diff --git a/backend/backend/__init__.py b/backend/backend/__init__.py
new file mode 100644
index 00000000..3b69d0d0
--- /dev/null
+++ b/backend/backend/__init__.py
@@ -0,0 +1,8 @@
+from logging import INFO
+
+from config.logging import config_logger
+
+config_logger(
+    logger_name=__name__,
+    log_level=INFO,
+)
diff --git a/backend/backend/analyze_sleep.py b/backend/backend/analyze_sleep.py
index bbe6c8f1..b2a60e04 100644
--- a/backend/backend/analyze_sleep.py
+++ b/backend/backend/analyze_sleep.py
@@ -1,5 +1,6 @@
 import json
 import falcon
+import logging
 
 from backend.request import ClassificationRequest
 from backend.response import ClassificationResponse
@@ -10,9 +11,12 @@
 from classification.model import SleepStagesClassifier
 from classification.features.preprocessing import preprocess
 
+_logger = logging.getLogger(__name__)
+
 
 class AnalyzeSleep:
     def __init__(self):
+        _logger.info("Initializing sleep stage classifier.")
         self.sleep_stage_classifier = SleepStagesClassifier()
 
     @staticmethod
@@ -55,6 +59,7 @@ def on_post(self, request, response):
         }
         """
 
+        _logger.info("Validating and parsing form fields and EEG file")
         try:
             form_data, file = self._parse_form(request.get_media())
             raw_array = get_raw_array(file)
@@ -67,16 +72,27 @@ def on_post(self, request, response):
                 raw_eeg=raw_array,
             )
         except (KeyError, ValueError, ClassificationError):
+            _logger.warn(
+                "An error occured when validating and parsing form fields. "
+                "Request parameters are either missing or invalid."
+            )
             response.status = falcon.HTTP_400
             response.content_type = falcon.MEDIA_TEXT
             response.body = 'Missing or invalid request parameters'
             return
 
+        _logger.info("Preprocessing of raw EEG data.")
         preprocessed_epochs = preprocess(classification_request)
+
+        _logger.info("Prediction of EEG data to sleep stages.")
         predictions = self.sleep_stage_classifier.predict(preprocessed_epochs, classification_request)
+
+        _logger.info("Computations of visualisation data & of sleep report metrics...")
         spectrogram_generator = SpectrogramGenerator(preprocessed_epochs)
         classification_response = ClassificationResponse(
             classification_request, predictions, spectrogram_generator.generate()
         )
 
         response.body = json.dumps(classification_response.response)
+
+        _logger.info("Request completed")
diff --git a/backend/backend/app.py b/backend/backend/app.py
index 9b272584..b7121492 100644
--- a/backend/backend/app.py
+++ b/backend/backend/app.py
@@ -1,8 +1,11 @@
 import falcon
+import logging
 
 from backend.ping import Ping
 from backend.analyze_sleep import AnalyzeSleep
 
+_logger = logging.getLogger(__name__)
+
 
 def App():
     app = falcon.App(cors_enable=True)
@@ -13,4 +16,8 @@ def App():
     analyze = AnalyzeSleep()
     app.add_route('/analyze-sleep', analyze)
 
+    _logger.info(
+        'Completed local server initialization. '
+        'Please go back to your browser in order to submit your sleep EEG file. '
+    )
     return app
diff --git a/backend/backend/request.py b/backend/backend/request.py
index 9e872cc4..d020af43 100644
--- a/backend/backend/request.py
+++ b/backend/backend/request.py
@@ -1,3 +1,4 @@
+import logging
 
 from classification.config.constants import EPOCH_DURATION
 from classification.config.constants import (
@@ -10,6 +11,8 @@
     ClassificationError,
 )
 
+_logger = logging.getLogger(__name__)
+
 
 class ClassificationRequest():
     def __init__(self, sex, age, stream_start, bedtime, wakeup, raw_eeg, stream_duration=None):
@@ -56,21 +59,27 @@ def _validate_timestamps(self):
             and has_got_out_of_bed_after_in_bed
             and has_respected_minimum_bed_time
         ):
+            _logger.warn("Received timestamps are invalid.")
             raise TimestampsError()
 
     def _validate_file_with_timestamps(self):
         has_raw_respected_minimum_file_size = self.raw_eeg.times[-1] > FILE_MINIMUM_DURATION
 
         if not has_raw_respected_minimum_file_size:
+            _logger.warn(f"Uploaded file must at least have {FILE_MINIMUM_DURATION} seconds of data.")
             raise FileSizeError()
 
         is_raw_at_least_as_long_as_out_of_bed = self.raw_eeg.times[-1] >= self.out_of_bed_seconds
 
         if not is_raw_at_least_as_long_as_out_of_bed:
+            _logger.warn(
+                "Uploaded file must at least last the time between the start of the "
+                f"stream and out of bed time, which is {self.out_of_bed_seconds} seconds.")
             raise TimestampsError()
 
     def _validate_age(self):
         is_in_accepted_range = ACCEPTED_AGE_RANGE[0] <= int(self.age) <= ACCEPTED_AGE_RANGE[1]
 
         if not(is_in_accepted_range):
+            _logger.warn(f"Age must be in the following range: {ACCEPTED_AGE_RANGE}")
             raise ClassificationError('invalid age')
diff --git a/backend/backend/spectrogram_generator.py b/backend/backend/spectrogram_generator.py
index d4f66296..30057e0f 100644
--- a/backend/backend/spectrogram_generator.py
+++ b/backend/backend/spectrogram_generator.py
@@ -20,6 +20,7 @@ def generate(self):
             self.epochs,
             fmin=self.spectrogram_min_freq,
             fmax=self.spectrogram_max_freq,
+            verbose=False,
         )
         psds_db = self._convert_amplitudes_to_decibel(psds)
 
diff --git a/backend/classification/__init__.py b/backend/classification/__init__.py
new file mode 100644
index 00000000..59d9b544
--- /dev/null
+++ b/backend/classification/__init__.py
@@ -0,0 +1,9 @@
+from logging import INFO
+
+from config.logging import config_logger
+
+config_logger(
+    logger_name=__name__,
+    log_level=INFO,
+    message_sublevel=True,
+)
diff --git a/backend/classification/features/extraction.py b/backend/classification/features/extraction.py
index e53667e9..0ed91efc 100644
--- a/backend/classification/features/extraction.py
+++ b/backend/classification/features/extraction.py
@@ -31,11 +31,6 @@ def get_eeg_features(epochs, in_bed_seconds, out_of_bed_seconds):
 
         features.append(channel_features)
 
-        print(
-            f"Done extracting {channel_features.shape[1]} features "
-            f"on {channel_features.shape[0]} epochs for {channel}\n"
-        )
-
     return np.hstack(tuple(features))
 
 
diff --git a/backend/classification/features/pipeline/utils.py b/backend/classification/features/pipeline/utils.py
index ca2ba215..ecbaebce 100644
--- a/backend/classification/features/pipeline/utils.py
+++ b/backend/classification/features/pipeline/utils.py
@@ -16,7 +16,7 @@ def get_psds_from_epochs(epochs):
     --------
     psds with associated frequencies calculated with the welch method.
     """
-    psds, freqs = psd_welch(epochs, fmin=0.5, fmax=30.)
+    psds, freqs = psd_welch(epochs, fmin=0.5, fmax=30., verbose=False)
     return psds, freqs
 
 
diff --git a/backend/classification/features/preprocessing.py b/backend/classification/features/preprocessing.py
index eb1b82cd..63468635 100644
--- a/backend/classification/features/preprocessing.py
+++ b/backend/classification/features/preprocessing.py
@@ -1,3 +1,4 @@
+import logging
 import mne
 from scipy.signal import cheby1
 
@@ -11,6 +12,8 @@
     HIGH_PASS_MAX_RIPPLE_DB,
 )
 
+_logger = logging.getLogger(__name__)
+
 
 def preprocess(classification_request):
     """Returns preprocessed epochs of the specified channel
@@ -20,13 +23,20 @@ def preprocess(classification_request):
     """
     raw_data = classification_request.raw_eeg.copy()
 
+    _logger.info("Cropping data from bed time to out of bed time.")
     raw_data = _crop_raw_data(
         raw_data,
         classification_request.in_bed_seconds,
         classification_request.out_of_bed_seconds,
     )
+
+    _logger.info(f"Applying high pass filter at {DATASET_HIGH_PASS_FREQ}Hz.")
     raw_data = _apply_high_pass_filter(raw_data)
+
+    _logger.info(f"Resampling data at the dataset's sampling rate of {DATASET_SAMPLE_RATE} Hz.")
     raw_data = raw_data.resample(DATASET_SAMPLE_RATE)
+
+    _logger.info(f"Epoching data with a {EPOCH_DURATION} seconds duration.")
     raw_data = _convert_to_epochs(raw_data)
 
     return raw_data
diff --git a/backend/classification/load_model.py b/backend/classification/load_model.py
index f1426be5..107100b5 100644
--- a/backend/classification/load_model.py
+++ b/backend/classification/load_model.py
@@ -1,4 +1,5 @@
 from datetime import datetime
+import logging
 from os import path, makedirs
 from pathlib import Path
 import re
@@ -12,6 +13,10 @@
 
 from classification.config.constants import HiddenMarkovModelProbability
 
+
+_logger = logging.getLogger(__name__)
+
+
 SCRIPT_PATH = Path(path.realpath(sys.argv[0])).parent
 
 BUCKET_NAME = 'polydodo'
@@ -50,9 +55,13 @@ def _has_latest_object(filename, local_path):
 
 def load_model():
     if not path.exists(MODEL_PATH) or not _has_latest_object(MODEL_FILENAME, MODEL_PATH):
-        print("Downloading latest model...")
+        _logger.info(
+            "Downloading latest sleep stage classification model... "
+            f"This could take a few minutes. (storing it at {MODEL_PATH})"
+        )
         _download_file(MODEL_URL, MODEL_PATH)
-    print("Loading model...")
+
+    _logger.info(f"Loading latest sleep stage classification model... (from {MODEL_PATH})")
     return onnxruntime.InferenceSession(str(MODEL_PATH))
 
 
@@ -67,8 +76,10 @@ def load_hmm():
         model_path = SCRIPT_PATH / HMM_FOLDER / hmm_file
 
         if not path.exists(model_path) or not _has_latest_object(hmm_file, model_path):
+            _logger.info(f"Downloading postprocessing model... (storing it at {model_path})")
             _download_file(url=f"{BUCKET_URL}/{hmm_file}", output=model_path)
 
+        _logger.info(f"Loading postprocessing model... (from {model_path})")
         hmm_matrices[hmm_probability.name] = np.load(str(model_path))
 
     return hmm_matrices
diff --git a/backend/classification/model.py b/backend/classification/model.py
index eb889153..a67e38d7 100644
--- a/backend/classification/model.py
+++ b/backend/classification/model.py
@@ -1,9 +1,12 @@
 """defines models which predict sleep stages based off EEG signals"""
+import logging
 
 from classification.features import get_features
 from classification.postprocessor import get_hmm_model
 from classification.load_model import load_model, load_hmm
 
+_logger = logging.getLogger(__name__)
+
 
 class SleepStagesClassifier():
     def __init__(self):
@@ -21,12 +24,14 @@ def predict(self, epochs, request):
         - request: instance of ClassificationRequest
         Returns: array of predicted sleep stages
         """
-
+        _logger.info("Extracting features...")
         features = get_features(epochs, request)
+        _logger.info(f"Finished extracting {features.shape[1]} features over {features.shape[0]} epochs.")
 
-        print(features, features.shape)
-
+        _logger.info("Classifying sleep stages from extracted features...")
         predictions = self._get_predictions(features)
+
+        _logger.info("Applying postprocessing step to the resulted sleep stages...")
         predictions = self._get_postprocessed_predictions(predictions)
 
         return predictions
diff --git a/backend/classification/parser/__init__.py b/backend/classification/parser/__init__.py
index f91a4116..f6f5b6f8 100644
--- a/backend/classification/parser/__init__.py
+++ b/backend/classification/parser/__init__.py
@@ -10,6 +10,8 @@
 The Cyton board logging format is also described here:
 [https://docs.openbci.com/docs/02Cyton/CytonSDCard#data-logging-format]
 """
+import logging
+
 from mne import create_info
 from mne.io import RawArray
 
@@ -18,6 +20,9 @@
 from classification.parser.sample_rate import detect_sample_rate
 
 
+_logger = logging.getLogger(__name__)
+
+
 def get_raw_array(file):
     """Converts a file following a logging format into a mne.RawArray
     Input:
@@ -27,15 +32,15 @@ def get_raw_array(file):
     """
 
     filetype = detect_file_type(file)
-    print(f"""
-    Detected {filetype.name} format.
-    """)
 
     sample_rate = detect_sample_rate(file, filetype)
-    print(f"""
-    Detected {sample_rate}Hz sample rate.
-    """)
 
+    _logger.info(
+        f"EEG data has been detected to be in the {filetype.name} format "
+        f"and has a {sample_rate}Hz sample rate."
+    )
+
+    _logger.info("Parsing EEG file to a mne.RawArray object...")
     eeg_raw = filetype.parser(file)
 
     raw_object = RawArray(
@@ -47,12 +52,12 @@ def get_raw_array(file):
         verbose=False,
     )
 
-    print(f"""
-        First sample values: {raw_object[:, 0]}
-        Second sample values: {raw_object[:, 1]}
-        Number of samples: {raw_object.n_times}
-        Duration of signal (h): {raw_object.n_times / (3600 * sample_rate)}
-        Channel names: {raw_object.ch_names}
-    """)
+    _logger.info(
+        f"Finished converting EEG file to mne.RawArray object "
+        f"with the first sample being {*(raw_object[:, 0][0]),}, "
+        f"with {raw_object.n_times} samples, "
+        f"with a {raw_object.n_times / (3600 * sample_rate):.2f} hours duration and "
+        f"with channels named {raw_object.ch_names}."
+    )
 
     return raw_object
diff --git a/backend/config/logging.py b/backend/config/logging.py
new file mode 100644
index 00000000..0f75b078
--- /dev/null
+++ b/backend/config/logging.py
@@ -0,0 +1,20 @@
+import logging
+import sys
+
+STD_OUTPUT_FORMAT = "[%(asctime)s - %(levelname)s]:\t%(message)s (%(name)s)"
+SUBLEVEL_OUTPUT_FORMAT = "[%(asctime)s - %(levelname)s]:\t\t%(message)s (%(name)s)"
+
+
+def config_logger(logger_name, log_level, message_sublevel=False):
+    """Configures logging with std output"""
+    logger = logging.getLogger(logger_name)
+    logger.setLevel(log_level)
+    logger.addHandler(_get_console_handler(message_sublevel))
+    logger.propagate = False
+
+
+def _get_console_handler(message_sublevel):
+    console_handler = logging.StreamHandler(sys.stdout)
+    formatter = SUBLEVEL_OUTPUT_FORMAT if message_sublevel else STD_OUTPUT_FORMAT
+    console_handler.setFormatter(logging.Formatter(formatter))
+    return console_handler