# How it Works

1. **First Scan:** Once a MatchAgent is loaded onto your website, it will scan and use the website's full context for optimizing content.
2. **UTM Rule Check:** The script reads UTM parameters in the page URL and checks them against the rules configured in your MatchAgent (e.g., `utm_source equals google`). If a match is found, personalization fires. If not, the page displays normally.
3. **Content Swap:** Only the elements you've mapped in the MatchAgent are updated — H1s, subheads, CTAs, badges, or sections. The rest of the page remains untouched.
4. **Caching:** After the first time a unique set of UTM keys is run through the system, those hyper-personalized variants will be cached for instantaneous retrieval on subsequent page loads.
5. **Editable Fields:** These cached AI-optimized content blocks are editable in the CampaignMatches tab of the dashboard.
6. **Tracking & Analytics:** Page views and click-through rates are measured for each CampaignMatch for a clear distinction of segment performance metrics. Individual buttons can also be tagged to measure more granular conversion data for page testing and reports.

Ad Click → UTM in URL → MatchAgent Checks Rules → Match Found → Content Swaps → Result Cached → Analytics Updated

## Caching Logic

### Cache Key Components

The cache key includes: content text, user prompt, URL parameters (like campaign, demographics), and MatchAgent configuration. This ensures that different contexts get different AI responses.

{% hint style="info" %}
**Cache invalidation:** A new cache version is created any time the content text, Custom User Prompt, URL parameters, or MatchAgent configuration changes. This means edits to copy or rules will generate a new variant entry in Analytics. Previous version data is retained for comparison.

**What does NOT trigger a new version:** Visitor traffic volume changes, plan upgrades, or dashboard-only settings that don't affect the MatchAgent configuration.
{% endhint %}


---

# Agent Instructions: Querying This Documentation

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

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://messagematch.gitbook.io/messagematch-docs/readme/how-it-works.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.