# Evi Analysis

{% hint style="info" %}
This feature is currently in its Beta testing phase. During this period, users may experience changes and updates as improvements are made based on feedback. We encourage users to provide input to help us refine and enhance the feature before its official release.
{% endhint %}

Instead of manually sifting through data tables and charts, you can simply type your questions, and the AI will process the underlying metrics, statistical significance, and objective performance to deliver concise and relevant answers.

Evi leverages CRO best practices and statistical data to provide clear, data-backed answers and recommendations.

The assistant can answer with text, table, and/or chart based on your needs. 

{% hint style="success" %}
Use case examples: 

* Explain the winning variation
* Challenge my hypothesis
* Give me CRO best practices based on my campaign results
  {% endhint %}

### Tips for Asking Effective Questions :

To get the most accurate and helpful responses from the AI, consider the following:

* **Wait for readiness** : if you are willing to make a decision, wait for the readiness to be ok at least on primary goals.
* **Be Specific**: The more precise your question, the better the AI can understand your intent.
  * *Good:* "Which variation performed best for the 'Conversion Rate' objective?"
  * *Avoid:* "What about the conversion rate?"
* **Reference Objectives and Metric**s: Clearly state the objectives and metrics you are interested in.
  * "What is the statistical significance for the 'Click-Through Rate' of Variation B versus Control?"
  * "Show me the difference in average revenue per user between all variations."
* **Ask for Comparison**s: The AI is excellent at comparing performance between variations.
  * "Compare the bounce rate of Variation A and Variation B."
  * "Which variation had the highest increase in sign-ups compared to the control?"
* **Inquire About Statistical Significance:**
  * "Is the difference in \[Metric] between \[Variation X] and \[Variation Y] statistically significant?"
  * "What is the p-value for the 'Add to Cart' objective?"
* **Ask for Summaries:**
  * "Summarize the overall performance of this A/B test."
  * "What are the key takeaways from this report?"
* **Identify Top/Worst Performers:**
  * "Which variation had the highest \[Metric]?"
  * "Which objective performed the worst for Variation C?"
* **Troubleshooting & Clarification**:
  * "Can you explain the 'Confidence Interval' for the 'Purchase Rate'?"

### Use the Evi Analysis:

{% stepper %}
{% step %}

### Access your report.

{% endstep %}

{% step %}

### On the top right corner, click on *Ask Copilot* button.

<figure><img src="/files/m73YENVlHFdf1l8i5r4h" alt="" width="563"><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### Select a question or enter your query.

<figure><img src="/files/wfRcRMJi5gPNFz3eSEp6" alt="" width="563"><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### Read the copilot's answer and get a better understanding of your report.&#x20;

<figure><img src="/files/xDAGkcMLi4Pxnp0GcO9g" alt="" width="563"><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

### Limitations

While powerful, the Evi Analysis has some limitations:

* It does not work in Frequentist mode
* Relies on Available Data: The AI can only analyze the data presented in your report. It cannot infer information not present in the underlying metrics and statistics.
* Statistical Interpretation: The AI provides statistical interpretations, but it's crucial for you, to apply domain expertise and strategic context to those interpretations.
* Complex Scenarios: For highly complex or multi-variate statistical modeling, you may still need to consult with a data analyst.
* No Predictive Capabilities (currently): The AI focuses on analyzing past performance; it does not predict future outcomes or design new experiments (unless specifically integrated in a future release).


---

# 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://docs.abtasty.com/reporting-and-performances/reporting/campaign-reporting/analysis-copilot.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.