diff --git a/en/develop/freqai/index.html b/en/develop/freqai/index.html
index 736204c65..cdd894132 100644
--- a/en/develop/freqai/index.html
+++ b/en/develop/freqai/index.html
@@ -1810,7 +1810,7 @@
 <li><strong>Extensibility</strong> - The generalized and robust architecture allows for incorporating any <a href="../freqai-configuration/#using-different-prediction-models">machine learning library/method</a> available in Python. Eight examples are currently available, including classifiers, regressors, and a convolutional neural network</li>
 <li><strong>Smart outlier removal</strong> - Remove outliers from training and prediction data sets using a variety of <a href="../freqai-feature-engineering/#outlier-detection">outlier detection techniques</a></li>
 <li><strong>Crash resilience</strong> - Store trained models to disk to make reloading from a crash fast and easy, and <a href="../freqai-running/#purging-old-model-data">purge obsolete files</a> for sustained dry/live runs</li>
-<li><strong>Automatic data normalization</strong> - <a href="../freqai-feature-engineering/#feature-normalization">Normalize the data</a> in a smart and statistically safe way</li>
+<li><strong>Automatic data normalization</strong> - <a href="../freqai-feature-engineering/#building-the-data-pipeline">Normalize the data</a> in a smart and statistically safe way</li>
 <li><strong>Automatic data download</strong> - Compute timeranges for data downloads and update historic data (in live deployments)</li>
 <li><strong>Cleaning of incoming data</strong> - Handle NaNs safely before training and model inferencing</li>
 <li><strong>Dimensionality reduction</strong> - Reduce the size of the training data via <a href="../freqai-feature-engineering/#data-dimensionality-reduction-with-principal-component-analysis">Principal Component Analysis</a></li>