diff --git a/src/python/strelka/scanners/scan_pdf.py b/src/python/strelka/scanners/scan_pdf.py
index dddb1b49..bcbc0dd0 100644
--- a/src/python/strelka/scanners/scan_pdf.py
+++ b/src/python/strelka/scanners/scan_pdf.py
@@ -10,7 +10,10 @@
 
 # hide PyMuPDF warnings
 fitz.TOOLS.mupdf_display_errors(False)
-
+phone_numbers = re.compile(
+    "\+?(?:\d{1,2})?\s?\(?\d{3}\)?[\s.-]\d{3}[\s.-]\d{2,4}?\-?\d{2,4}?",
+    flags=0,
+)
 
 class ScanPdf(strelka.Scanner):
     """Collects metadata and extracts files from PDF files."""
@@ -55,6 +58,22 @@ def scan(self, data, file, options, expire_at):
             self.event['subject'] = reader.metadata['subject']
             self.event['title'] = reader.metadata['title']
             self.event['xrefs'] = reader.xref_length() - 1
+            
+            #collect phones
+            phones = []
+            for i in range(self.event["pages"]):
+                phones.extend(
+                    [
+                        re.sub("[^0-9]", "", x)
+                        for x in re.findall(
+                            phone_numbers,
+                            reader.get_page_text(i).replace(
+                                "\t", " "
+                            ),
+                        )
+                    ]
+                )
+            self.event["phones"] = list(set(phones))
 
             # iterate through xref objects
             for xref in range(1, reader.xref_length()):