> For the complete documentation index, see [llms.txt](https://docs.hackle.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.hackle.io/en/ab-test/mab-test/mab-vs-ab-test.md).

# MAB Test vs A/B Test

{% hint style="info" %}
A MAB (Multi-Armed Bandit) Test is an experiment method that automatically adjusts traffic per group to maximize a success metric.
{% endhint %}

![](https://143363954-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FyAVlNQhS253reum0Ce94%2Fuploads%2Fgit-blob-e80801a184bf188e7aff21b9f66ef61030bd705c%2F02a164f-MAB_KOR_282ea19f721fbb14.png?alt=media)

<table><thead><tr><th width="168.42578125">Category</th><th>A/B Test</th><th>MAB Test</th></tr></thead><tbody><tr><td>Traffic distribution</td><td>Fixed per group</td><td>Automatically adjusted based on performance</td></tr><tr><td>Main purpose</td><td>Comparing results and making decisions</td><td>Maximizing the success metric</td></tr><tr><td>Best suited for</td><td>Long-term, careful validation</td><td>Experiments that need fast optimization</td></tr><tr><td>Duration</td><td>4 weeks or more</td><td>Shorter than an A/B Test (about 1~2 weeks)</td></tr></tbody></table>

### A/B Test

An A/B Test is an experiment method where a fixed amount of traffic is allocated to each group under a controlled environment over a period of time, and decisions are made by comparing performance between groups.

It allocates fixed traffic to each group and compares performance and statistical significance over a sufficient period.

{% hint style="success" %}
An A/B Test is a good fit when you want to compare performance and statistical significance over a sufficient period.
{% endhint %}

### MAB Test

The main goal of MAB is to answer "Which test group shows the greatest reward (= success metric optimization)?"

Because it automatically allocates more traffic to the better-performing groups, it is a good fit for quickly maximizing a single success metric.

{% hint style="success" %}
A MAB Test is suitable for maximizing conversions for short and temporary experiences where changes are not permanent.

* When you want to identify the optimal option within a short period and expose it to more users
* Promotional offers, headline tests, webinar registration pages, and so on
* When continuous optimization is needed, such as for algorithm tests like search and recommendation logic
  {% endhint %}

{% hint style="warning" %}
Since traffic distribution is determined by a single success metric, decisions may differ from those that consider multiple metrics.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.hackle.io/en/ab-test/mab-test/mab-vs-ab-test.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.