diff --git a/README.md b/README.md
index 1a8492f..91b2e5b 100644
--- a/README.md
+++ b/README.md
@@ -11,7 +11,12 @@
 
 **Find the best hyperparameters for Embedding your data in RAG Pipeline**
 
-![vectorboard banner image](docs/images/benchmark.png)
+Get a detailed report on differences between each hyperparameter combination:
+
+- Performance and embedding time
+- Quality of response on the Eval queries
+
+> More benchmark properties, such as cost estimation and Automated accuracy metric is upcoming.
 
 # TL;DR
 
@@ -105,6 +110,16 @@ Finally, view the results in a Gradio app using `.results()` method. To get a pu
 search.results(share=True)
 ```
 
+## What's included in the results?
+
+Several metrics are included in the results page in the Gradio app:
+
+- Total runtime and embedding time per experiment (a combination of hyperparameters in the `param_grid`)
+  ![vectorboard banner image](docs/images/benchmark.png)
+
+- Response on each `eval_query` per `Experiment` in a table format. This way you can see which experiment is giving the correct answer and you can easily compare them with each other.
+  ![gradio image](docs/images/gradio-screenshot-001.png)
+
 # Overview and Core Concepts
 
 RAG (Retrieval-Augmented Generation) revolutionizes the way we approach question-answering and text generation tasks by combining the capabilities of retrieval-based and generative models. While it excels in many areas, one of the critical factors for its success is the quality of embeddings.
diff --git a/docs/images/gradio-screenshot-001.png b/docs/images/gradio-screenshot-001.png
new file mode 100644
index 0000000..fe70c8f
Binary files /dev/null and b/docs/images/gradio-screenshot-001.png differ