diff --git a/image_prediction/image_extractor/extractors/parsable.py b/image_prediction/image_extractor/extractors/parsable.py
index d115059..784a54f 100644
--- a/image_prediction/image_extractor/extractors/parsable.py
+++ b/image_prediction/image_extractor/extractors/parsable.py
@@ -8,9 +8,9 @@ from typing import List, Union
 import fitz
 import numpy as np
 from PIL import Image
-from funcy import merge, compose, rcompose, keep
-from pymonad.maybe import Maybe, Nothing, Just
-from pymonad.tools import curry
+from funcy import merge, compose, rcompose, keep, lfilter
+from pymonad.either import Right, Left, Either
+from pymonad.tools import curry, identity
 
 from image_prediction.exceptions import InvalidBox, BadXref
 from image_prediction.image_extractor.extractor import ImageExtractor, ImageMetadataPair
@@ -48,7 +48,7 @@ class ParsablePDFImageExtractor(ImageExtractor):
         metadata = extract_valid_metadata(self.doc, page)
 
         maybe_image_metadata_pairs = map(partial(metadatum_to_image_metadata_pair, self.doc), metadata)
-        image_metadata_pairs = [pair.value for pair in maybe_image_metadata_pairs if pair.is_just()]
+        image_metadata_pairs = keep(right, maybe_image_metadata_pairs)
         clear_caches()
 
         image_metadata_pairs = stitch_pairs(list(image_metadata_pairs), tolerance=self.tolerance)
@@ -56,6 +56,12 @@ class ParsablePDFImageExtractor(ImageExtractor):
         yield from image_metadata_pairs
 
 
+def right(pair: Either):
+    if pair.is_right():
+        return pair.either(identity, identity)
+    logger.warning(f"Skipping bad image. reason: {pair.either(identity, identity)}")
+
+
 def extract_pages(doc, page_range):
     page_range = range(page_range.start + 1, page_range.stop + 1)
     pages = map(doc.load_page, page_range)
@@ -63,14 +69,14 @@ def extract_pages(doc, page_range):
     yield from pages
 
 
-def validate_image(image: Image.Image) -> Maybe:
+def validate_image(image: Image.Image) -> Either:
     try:
         # TODO: stand-in heuristic for testing if image is valid => find cleaner solution (RED-5148)
         image.resize((100, 100)).convert("RGB")
-        return Just(image)
+        return Right(image)
     except (OSError, Exception):
         logger.warning(f"Invalid image encountered.")
-        return Nothing
+        return Left("Invalid image.")
 
 
 def extract_valid_metadata(doc: fitz.fitz.Document, page: fitz.fitz.Page):
@@ -82,9 +88,9 @@ def extract_valid_metadata(doc: fitz.fitz.Document, page: fitz.fitz.Page):
     )(page)
 
 
-def metadatum_to_image_metadata_pair(doc, metadatum: dict) -> Maybe:
+def metadatum_to_image_metadata_pair(doc, metadatum: dict) -> Either:
     maybe_image = xref_to_maybe_image(doc, metadatum[Info.XREF]).bind(validate_image)
-    maybe_image_metadata_pair = make_maybe_image_metadata_pair(maybe_image, Just(metadatum))
+    maybe_image_metadata_pair = make_maybe_image_metadata_pair(maybe_image, Right(metadatum))
     return maybe_image_metadata_pair
 
 
@@ -111,21 +117,22 @@ def get_metadata_for_images_on_page(page: fitz.Page):
 
 
 @lru_cache(maxsize=None)
-def xref_to_maybe_image(doc, xref) -> Maybe:
+def xref_to_maybe_image(doc, xref) -> Either:
     try:
-        return Just(extract_image(doc, xref))
+        return Right(extract_image(doc, xref))
     except BadXref:
-        return Nothing
+        return Left("Bad xref.")
 
 
-def make_maybe_image_metadata_pair(image: Maybe, metadata: Maybe):
+def make_maybe_image_metadata_pair(image: Either, metadata: Either):
     # haskell.org/tutorial/monads.html
     # (>>)             :: m a -> m b -> m b
-    return Just(make_image_metadata_pair).amap(image).amap(metadata)
+    return Right(make_image_metadata_pair).amap(image).amap(metadata)
+    # TODO: Somehow metadata needs to be added to Lefts for logging the reference to the invalid image
 
 
 @curry(2)
-def make_image_metadata_pair(image: Image.Image, metadatum: dict) -> Just:
+def make_image_metadata_pair(image: Image.Image, metadatum: dict) -> ImageMetadataPair:
     return ImageMetadataPair(image, metadatum)