diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py
index e130f2dbd..8299bfb16 100644
--- a/youtube_dl/extractor/npo.py
+++ b/youtube_dl/extractor/npo.py
@@ -221,13 +221,14 @@ class SchoolTVIE(NPOIE):
     def _real_extract(self, url):
         video_id = url.rstrip('/').split('/')[-1]
 
-        # TODO Find out how we could obtain this automatically
-        #      Otherwise this extractor might break each time SchoolTV deploys a new release
-        build_id = 'b7eHUzAVO7wHXCopYxQhV'
+        build_id = self._search_nextjs_data(
+            self._download_webpage(url, video_id),
+            video_id,
+        )['buildId']
 
         metadata_url = 'https://schooltv.nl/_next/data/' \
                        + build_id \
-                       + '/item/' \
+                       + '/video-item/' \
                        + video_id + '.json'
 
         metadata = self._download_json(metadata_url,
@@ -304,7 +305,7 @@ class VPROIE(NPOIE):
         formats = []
         for result in results:
             formats.extend(self._extract_formats_by_product_id(result, video_id))
-            break  # TODO find a better solution, VPRO pages can have multiple videos embedded
+            break
 
         if not formats:
             raise ExtractorError('Could not find a POMS product id in the provided URL, '