries, so the good news is that you don’t need to create them from scratch.

For instance, Chroma uses an Approximate Nearest Neighbor (ANN) algorithm called Hierarchical Navigable Small World (HNSH) to find similar documents.

If you want to get their score and order them, you can run the following:


```python
results = docs.similarity_search_with_score(query="Find a book about Nutrition")
for doc, score in results:
    print(f"{score}: {doc.page_content}"}
```


However, this [documentation](https://api.python.langchain.com/en/latest/chroma/vectorstores/langchain_chroma.vectorstores.Chroma.html) has quite limited ranking results. A better approach is to use Meilisearch ranking rules, which are more straightforward.

By default, these are the ranking rules that can be tweaked:

* **"words”:** Sorts results by decreasing number of matched query terms
* **"typo":** Sorts results by increasing number of typos
* **"proximity":** Sorts results by increasing distance between matched query terms
* **"attribute":** Sorts results based on the attribute ranking order
* **"sort":** Sorts results based on parameters decided at query time
* **"exactness":** Sorts results based on the similarity of the matched words with the query words.

We can already see that the ranking mechanism can go beyond simply similarity. To rank the results, all you need to do is change the order of these terms in the query according to your needs:

```python
client.index('movies').update_ranking_rules([
    'typo',
    'words',
    'sort',
    'proximity',
    'attribute',
    'exactness',
    'release_date:asc',
    'rank:desc'
])
```


You can now search with a certain number of results (limit), and their relevance will be based on the updated ranking order.

This function is much simpler to implement and takes into account many other rules. Meilisearch streamlines the ranking process without the need to explore multiple libraries or create ranking algorithms from scratch.


## Can I make a search engine in Python for free?

You can use Python frameworks, such as LangChain, paired with open-source vector databases like Chroma. However, this strategy has limitations, such as limited ranking mechanisms, and requires additional steps and preprocessing.

For ease of implementation, more customization, and fast document retrieval, the best approach is to use [self-hosted Meilisearch](https://www.meilisearch.com/docs/learn/self_hosted/getting_started_with_self_hosted_meilisearch). You can use your machine or a VPS, which comes with a price.

You can also have free access to Meilisearch Cloud with a [14-day trial](https://www.meilisearch.com/cloud).


## What are the best open-source search engines for Python?

The best open-source search engines have comprehensive documentation and a large community of developers who share their issues and achievements.

Open-source search engine tools should also be easy to set up and provide examples for the community. These are the three main open-source platforms that support Python:


## Meilisearch

[Meilisearch](https://cloud.meilisearch.com/) is an open-source, lightning-fast search engine designed for developers and enterprises seeking to embed intuitive, scalable search experiences into applications through its RESTful API. 

Focusing on simplicity and performance, it provides advanced features such as typo tolerance and faceted search.

The [documentation](https://www.meilisearch.com/docs/reference/api/overview) is clear, easy to follow, and has examples. There’s a [Discord](https://discord.com/invite/meilisearch) group for developers to share their work or find solutions, and a well-structured [GitHub](https://github.com/meilisearch/meilisearch) repository.


## Qdrant

[Qdrant](https://qdrant.tech/) is an open-source vector database and vector search engine built in Rust. It efficiently handles similarity searches on high-dimensional vectors, making it ideal for tasks like recommendation systems, semantic search, and anomaly detection.

The Qdrant’s RESTful API supports multiple languages, including [Python](https://python-client.qdrant.tech/). The documentation is vast and can get overwhelming when you need to find the right steps to build a Python search engine. However, it also provides code examples, a [GitHub](https://github.com/qdrant/qdrant-client) repository, and a [Discord](https://discord.com/invite/tdtYvXjC4h) community.


## Elasticsearch

Elasticsearch is an open-source, distributed search and analytics engine with a scalable data store and vector database for various use cases.

The Python client for Elasticsearch is [well documented](https://www.elastic.co/docs/reference/elasticsearch/clients/python/getting-started) and provides the right tutorials to start building a search engine seamlessly.

They have a [GitHub](https://github.com/elastic/elasticsearch-py) repository where you can find examples, more information about the Python SDK, and a [ticket system](https://github.com/elastic/docs-content/issues/new?template=issue-report.yaml&link=https://www.elastic.co/docs/reference/elasticsearch/clients/python/getting-started&labels=source:web) on GitHub to address issues.

Elasticsearch also provides a [Python DSL](https://www.elastic.co/docs/reference/elasticsearch/clients/python/elasticsearch-dsl) module that aims to help with writing and running queries against Elasticsearch in a more convenient and idiomatic way.


## What programming languages besides Python are used to build search engines?

Python is not the only programming language that allows you to build AI-powered search engines. Some common programming languages you can use are:


* **JavaScript:** Learn how to [build a search engine in Javascript](https://github.com/meilisearch/meilisearch-js). 

* **PHP:** Learn how to [build a search engine in PHP](https://github.com/meilisearch/meilisearch-php).

* **Golang:** Learn how to [build a search engine with Golang](https://github.com/meilisearch/meilisearch-go).


## High-performance Python search engines with Meilisearch

While setting up the Python search engine, we realised the importance of having a unified package with its vector database, seamlessly embeds the documents, and provides tools for easy filtering and ranking results.

Using multiple libraries and Python frameworks can be overwhelming, leading to more problems than solutions. The frameworks can be limited in what you are allowed to do, which may force you to switch to another one or build from scratch, both of which can be time-consuming and resource-intensive.

Clarity, good documentation, and ease of use are the keys to building high-performing Python search engines. You don’t need to master several frameworks, libraries, and look for information in countless forums and YouTube videos.


How to Make a Search Engine in Python: Step-by-Step Tutorial

With React’s power and Meilisearch’s functionalities, you can build a search engine that feels fast, clean, and smart from the start for the best user experience.

In this guide, we’ll show you how to build a real-time search engine in React using Meilisearch’s InstantSearch integration. You’ll install the tools, set up your UI, and customize it to fit your app.

We’ll cover:

* Setting up Meilisearch in your React project
* Adding a search box and displaying results
* Customizing the experience with your own React components

Whether it’s a basic SPA or a full-stack project, this guide outlines the core steps for building a client-side search engine with [Meilisearch for React](https://www.meilisearch.com/with/react).

## 1. Set up Meilisearch in your React app

To connect your React app with Meilisearch, you must first install two dependencies: `react-instantsearch` for frontend components and `@meilisearch/instant-meilisearch` to bridge Meilisearch with InstantSearch.

This setup gives you a fully functional React search engine that updates as users type. You’ll be able to create a powerful search bar with minimal code, making it ideal for web apps that need fast performance and clean user interfaces.

```js
npm install react-instantsearch @meilisearch/instant-meilisearch
```


Once installed, import React and the required InstantSearch components:

```js
import 'instantsearch.css/themes/algolia.css'; // Import default InstantSearch styling
import React from 'react';
import { InstantSearch, SearchBox, Hits, Highlight } from 'react-instantsearch-dom';
import { instantMeiliSearch } from '@meilisearch/instant-meilisearch';

const searchClient = instantMeiliSearch(
  'https://your-meilisearch-url.com',
  'your-public-api-key'
);

const App = () => (
  <InstantSearch indexName="your-index" searchClient={searchClient}>
    <SearchBox />
    <Hits hitComponent={Hit} />
  </InstantSearch>
);

const Hit = ({ hit }) => (
  <div className="result-card">
    <h3>{hit.name}</h3>
    <p>{hit.category}</p>
  </div>
);

export default App;
```


This installs everything you need to get Meilisearch talking to your React frontend. Next, we’ll plug in your Meilisearch instance and show real search results.

## 2. Connect to Meilisearch and fetch search results

Now it’s time to connect your React frontend to your Meilisearch instance and return live results. This is where your React search engine comes to life.

Use the **instantMeiliSearch** function to link your frontend with the backend. You’ll need your Meilisearch host URL and public API key, which are available if you're running a local instance or using Meilisearch Cloud.

Here’s how to set it up:

```js
const searchClient = instantMeiliSearch(
  'https://your-meilisearch-url.com',
  'your-public-api-key'
)
```


Then wrap your app with the **InstantSearch** component:

```js
<InstantSearch indexName="your-index" searchClient={searchClient}>
  <SearchBox />
  <Hits hitComponent={Hit} />
</InstantSearch>
```


With this in place, you can fetch and display results as the user types. Let’s now see how to format those results using custom React components.


## 3. Customize search results with your own components

Once your Meilisearch setup is live, the next step is making your results native to your app. That means creating your UI components. Instead of sticking to default formatting, you can control how each result looks with a custom React function.

The **hitComponent** prop lets you define exactly what shows up. Here’s a simple example:


```js
const Hit = ({ hit }) => (
  <div className="result-card">
    <h3>{hit.name}</h3>
    <p>{hit.category}</p>
  </div>
)
```


From here, you can import images, pricing, ratings, or even quick links. You can add your own CSS, layer logic with React hooks, and use dynamic onChange events to update results in real time. You can customize your placeholder text for the search input, apply any className you want, and style it to match your frontend design.

You can even optimize your app.js using code splitting, especially when working with larger web applications. This level of control makes Meilisearch a strong fit for real-world use. You’re not locked into a preset theme; your search results can feel like they were always part of your app’s HTML page.

## 4. Add filters and facets for a better search UX

When users are searching, they don’t always know the exact term. That’s why filters matter. With Meilisearch, you can add faceted search to let users narrow down results by category, tags, brand, or anything else.

This is especially useful in web applications where fast and accurate filtering improves user experience. You can use InstantSearch widgets like **RefinementList** or **Menu**, and apply your own className to keep styling consistent across your layout.

First, make sure the fields you want to filter by are listed in **filterableAttributes** in your index settings.


```js
{
  "filterableAttributes": ["category", "brand", "price"]
}
```


Then use the **RefinementList** or **Menu** widgets from InstantSearch to build filter UIs in your React app. This improves the user experience by making your search engine feel more like Google Search or a modern e-commerce site.

[Faceted search](https://www.meilisearch.com/blog/nested-hierarchical-facets-guide) is fast and easy with Meilisearch and works great even on large datasets. With filters in place, your front-end search engine will be smart and lightning-quick.


## 5. Track user behavior with search analytics

Once your search engine is live, understanding how people use it helps you improve design and ranking.

Meilisearch Cloud includes built-in analytics that show you what users are searching for, what they’re not finding, and how your results perform across real web traffic. Whether using Meilisearch in a Next.js project or a full Node.js backend, the data flows cleanly into your stack.

You can monitor:

* Top search queries
* Searches returning no results
* User behavior patterns across your web pages

This gives you the clarity to improve content, tweak filters, and boost overall ranking relevance. These analytics offer a real edge for e-commerce and web development teams working with a modern JavaScript library. They also improve user experience and keep your frontend useful.

If you’re self-hosting Meilisearch, you can still track behavior using **JavaScript**, **useEffect**, or a custom logging layer. But with Meilisearch Cloud, everything’s built in and ready.

## Can you build a search engine in React for free?

Technically, you can build a search engine in React for free since React itself is free. Meilisearch is open source, and its self-hosted version lets you deploy a React web app and backend without paying a cent.

You can handle the frontend with `react-instantsearch` and `@meilisearch/instant-meilisearch`, then run everything on your local machine or private server.

That said, some things can still cost you:

* API rate limits if you use third-party data sources
* Hosting costs if you're not running locally
* Development time and maintaining your infrastructure
* No built-in analytics unless you’re on Meilisearch Cloud

So while the tech stack is free, scaling and managing it long term may not be. And while your React application can run for free, growing it at scale takes resources. Next, let’s look at what it takes to build a search engine using only JavaScript (without React).

## Can you build a search engine in JS without React?

You can build a full-featured search engine using just JavaScript without React.

With Meilisearch’s JavaScript client, you can connect directly to the backend, send search queries, and display results however you like using plain HTML, CSS, and vanilla JS. You won’t get prebuilt components like you do with React, but you gain full control over the structure and design.

Here’s what the basic flow looks like:

* Install the Meilisearch JavaScript SDK
* Initialize your client with a host and API key
* Send search queries to your Meilisearch index
* Parse the JSON response
* Render results manually in the DOM using **innerHTML** or template literals

Want a full walkthrough? Check out this tutorial to [learn how to build a search engine in JavaScript](https://www.meilisearch.com/blog/javascript-search-engine).

## Build responsive React search interfaces with Meilisearch

Building a search engine in ReactJS is manageable for both beginners and experienced React developers. Using Meilisearch’s JavaScript library, you can connect a React frontend to a flexible backend that supports real-time results and faceted filtering.

This setup works with popular web development tools like Next.js, supports client-side and SSR (server-side rendering), and gives you control over markup and component structure.

Whether you’re building a full-stack project or a simple standalone interface, Meilisearch for React has the building blocks to create a functional search engine that fits into modern frontend workflows.

How to Build a Search Engine in React: Step-by-Step Tutorial

Marketplace search engines are specialized tools designed to help users find what they need quickly and efficiently within an online marketplace such as Etsy, Amazon, eBay, and Walmart. 

They must be fast, typo-tolerant, offer faceted filtering, and support different input formats (text, words, documents, images, voice, and more). 

When done correctly, marketplace search engines impact user satisfaction, conversion rates, and sales by ensuring shoppers find precisely what they need. 

Among the best tools for creating marketplace search engines are Elasticsearch, Algolia, and Meilisearch. 

Let’s explore these and other alternatives further and learn how to seamlessly set up a marketplace search engine with Meilisearch.

## What is a marketplace search engine?

Marketplace websites are e-commerce websites, but instead of relying on products from a single vendor, they aggregate and rank listings from numerous sellers, each with varying prices, availability, and ratings.

Since marketplaces have dynamic inventories that change frequently, the search engine must process vast and numerous indices with different data types in real-time while balancing relevance, speed, and accuracy for a seamless shopping experience.

Therefore, a marketplace search engine is a robust AI-powered system that can be complex to build from scratch.

## How does a marketplace search engine work?

A marketplace search engine delivers fast, accurate results by combining advanced indexing, real-time querying, and intelligent ranking. Unlike traditional search systems, it must handle constantly changing inventory from multiple sellers while ensuring relevance and speed.

![How does a marketplace search engine work_.png](https://unable-actionable-car.media.strapiapp.com/How_does_a_marketplace_search_engine_work_feb03bc64a.png)

There are three key steps for a marketplace search engine to work:

1. **Indexing and Updating:** Listings from multiple vendors are converted into vector embeddings—numerical representations of the raw data format (documents, text, images, etc.). This allows the engine to understand the semantic approximation between products even when search terms don’t match exactly. Since sellers update their listings frequently, the system must be able to run constantly to update the semantic space with new and updated embeddings.
2. **Query Processing & Typo Tolerance:** The engine analyzes the query, accounting for typos, synonyms, and intent. Vector search enables semantic matching, returning relevant products even if the wording differs. Filters (price, brand, ratings) are applied dynamically to narrow results. Note that the query can tackle multiple indices, or a unified index that aggregates the data from the different listings.
3. **Ranking & Personalization:** Results are ranked using machine learning (ML) algorithms that weigh factors like popularity, seller reputation, price competitiveness, and user behavior. 

## Why is a search engine important for an online marketplace?

A fast and accurate AI search engine is key for online marketplaces to improve customer satisfaction, reduce bounce rates, and increase revenue.

A marketplace that does not rely on AI-powered search engines will lose market share to competitors that provide a better user experience. 

A good example of a marketplace that succeeded in providing faster and more relevant results is [Qogita](https://www.meilisearch.com/blog/qogita-simplifies-b2b-trade), which integrates the Meilisearch API into its infrastructure. 

*“Meilisearch has empowered us to deliver lightning-fast and highly relevant search results tailored to our customers' needs,*” said Ivo Silva, Engineering Director at Qogita.

Like Qogita, marketplaces can reduce their engineering costs while increasing their revenue by using intelligent search engines that are easy to set up through seamless APIs in multiple programming languages.

## What are the key features of a good marketplace search engine?

The key features of a good marketplace search engine matter at scale because multiple listings from different vendors must be indexed, and both the sellers and listings increase over time.

Taking scalability as one of the main pillars of a marketplace, the following are the most essential features of a good marketplace search engine.

### Speed and instant results

The system should be able to query all indices or a unified index with embeddings from multiple vendors in almost real-time to provide low-latency search-as-you-type results. 

According to [G2 reviews](https://www.g2.com/products/meilisearch/reviews), Meilisearch is one of the tools that excels in speed for large datasets: *“In terms of performance, it’s fast and handles searches and indexing smoothly, even with larger datasets.”*

### Relevance and ranking customization

When delivering the most relevant results, the search system should be flexible and adaptable to the marketplace’s needs.

Some may prioritize price, while others the reviews, or even a more complex ranking customization.

### Faceted filtering and navigation

It is necessary to narrow down listings across multiple vendors efficiently. 

Faceted search helps users [refine results](https://www.meilisearch.com/blog/nested-hierarchical-facets-guide). For example, a buyer searching for "running shoes" can quickly drill down to "men’s," "under $100," and "4+ stars" without manually sifting through pages of mismatched products.

Additionally, faceted filtering works hand-in-hand with ranking algorithms.

### Typo tolerance and error correction

Vector search converts inputs to vector embeddings, making the results less sensitive to typos and errors from the user. For instance, Meilisearch has default [typo tolerance rules](https://www.meilisearch.com/blog/typo-tolerance) that allow customization for your project’s needs.

### Personalization and contextual results

Past searches, views, clicks, and add-to-basket activities should be considered insightful data for future search queries made by the same user to enhance ranking algorithms and increase accuracy over time.

## What are the common types of marketplace search engines?

Marketplace search engines vary in complexity, from basic keyword matching to AI-driven intelligence. Each type serves different use cases.

| **Search engine type** | **Best for** |
| --- | --- |
| **Keyword-based search engines** | Precise product searches based on product numbers or models. |
| **Semantic search engines** | Understanding the user’s intent. |
| **AI-powered search engines** | Understanding the context and the user’s behavior. |
| **Federated search engines** | Cross-platform searches. |

Let’s take a look at the common types of marketplace search engines individually:

* **Keyword-based search engines:** Relies on exact or partial word matches. They are ideal for structured data but [struggle with synonyms or intent](https://www.meilisearch.com/blog/hybrid-search). It is the fastest way to get results.
* **Semantic search engines:** Uses natural language processing (NLP) to interpret meaning, improving results for vague queries.
* **AI-powered search engines:** Relies on ML and vector embeddings to deliver highly relevant results that understand intent and context.
* **Federated search engines:** Queries multiple indices (e.g., products, reviews) simultaneously, unifying seller catalogs.

## What are examples of successful marketplace search engines?

Marketplace search engines power some of the world’s largest product platforms.

* **Facebook Marketplace:** This platform uses location-based search, keyword matching, and personalized filtering for local buying/selling, prioritizing proximity and recent listings.
* **Amazon Marketplace:** AI-powered semantic and vector search, combining personalized recommendations, typo tolerance, and real-time inventory updates.
* **Etsy:** Focuses on the semantic search for handmade/unique items, understanding natural queries like "bracelet made of wood with a tribal look."
* **eBay:** Relies on keyword search, filter-based search, and vector search. Its hybrid system is optimized for exact product matches (e.g., "PlayStation 5 sealed box") and enhances search relevance across massive, constantly changing inventories.
* **Airbnb:** Combines faceted search (price, amenities) with location-based discovery. Strong use of filter personalization.
* **Fiverr:** Incorporates natural language understanding to interpret the intent behind user queries, e.g., understanding that “logo animation” and “animated logo” mean the same thing.

Each platform uses different search techniques depending on the products it aims to sell. For instance, Airbnb requires more filters than other marketplaces because customers require more fine-tuning (e.g., search by price, reviews, location, amenities). 

A robust marketplace search engine is not driven by the most top-notch technology, but by the customer's needs.

## What are the best software tools for building a marketplace search engine?

Choosing the right search tool is critical for marketplace performance. Here are some top solutions balancing speed, scalability, and relevance.

### 1. Meilisearch

![image2.png](https://unable-actionable-car.media.strapiapp.com/image2_0b644b6760.png)

This open-source search engine delivers instant, typo-tolerant results with minimal setup. Its RESTful API makes integration straightforward, while features like filtering and sorting help users find what they need quickly. The lightweight design ensures low latency even with growing product catalogs.

**Best for:** Due to its performance and speed for data retrieval, it is best for enterprises that manage various data sources and require typo tolerance.

<Callout intent="tip">

Meilisearch is the best way to implement marketplace search if you need speed and the ability to manage many data sources. [Get started for free](https://www.meilisearch.com/cloud)!

</Callout>

### 2. Elasticsearch

![image11.png](https://unable-actionable-car.media.strapiapp.com/image11_2eef4eefca.png)

[Source](https://www.elastic.co)

Built on Apache Lucene, Elasticsearch handles complex queries across massive datasets. Its distributed architecture scales horizontally, while powerful analytics capabilities provide insights into search patterns. The ecosystem includes Kibana for visualization and Elastic APM for performance monitoring.

**Best for:** Large-scale marketplaces requiring complex queries and big data handling.

### 3. Algolia

![image3.png](https://unable-actionable-car.media.strapiapp.com/image3_7fefefa566.png)

[Source](https://www.algolia.com/)

Algolia's hosted API delivers AI-powered, accurate results. Features include faceted navigation, personalization, and A/B testing for optimization. The platform handles indexing and infrastructure, freeing teams to focus on user experience.

**Best for:** Marketplaces prioritizing speed, personalization, and analytics.

### 4. Typesense

![image9.png](https://unable-actionable-car.media.strapiapp.com/image9_be714c1d01.png)

[Source](https://www.codemanas.com/what-is-search-with-typesense/)

This open-source alternative combines Algolia-like simplicity with self-hosting benefits. Built in C++ for speed, it offers typo tolerance, faceting, and geosearch. The small memory footprint makes it cost-effective to run at scale.

**Best for:** Teams needing a high-performance search they can fully control and customize.

### 5. Solr

![image4.png](https://unable-actionable-car.media.strapiapp.com/image4_b6ddc59db2.png)

[Source](https://solr.apache.org/guide/solr/latest/query-guide/query-screen.html)

Apache Solr powers some of the world's largest search implementations. Its extensive plugin system supports multilingual search, advanced relevancy tuning, and machine learning integration. In addition, Apache Solr is entirely free and open-source.

**Best for:** Large marketplaces with complex search requirements and technical resources.

## How to create a marketplace search engine with Meilisearch

Here is a brief step-by-step process to create a marketplace search engine using Meilisearch:

### 1. Set up a Meilisearch instance

You can deploy Meilisearch easily using Meilisearch Cloud (with a free trial) or self-host it locally or via a Docker image.

After deployment, obtain your database URL and API keys (Search API Key for read-only access, Admin API Key for updating data/settings).

### 2. Prepare your marketplace data

Collect your product or marketplace listing data in a supported format such as JSON or CSV. Ensure each document has a unique identifier (ID) for indexing and updates.

### 3. Create an index and upload documents

Use Meilisearch API or SDKs (e.g., Python SDK) to create an index (e.g., “products”) and add your marketplace documents to it. Example Python snippet to add documents:

```python
import meilisearch  

client = meilisearch.Client('YOUR_MEILI_URL', 'YOUR_ADMIN_API_KEY')  

index = client.index('products')  

index.add_documents(your_documents_list)
```

### 4. Configure search settings

Customize ranking rules to prioritize relevant results (e.g., by popularity, recency).

Set searchable attributes (e.g., product name, category, description) and define synonyms or stop words if needed.

### 5. Implement search interface

Build your frontend search UI that queries Meilisearch using the Search API Key for real-time, typo-tolerant, and faceted search experiences.

Support features like “search-as-you-type,” filters, and highlighting to enhance the user experience.

### 6. Optimize and maintain

Continuously update your index with new or changed marketplace data.

Use Meilisearch’s language-specific tokenization and normalization for better full-text search relevance.

Optionally, integrate AI-powered search for semantic search capabilities using embeddings and OpenAI.

## What are the challenges of scaling a marketplace search engine?

Scaling a marketplace search engine presents technical and business difficulties that increase as the company scales. 

One of the biggest challenges is maintaining search speeds and relevancy as the product catalogs expand. For instance,  a query for "gaming PC" requires the system to ensure that all listings are correctly updated and indexed. Next, the ranking algorithm must be sophisticated enough to output the best results from hundreds of products, usually backed by filters and user behavior.

Poor relevance may lead to increased bounce rates, lost conversions, and weak customer satisfaction without proper tuning.

In addition to real-time filtering, typo tolerance, and multilingual options are also resource-intensive at scale. For instance, the latter needs to handle non-Latin characters or perform translations at the query level, which leads to higher infrastructure costs.

Solutions like Meilisearch directly address these drawbacks with a lightweight architecture that delivers fast, typo-tolerant search even with massive catalogs. Its intuitive relevance tuning and efficient indexing help marketplaces scale smoothly without excessive infrastructure investment.

## Start creating smarter search engines today!

A good marketplace search engine should deliver lightning-fast and accurate results based on multiple listings. As the platform scales, the system should be able to update the indices frequently without compromising query latency.

Modern solutions like Meilisearch are adaptable to different business needs, open-source, and able to deal with scalability challenges while offering other important features such as multilingual support, typo tolerance, faceted filtering, and AI-enhanced ranking algorithms.

Marketplace search engine: How to make one, top tools, & more

A searchable knowledge base is a structured information hub, like Frequently Asked Questions (FAQs) or guides, built for quick and easy retrieval. It works by indexing content and using algorithms like typo-tolerant search to deliver precise results instantly. 

Some benefits include slashing the volume of support tickets and boosting customer loyalty by proxy of independence. Common knowledge base types can be *internal*, for employee resources, and *external*, for customer help centers. 

In this post, we’ll show you a couple of top tools for building a searchable knowledge base, and we’ll also show you how to build one using our own platform, Meilisearch. So, let’s get started. 

## What is a searchable knowledge base?

A searchable knowledge base is a structured repository or database of information, such as articles, FAQs, or guides, designed for effortless search and retrieval. Picture a digital library where answers to your questions are just one search away. 

Unlike static or poorly organized knowledge bases that force users to manually sift through documents, a searchable system uses indexing, search algorithms, and metadata to deliver precise results amazingly fast. This ensures that users – whether customers or employees – find what they need without frustration.

First, let’s explore how these systems actually function.

## How does a searchable knowledge base work? 

A searchable knowledge base delivers answers with speed and precision by virtue of how it is built.

![image6.png](https://unable-actionable-car.media.strapiapp.com/image6_a919b23e13.png)

Here’s how it works:

* **Step 1: Indexing** refers to mapping out content, such as articles or FAQ pages, so the search engine knows exactly where to look. This process breaks down text into searchable components, making the search process fast.
* **Step 2:** When a user enters a query, algorithms scan the index produced in Step 1 to **find matches**. They rank results based on relevance, using metrics like keyword frequency or proximity. Tools like Meilisearch enhance this with [typo tolerance](https://www.meilisearch.com/blog/typo-tolerance), catching misspellings for better accuracy.
* **Step 3:** Metadata like meta tags, subcategories, or dates adds context to enable **result filtering**. For example, searching “billing issues” can be narrowed to articles tagged “invoices.” [Faceted search](https://www.meilisearch.com/blog/nested-hierarchical-facets-guide), also supported by Meilisearch, allows users to refine results by specific attributes.
* **Step 4:** If all goes well, the search engine produces **accurate and relevant results**, often with highlighted keywords or snippets. For example, a query for “reset password” will pull up the exact guide instantly, saving users from manual browsing.

This process ensures users get answers quickly, which brings us to the tangible benefits.

## What are the benefits of searchable knowledge bases?

Searchable knowledge bases help organizations manage information and deliver measurable advantages for users and intra-company teams. 

![image5.png](https://unable-actionable-car.media.strapiapp.com/image5_e5df53a1a1.png)

Here’s why they’re a game-changer.

### Improved user self-service success

Empowering users to find answers independently is a cornerstone of modern customer support. A searchable knowledge base lets customers resolve issues without contacting company support, increasing customer independence. 

According to the [Zendesk Customer Experience Trends Report 2020](https://www.zendesk.com/blog/zendesk-customer-experience-trends-report-2020/), 69% of buyers prefer to solve problems independently. Thus, such a self-serve model fosters user confidence and cuts down on repetitive inquiries.

### Increased operational efficiency

Knowledge bases free customer support teams from answering routine questions by enabling self-service in customer support.

[Gorgias reports](https://www.gorgias.com/blog/automation-impact-on-cx-data) that a well-implemented AI-powered knowledge base can decrease the ticket-to-order ratio by up to 27%. Reducing support agent contacts allows staff to focus on complex issues, streamlining operations, and reducing roadblocks to project progression.

### Enhanced customer satisfaction

Quick access to solutions is a market-respected way to keep customers happy. A searchable knowledge base delivers answers in milliseconds, reducing wait times and frustration.

[Research by Deloitte](https://www.deloittedigital.com/us/en/insights/research/personalizing-growth.html) reported that brands that incorporate personalization into their customer service, an aspect that searchable knowledge bases facilitate, are 71% more likely to experience customer loyalty.

### Better internal knowledge management

Internally, a searchable knowledge base centralizes critical information, from policies to troubleshooting guides. This reduces knowledge silos, ensuring teams stay aligned and the decision-making individuals stay informed.

Employees can quickly access up-to-date resources, improving inter-departmental collaboration.

These benefits vary depending on the knowledge base you implement, so let’s break down the different types.

## What are the different types of knowledge bases?

Knowledge bases are versatile and can be tailored to serve specific audiences, purposes, and use cases. Understanding the types helps you choose the right one for your needs.

### Internal knowledge base

An internal knowledge base is designed with a company’s employees in mind.

It usually houses resources like onboarding guides, standard operating procedures (SOPs), and troubleshooting FAQs. It streamlines knowledge sharing and typically uses [enterprise search](https://www.meilisearch.com/blog/enterprise-search), ensuring team members have instant access to the information they need to excel in their roles.

### External knowledge base

External knowledge bases, on the other hand, are more customer-facing and include FAQ sections, help centers, and support articles.

They empower users to find answers independently without ringing the help desk. By providing accessible, clear information, they reduce support inquiries and enhance the customer experience.

### Product knowledge base

As the name suggests, a product knowledge base is focused on product-specific documentation.

It features developer docs, tutorials, and user guides. Customers and internal teams rely on this to use and support a product.

### IT knowledge base

An IT knowledge base keeps IT teams happy and stocked with the information they need, when they need it.

It provides system information, troubleshooting guides, and network documentation. It serves as internal support, enabling IT staff to resolve technical issues efficiently and reduce system downtime.

### HR knowledge base

An HR knowledge base not only keeps employees informed about their rights in the company but also lightens the HR team's workload.

It stores company policies, employee benefits, and compliance documents. Employees get easy access to resources like vacation policies or handbooks, thus promoting transparency and ensuring compliance across the organization.

## What are the key features of an effective searchable knowledge base?

A functional, searchable knowledge base needs the following to ensure usability, security, and relevance.

![image3.png](https://unable-actionable-car.media.strapiapp.com/image3_7dab147cc6.png)

### Advanced search functionality

Robust search capabilities are non-negotiable for effective search. Features like typo tolerance, ranking rules, and faceted search make finding relevant information intuitive.

For example, Meilisearch’s typo tolerance ensures users get correct results even with misspellings, e.g., “invices” instead of “invoices”. Similarly, faceted search lets them filter by categories like “billing” or “technical.”

### Real-time content updates

Keeping content up-to-date is vital for *the most* accurate results. Real-time updates, whether through versioning or live update systems, ensure users always have access to the latest information. This eliminates reliance on outdated guides, ensuring accuracy and establishing trust.

### User-friendly content organization

Logical and user-focused content organization enhances usability. Intuitive menus and effective tagging help users navigate knowledge bases effortlessly.

Additionally, clear categories, such as “Getting Started” or “Troubleshooting” for product documentation, further improve user navigation in the answer-finding process.

### Access control and permissions

Controlling access and permissions is essential, especially for internal knowledge bases. Defining access levels progressively up the organizational hierarchy ensures the right decisions are made at the right level.

Similarly, user-role-defined permissions ensure sensitive or department-specific content is only visible to authorized users. Together, access and permission controls protect data integrity and confidentiality.

### Analytics and feedback mechanisms

Understanding user behavior is key to system improvement. Studying user analytics may reveal popular searches, content gaps, and where the search process leaves them wanting more. 

Furthermore, feedback tools like “Was this helpful?” buttons help refine articles by identifying potential areas of improvement. This data-driven approach keeps the knowledge base relevant.

### Mobile accessibility

In the age of mobile first, a knowledge base should be available on hand-held devices such as smartphones and tablets. Through responsive design, users can search and access information anywhere, increasing convenience and accessibility.

Next, let’s look at the best tools to bring these features to life.

## What are the best software tools for building a searchable knowledge base?

The right tool depends on your organization’s size, needs, and technical capabilities. Here are the top five options to help you make that choice.

### 1. Meilisearch

![image4.gif](https://unable-actionable-car.media.strapiapp.com/image4_78cac7e82a.gif)

Source: Meilisearch

Meilisearch is both an open-source and premium search engine that delivers lightning-fast, relevant search experiences. With search-as-you-type results in under 50 milliseconds, typo tolerance, and [hybrid search](https://www.meilisearch.com/blog/hybrid-search) combining keyword and semantic capabilities, it’s perfect for building intuitive knowledge bases. Plus, its plug-and-play setup, [RAG-based infrastructure](https://www.meilisearch.com/blog/how-to-build-rag), and SDKs make integration into virtually any infrastructure a breeze.

**Best for**: Organizations seeking a user-friendly, scalable, and [intelligent search solution](https://www.meilisearch.com/blog/intelligent-search).

<Callout intent="tip">

#### Enhance your knowledge base with Meilisearch

Experience lightning-fast search and advanced functionalities to make your knowledge base more user-friendly and effective. [Try Meilisearch for free!](https://www.meilisearch.com/cloud)

</Callout>

### 2. Atlassian Confluence

![image9.png](https://unable-actionable-car.media.strapiapp.com/image9_08e044307d.png)

Source: Atlassian Confluence

Atlassian Confluence is a collaborative platform built specifically for creating and managing internal knowledge bases. It offers real-time editing, version control, whiteboards, and robust search functionalities.

It also comes equipped with templates for organizing content. One of its defining features is seamless integration with Jira, Slack, and many other tools, making it ideal for teams within and outside the Atlassian ecosystem.

**Best for**: Teams needing a collaborative, enterprise-grade documentation platform.

### 3. Zendesk

![image2.png](https://unable-actionable-car.media.strapiapp.com/image2_31cb14fcd5.png)

Source: Zendesk

Zendesk’s help center software includes a powerful knowledge base for customer self-service. It artfully combines generative and agentic AI with content management and user analytics to reduce ticket volume by up to 15%.

Plus, its intuitive interface and reporting tools make it a strong choice for support-focused organizations.

**Best for**: Businesses wanting an integrated customer support and knowledge base solution.

### 4. Helpjuice

![image8.png](https://unable-actionable-car.media.strapiapp.com/image8_3113a2338b.png)

Source: Helpjuice

Helpjuice is another user-friendly inter-team knowledge base platform with AI-assisted content creation and typo-tolerant search. It offers rich formatting for easy content creation within the platform, multi-language support for truly global teams, and analytics to track user behavior.

Its AI-powered chatbot and search functionality, coupled with multiple customization options, make it ideal for global teams needing tailored solutions.

**Best for**: Companies prioritizing ease of use and advanced content management.

### 5. Notion

![image11.png](https://unable-actionable-car.media.strapiapp.com/image11_1490ca9d19.png)

Source: Notion

Notion is a versatile workspace combining wikis, documentation, and project management. Its flexible infrastructure, powerful search capabilities, and AI features make it great for creating all kinds of knowledge bases.

It has remarkable security features, while teamspaces and integrations enhance collaboration across tools.

**Best for**: Teams wanting an all-in-one platform for knowledge and project management.

Now, let’s dive into building a knowledge base with Meilisearch.

## How to create a searchable knowledge base with Meilisearch?

Meilisearch transforms a basic knowledge base into a superior searchable experience with speed, accuracy, and relevance.

![image12.png](https://unable-actionable-car.media.strapiapp.com/image12_66b0430409.png)

Here’s a step-by-step guide to get you started.


### 1. Set up Meilisearch

Getting started with Meilisearch is a breeze. Choose Meilisearch Cloud for a managed, hassle-free setup, or self-host using Docker, a binary, or other methods. 

For self-hosting, install with a command like:

``` 
# Install Meilisearch
curl -L https://install.meilisearch.com | sh
```

Secure your instance with API keys (master, admin, search) to control access and permission levels, maintaining a tight grip on your knowledge base’s protection.

### 2. Create an index

An index is Meilisearch’s container for your searchable documents, like articles or FAQs. Create one via the web UI or API, naming it something relevant, such as “knowledge-base.”

This organizes your content, setting the stage for efficient searches. Here, a REST API call like `POST /indexes` with `{"uid": "knowledge-base"}` would do the trick.


### 3. Add documents

Populate your index with content formatted as JSON, NDJSON, or CSV files. Make sure to assign a unique identifier (e.g., “id”) to each document. Use the web UI for small datasets or the REST API/SDKs for bulk uploads.

Here’s a Python script example:

```python
# In the command line:
# pip3 install meilisearch

# In your .py file:
import meilisearch
import json

client = meilisearch.Client('http://localhost:7700', 'your_master_key')

json_file = open('knowledge_base.json', encoding='utf-8')
documents = json.load(json_file)


json_file = open('knowledge_base.json', encoding='utf-8')
documents = json.load(json_file)


client.index('knowledge-base').add_documents(documents)
```

This code assumes you have a `knowledge_base.json` file with content like this:

```json
[  {
    "id": 1,
    "title": "How to Reset Password",
    "content": "Step-by-step guide to reset your password."
  },
  {
    "id": 2,
    "title": "Account Settings",
    "content": "Manage your account preferences."
  }
]
```

Save this JSON in `knowledge_base.json`, and the code will upload it to the `knowledge-base index`.


### 4. Configure search settings

Customize how searches work by defining searchable attributes (e.g., “title,” “content”) and ranking rules to prioritize results, like favoring recent or relevant documents.

Enable filters and facets for users to narrow results by categories or dates. For instance, set searchable attributes via the API: `POST /indexes/knowledge-base/settings` with `{"searchableAttributes": ["title", "content"]}`.

### 5. Integrate search into your application

Bring search into your app using Meilisearch’s REST API or SDKs (Node.js, Python, etc.). Query the index from a search bar and display results with highlighted matches.

Here’s a JavaScript example:

```javascript
const { MeiliSearch } = require('meilisearch');
const client = new MeiliSearch({ host: 'http://localhost:7700' });
client.index('knowledge-base').search('reset password').then((results) => {
    console.log(results.hits);
});
```

This retrieves and logs search results, ready for display.

### 6. Advanced features

Meilisearch’s advanced features elevate your knowledge base and take it from what it is to what it can be.

Typo tolerance ensures that accurate results appear despite misspellings, while synonyms and stop words improve relevance by handling related terms or ignoring common words. Hybrid search blends keyword and semantic search for context-aware results, ideal for complex queries. 

And you can configure and optimize these via the API to fine-tune your search experience.

With these steps, your knowledge base is ready to shine!

## Over to You!

A searchable knowledge base empowers users and simplifies information access across platforms quickly. Meilisearch enables you to create a fast, intuitive system tailored to your organization's requirements, whether for customer support, internal documentation, or product documentation.

This helps reduce support tickets, improve efficiency, and boost customer satisfaction, giving your company a strategic advantage. Start now and find out how Meilisearch enables your users to find answers.

Searchable knowledge base: How to make one, top tools, & more

When we discovered that 70% of our team was already using AI tools daily—but with wildly varying results—we knew we needed a systematic approach. Not another vague "AI transformation" initiative, but a practical framework for separating AI hype from real value.

Here's the concrete system we developed to evaluate and implement AI tools across our company and what we've learned so far.

## The surprising reality we uncovered

When we began our AI journey at Meilisearch, we expected to find pockets of adoption. Instead, we discovered something more interesting: **universal experimentation but inconsistent value**.

### The adoption paradox

Our company-wide assessment revealed that every single team member had already experimented with AI tools, with an impressive 70% incorporating them into daily workflows.

![image.png](https://unable-actionable-car.media.strapiapp.com/image_40601b4a7c.png)

![image2.png](https://unable-actionable-car.media.strapiapp.com/image2_a867a1c935.png)

But the numbers told only half the story:

Despite high adoption rates, team sentiment about AI's value showed surprising variation. Many were using the tools but questioning whether they were truly beneficial or just creating a new kind of busywork.

![image3.png](https://unable-actionable-car.media.strapiapp.com/image3_67ad5244d9.png)

![image4.png](https://unable-actionable-car.media.strapiapp.com/image4_4a6d31dcca.png)

This gap between usage and perceived value became our first critical insight.

### Four critical insights that changed our approach

<Callout intent="note">

**Previously Shared Insights:**

- 🧠 [𝐃𝐨𝐞𝐬 𝐀𝐈 𝐛𝐞𝐜𝐨𝐦𝐞 𝐥𝐞𝐬𝐬 𝐯𝐚𝐥𝐮𝐚𝐛𝐥𝐞 𝐚𝐬 𝐲𝐨𝐮𝐫 𝐞𝐱𝐩𝐞𝐫𝐭𝐢𝐬𝐞 𝐠𝐫𝐨𝐰𝐬?](https://www.linkedin.com/posts/gillianmcauliffe_aitransformation-professionalgrowth-futureofwork-activity-7292477225889660928-7UXm/?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAXkmQUBijf17cl9IKQZhMWLN67r9deqvAc) How your relationship with AI changes at different levels of expertise.
- 🔍 [**Unlocking Our AI Knowledge-Sharing Blueprint:**](https://www.linkedin.com/posts/gillianmcauliffe_aitransformation-futureofwork-knowledgesharing-activity-7290061146634485762-fm1D/?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAXkmQUBijf17cl9IKQZhMWLN67r9deqvAc) Understanding what people want to know to get started.
- 🤝 [Preserving Human Connection in an AI-Enhanced Workplace](https://www.linkedin.com/posts/gillianmcauliffe_aitransformation-teamwork-futureofwork-activity-7287036478092632066-boTG/?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAXkmQUBijf17cl9IKQZhMWLN67r9deqvAc): Will we lose what's special about our teams by bringing in AI?

</Callout>

Through in-depth interviews across all departments, we uncovered patterns that wouldn't have appeared in simple usage statistics:

1. 🌿 **Environmental impact:** Our teams have raised thoughtful concerns about the environmental impact of AI.
2. **🤔 The expertise paradox:** As team members' domain expertise increases, their perception of AI's value often follows a U-curve. Beginners find AI helpful for learning, experts leverage it as a powerful accelerator, but those in the middle sometimes find it more hindrance than help.
3. **🎯 The trust dilemma:** AI's "inconsistent brilliance" creates a fundamental trust issue—sometimes transformative, sometimes bafflingly off-target—making teams reluctant to integrate it into mission-critical processes.
4. **✨ Creative acceleration:** Despite skepticism, nearly everyone reported dramatic improvements in creative processes—from eliminating writer's block to transforming editing workflows.
5. 🔭 **Prompt sophistication gap:** Most team members were still using basic prompting techniques, unaware of how dramatically their results could improve with more advanced approaches.

This investigation revealed we weren't facing an adoption problem but a **value extraction challenge**. People were using AI tools but were not consistently getting meaningful results. We needed a framework to help teams identify where AI could actually create value—and where it was just adding complexity.

## ⚖️ Our solution: the AI value scoring matrix

Based on these insights, we developed a systematic framework to cut through the AI hype cycle. Rather than chasing the latest tools or models, we created a data-driven method to identify where AI could create genuine value for our specific context.

**The value scoring matrix:** Each potential AI use case is evaluated across multiple weighted dimensions aligned with our organizational priorities. This transforms the abstract question of "where should we use AI?" into a concrete, comparable set of scores.

**Our Six Critical Dimensions:**

| **Dimension** | **Weight** | **What We Assess** | **Why It Matters** |
|--------------------|------------|--------------------------------------------------------------------|----------------------------------------------------------------------------------|
| Feasibility | 20% | Implementation difficulty with current skills/resources | Prevents us from chasing technically impressive but impractical applications |
| Revenue Impact | 25% | Potential to increase top-line growth | Ensures AI initiatives contribute to business success |
| Data Readiness | 15% | Availability and quality of necessary data | Many AI projects fail due to data issues, not technology limitations |
| Time to Value | 15% | Speed to tangible results | Builds momentum with quick wins |
| Risk | 10% | Potential challenges or downsides | Protects against unintended consequences |
| Internal Impact | 15% | Number of people/teams benefiting | Prioritizes broad improvements over narrow optimizations |

Here's how we applied this framework to evaluate potential AI initiatives:

This systematic approach cut through subjective debates about which AI tools to adopt, replacing them with evidence-based decisions that aligned with our strategic priorities.

## Early results: beyond the AI hype cycle

Our scoring matrix has already transformed how we approach AI at Meilisearch:

### 1\. From endless options to strategic focus

Instead of drowning in the constant stream of new AI products, our teams now have a clear filter: "Does this tool address one of our high-scoring use cases?" This has dramatically reduced distractions and FOMO while increasing meaningful adoption.

### 2\. Balanced investment: "Ferraris" and "cargo ships"

We're strategically balancing our implementation:

- **"Ferrari" projects:** Quick wins that generate excitement and momentum (e.g., automating our internal documentation updates)
- **"Cargo ship" initiatives:** Substantial efforts with transformative potential (e.g., reimagining our customer support workflow with AI assistance)

### 3\. Creating a dedicated space for experimentation

One crucial lesson: AI adoption requires protected time. We've integrated dedicated "AI exploration blocks" into our work schedules and upcoming offsite, focusing specifically on our highest-scoring use cases.

### 4\. Addressing real concerns

The scoring matrix naturally incorporates ethical considerations through the risk dimension. We've operationalized this by:

- Favoring smaller, more efficient models when possible
- Prioritizing open-source options that allow us to run models locally
- Building in time for human review of AI-generated content

## The framework in action: a case study

After applying our scoring matrix to use cases identified through internal discussions, we quickly surfaced high-impact opportunities. One standout example was improving our code review process—specifically, reducing time to first review for pull requests (PRs).

This use case scored highly across multiple dimensions:

- High feasibility with existing tools
- Direct impact on development velocity
- Clear time-to-value metrics
- Broad internal impact across engineering teams

We implemented CodeRabbit, an AI-powered code review assistant that provides immediate PR feedback. While maintaining our essential human review process, the tool identifies common issues and provides automated PR descriptions. After a two-week trial period, we measured both quantitative metrics and gathered team feedback.

The results validated our scoring framework's prediction:

- Decreased average time to first review
- Improved PR documentation quality
- Positive team adoption and engagement
- Cumulative time savings across the engineering organization

## What's next: your turn

We're sharing this framework because we believe it can help other organizations cut through AI hype and focus on value creation.

The key to successful AI adoption isn't about having the fanciest tools—it's about having the clearest framework for identifying where AI truly adds value in your specific context.

🚀 How we're making AI work at Meilisearch

Hey, here's a recap of everything that happened at Meilisearch during April.

## 🚀 Meilisearch 1.14

[![meilisearch-1.14.png](https://unable-actionable-car.media.strapiapp.com/meilisearch_1_14_166a7ff64e.png)](https://www.meilisearch.com/blog/meilisearch-1-14)

Meilisearch 1.14 improves AI search with composite embedders and embedding cache. It also adds granular filterable attributes and batch document retrieval by ID.

[Read the release notes](https://www.meilisearch.com/blog/meilisearch-1-14)

## 📰 In the news

### 🔢 Why you shouldn't use a vector database for RAG

The contrarian take on building better retrieval augmented generation systems.

[Read the blog](https://www.meilisearch.com/blog/vector-dbs-rag)

### ⚙️ What is full-text search, and how does it work?

Learn the inner workings of a full-text search engine.

[Read the blog](https://www.meilisearch.com/blog/how-full-text-search-engines-work)

### 🎙️ Coruzant podcast with Quentin de Quelen

Listen to our CEO discuss his entrepreneur journey and all things Meilisearch.

[Listen to the podcast](https://coruzant.com/profiles/quentin-de-quelen/)

### 🔝 Top 10 alternatives to Elasticsearch

Discover their key features, pricing, pros, and cons based on users feedback.

[Read the blog](https://www.meilisearch.com/blog/elasticsearch-alternatives)

### 🔍 What is federated search?

Discover the different types of federated search, and their benefits and challenges.

[Read the blog](https://www.meilisearch.com/blog/what-is-federated-search)


## 🤗 Made with Meilisearch

[Agora](https://www.meilisearch.com/blog/Agora) uses Meilisearch hybrid search to provide instant search results across various data sources to build a fast and intuitive ecommerce experience.

[![Agora gif.gif](https://unable-actionable-car.media.strapiapp.com/Agora_gif_eca98f7605.gif)](https://www.meilisearch.com/blog/Agora)

_We’re always looking to promote new projects from the community. Post yours in the [#built-with-meilisearch](https://dub.sh/meili-discord?utm_campaign=newsletter&utm_source=blog&utm_medium=april-25-updates) Discord channel!_

---

Receive this updates in your inbox by subscribing to our [monthly newsletter](https://dub.sh/meili-newsletter?utm_campaign=newsletter&utm_source=blog&utm_medium=april-25-updates). To learn more about Meilisearch's future and help shape it, take a look at our [roadmap](https://roadmap.meilisearch.com/?utm_campaign=newsletter&utm_source=blog&utm_medium=april-25-updates) and participate in our [Product Discussions](https://github.com/orgs/meilisearch/discussions).

For anything else, join our developers community on [Discord](https://dub.sh/meili-discord?utm_campaign=newsletter&utm_source=blog&utm_medium=april-25-updates).

Meilisearch latest news and company updates