diff --git a/poetry.lock b/poetry.lock
index 0e8059e..0138151 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -2217,4 +2217,4 @@ watchmedo = ["PyYAML (>=3.10)"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.10"
-content-hash = "ea702bbac9344368a96b5da6a28ca66ebbd3be7ad17daf7e80fd1d34025b618e"
+content-hash = "f69f2921f19115d4b57493f6208e29aca65b0fd78dec3f98792cf48bad590dcf"
diff --git a/pyproject.toml b/pyproject.toml
index c6228bf..051814a 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -49,6 +49,7 @@ matplotlib = "^3.8.3"
 matplotlib-label-lines = "^0.7.0"
 seaborn = "^0.13.2"
 plotext = "^5.2.8"
+pandas = "^2.2.1"
 
 [tool.bandit]
 exclude_dirs = [".venv/"]
diff --git a/src/hpc_multibench/plot/export_data.py b/src/hpc_multibench/plot/export_data.py
new file mode 100755
index 0000000..7385ffc
--- /dev/null
+++ b/src/hpc_multibench/plot/export_data.py
@@ -0,0 +1,42 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""A set of functions to export the results of a test bench run."""
+
+import pandas as pd
+
+from hpc_multibench.plot.plot_data import split_metric_uncertainty
+from hpc_multibench.run_configuration import RunConfiguration
+from hpc_multibench.uncertainties import UFloat
+from hpc_multibench.yaml_model import ExportModel
+
+
+def export_data(
+    plot: ExportModel,
+    all_metrics: list[tuple[RunConfiguration, dict[str, str | UFloat]]],
+) -> None:
+    """Construct and export a pandas data frame from the metrics."""
+    df_data: dict[str, list[float | str]] = {}
+    for run_configuration, metrics in all_metrics:
+        row_data: dict[str, float | str] = {"Run configuration": run_configuration.name}
+
+        for metric in metrics:
+            value, error = split_metric_uncertainty(metrics, metric)
+            row_data[metric] = value
+            if error is not None:
+                row_data[f"{metric} error"] = error
+
+        for column, cell in row_data.items():
+            if column not in df_data:
+                df_data[column] = []
+            df_data[column].append(cell)
+
+    export_df = pd.DataFrame(df_data)
+
+    if plot.export_path is None:
+        print(export_df.to_string())
+    elif plot.export_format == "csv":
+        export_df.to_csv(plot.export_path)
+    else:
+        raise NotImplementedError(
+            f"Export format '{plot.export_format}' not supported!"
+        )
diff --git a/src/hpc_multibench/plot/plot_matplotlib.py b/src/hpc_multibench/plot/plot_matplotlib.py
index 3ded549..e7f8cdb 100755
--- a/src/hpc_multibench/plot/plot_matplotlib.py
+++ b/src/hpc_multibench/plot/plot_matplotlib.py
@@ -14,9 +14,6 @@
 from hpc_multibench.uncertainties import UFloat
 from hpc_multibench.yaml_model import BarChartModel, LinePlotModel, RooflinePlotModel
 
-# from labellines import labelLines
-
-
 sns.set_theme()
 
 
@@ -78,6 +75,7 @@ def draw_roofline_plot(
         plt.plot(x, y, label=label)
     for label, (x, y) in roofline.compute_bound_ceilings.items():
         plt.plot(x, y, label=label)
+    # from labellines import labelLines
     # for ax in plt.gcf().axes:
     #     labelLines(ax.get_lines())
     for name, (x_point, y_point, x_err, y_err) in data.items():
diff --git a/src/hpc_multibench/test_bench.py b/src/hpc_multibench/test_bench.py
index 7fe153f..a2b065c 100755
--- a/src/hpc_multibench/test_bench.py
+++ b/src/hpc_multibench/test_bench.py
@@ -18,6 +18,7 @@
 
 from typing_extensions import Self
 
+from hpc_multibench.plot.export_data import export_data
 from hpc_multibench.plot.plot_matplotlib import (
     draw_bar_chart,
     draw_line_plot,
@@ -82,7 +83,7 @@ def __init__(
         # Validate that all configurations named in the test bench are defined
         # in the test plan
         for run_configuration_name in bench_model.run_configurations:
-            if run_configuration_name not in self.run_configuration_models.keys():
+            if run_configuration_name not in self.run_configuration_models:
                 raise RuntimeError(
                     f"'{run_configuration_name}' not in list of"
                     " defined run configurations!"
@@ -100,10 +101,7 @@ def instantiations(self) -> list[dict[str, Any]]:
             (
                 [[(key, value)] for value in values]
                 if isinstance(key, str)
-                else [
-                    [(k, v) for k, v in zip(key, setting, strict=True)]
-                    for setting in values
-                ]
+                else [list(zip(key, setting, strict=True)) for setting in values]
             )
             for key, values in self.bench_model.matrix.items()
         ]
@@ -424,3 +422,6 @@ def report(self) -> None:
 
         for roofline_plot in self.bench_model.analysis.roofline_plots:
             draw_roofline_plot(roofline_plot, aggregated_metrics)
+
+        for export_schema in self.bench_model.analysis.data_exports:
+            export_data(export_schema, aggregated_metrics)
diff --git a/src/hpc_multibench/yaml_model.py b/src/hpc_multibench/yaml_model.py
index dd74c30..c6c3dad 100755
--- a/src/hpc_multibench/yaml_model.py
+++ b/src/hpc_multibench/yaml_model.py
@@ -82,11 +82,12 @@ class RooflinePlotModel(BaseModel):
     ert_json: Path
 
 
-# class ExportModel(BaseModel):
-#     """A Pydantic model for a exporting a set of metrics."""
+class ExportModel(BaseModel):
+    """A Pydantic model for a exporting a set of metrics."""
 
-#     metrics: list[str]
-#     filename: Path | None = None
+    export_path: Path | None
+    export_format: str = "csv"
+    # metrics: list[str] | None = None
 
 
 class AnalysisModel(BaseModel):
@@ -98,7 +99,7 @@ class AnalysisModel(BaseModel):
     line_plots: list[LinePlotModel] = []
     bar_charts: list[BarChartModel] = []
     roofline_plots: list[RooflinePlotModel] = []
-    # exports: list[BarChartModel] = []
+    data_exports: list[ExportModel] = []
 
 
 class RerunModel(BaseModel):