diff --git a/examples/create-database.rs b/examples/create-database.rs
index b66bf7f5b..e7a8e72e4 100644
--- a/examples/create-database.rs
+++ b/examples/create-database.rs
@@ -10,6 +10,7 @@ use meilidb::database::schema::{Schema, SchemaBuilder, STORED, INDEXED};
 use meilidb::database::update::PositiveUpdateBuilder;
 use meilidb::tokenizer::DefaultBuilder;
 use meilidb::database::Database;
+use meilidb::DocumentId;
 
 #[derive(Debug, StructOpt)]
 pub struct Opt {
@@ -67,7 +68,7 @@ fn index(schema: Schema, database_path: &Path, csv_data_path: &Path) -> Result<D
             }
         };
 
-        let document_id = calculate_hash(&document.id);
+        let document_id = DocumentId(calculate_hash(&document.id));
         update.update(document_id, &document).unwrap();
     }
 
diff --git a/src/data/doc_indexes.rs b/src/data/doc_indexes.rs
index 832740e4e..5d451f83c 100644
--- a/src/data/doc_indexes.rs
+++ b/src/data/doc_indexes.rs
@@ -156,13 +156,16 @@ unsafe fn into_u8_slice<T>(slice: &[T]) -> &[u8] {
 #[cfg(test)]
 mod tests {
     use super::*;
+
     use std::error::Error;
 
+    use crate::DocumentId;
+
     #[test]
     fn builder_serialize_deserialize() -> Result<(), Box<Error>> {
-        let a = DocIndex { document_id: 0, attribute: 3, attribute_index: 11 };
-        let b = DocIndex { document_id: 1, attribute: 4, attribute_index: 21 };
-        let c = DocIndex { document_id: 2, attribute: 8, attribute_index: 2 };
+        let a = DocIndex { document_id: DocumentId(0), attribute: 3, attribute_index: 11 };
+        let b = DocIndex { document_id: DocumentId(1), attribute: 4, attribute_index: 21 };
+        let c = DocIndex { document_id: DocumentId(2), attribute: 8, attribute_index: 2 };
 
         let mut builder = DocIndexesBuilder::memory();
 
@@ -183,9 +186,9 @@ mod tests {
 
     #[test]
     fn serialize_deserialize() -> Result<(), Box<Error>> {
-        let a = DocIndex { document_id: 0, attribute: 3, attribute_index: 11 };
-        let b = DocIndex { document_id: 1, attribute: 4, attribute_index: 21 };
-        let c = DocIndex { document_id: 2, attribute: 8, attribute_index: 2 };
+        let a = DocIndex { document_id: DocumentId(0), attribute: 3, attribute_index: 11 };
+        let b = DocIndex { document_id: DocumentId(1), attribute: 4, attribute_index: 21 };
+        let c = DocIndex { document_id: DocumentId(2), attribute: 8, attribute_index: 2 };
 
         let mut builder = DocIndexesBuilder::memory();
 
diff --git a/src/database/blob/positive/blob.rs b/src/database/blob/positive/blob.rs
index fd6f2251d..3687bc1bb 100644
--- a/src/database/blob/positive/blob.rs
+++ b/src/database/blob/positive/blob.rs
@@ -201,13 +201,16 @@ impl<W: Write, X: Write> PositiveBlobBuilder<W, X> {
 #[cfg(test)]
 mod tests {
     use super::*;
+
     use std::error::Error;
 
+    use crate::DocumentId;
+
     #[test]
     fn serialize_deserialize() -> Result<(), Box<Error>> {
-        let a = DocIndex { document_id: 0, attribute: 3, attribute_index: 11 };
-        let b = DocIndex { document_id: 1, attribute: 4, attribute_index: 21 };
-        let c = DocIndex { document_id: 2, attribute: 8, attribute_index: 2 };
+        let a = DocIndex { document_id: DocumentId(0), attribute: 3, attribute_index: 11 };
+        let b = DocIndex { document_id: DocumentId(1), attribute: 4, attribute_index: 21 };
+        let c = DocIndex { document_id: DocumentId(2), attribute: 8, attribute_index: 2 };
 
         let mut builder = PositiveBlobBuilder::memory();
 
@@ -228,9 +231,9 @@ mod tests {
 
     #[test]
     fn serde_serialize_deserialize() -> Result<(), Box<Error>> {
-        let a = DocIndex { document_id: 0, attribute: 3, attribute_index: 11 };
-        let b = DocIndex { document_id: 1, attribute: 4, attribute_index: 21 };
-        let c = DocIndex { document_id: 2, attribute: 8, attribute_index: 2 };
+        let a = DocIndex { document_id: DocumentId(0), attribute: 3, attribute_index: 11 };
+        let b = DocIndex { document_id: DocumentId(1), attribute: 4, attribute_index: 21 };
+        let c = DocIndex { document_id: DocumentId(2), attribute: 8, attribute_index: 2 };
 
         let mut builder = PositiveBlobBuilder::memory();
 
diff --git a/src/database/database_view.rs b/src/database/database_view.rs
index f43d65439..c8eed37c6 100644
--- a/src/database/database_view.rs
+++ b/src/database/database_view.rs
@@ -100,7 +100,7 @@ where D: Deref<Target=DB>
 {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         let mut options = ReadOptions::new();
-        let lower = DocumentKey::new(0);
+        let lower = DocumentKey::new(DocumentId(0));
         options.set_iterate_lower_bound(lower.as_ref());
 
         let mut iter = self.snapshot.iter_opt(options);
diff --git a/src/database/document_key.rs b/src/database/document_key.rs
index 529bc1b75..9104df5f6 100644
--- a/src/database/document_key.rs
+++ b/src/database/document_key.rs
@@ -19,7 +19,7 @@ impl DocumentKey {
 
         let mut wtr = Cursor::new(&mut buffer[..]);
         wtr.write_all(b"doc-").unwrap();
-        wtr.write_u64::<NativeEndian>(id).unwrap();
+        wtr.write_u64::<NativeEndian>(id.0).unwrap();
 
         DocumentKey(buffer)
     }
@@ -43,7 +43,8 @@ impl DocumentKey {
     }
 
     pub fn document_id(&self) -> DocumentId {
-        (&self.0[4..]).read_u64::<NativeEndian>().unwrap()
+        let id = (&self.0[4..]).read_u64::<NativeEndian>().unwrap();
+        DocumentId(id)
     }
 }
 
@@ -88,7 +89,8 @@ impl DocumentKeyAttr {
     }
 
     pub fn document_id(&self) -> DocumentId {
-        (&self.0[4..]).read_u64::<NativeEndian>().unwrap()
+        let id = (&self.0[4..]).read_u64::<NativeEndian>().unwrap();
+        DocumentId(id)
     }
 
     pub fn attribute(&self) -> SchemaAttr {
diff --git a/src/database/mod.rs b/src/database/mod.rs
index 99fc5228e..2351c658c 100644
--- a/src/database/mod.rs
+++ b/src/database/mod.rs
@@ -194,6 +194,7 @@ mod tests {
     use serde_derive::{Serialize, Deserialize};
     use tempfile::tempdir;
 
+    use crate::DocumentId;
     use crate::tokenizer::DefaultBuilder;
     use crate::database::update::PositiveUpdateBuilder;
     use crate::database::schema::{SchemaBuilder, STORED, INDEXED};
@@ -238,8 +239,8 @@ mod tests {
         let mut update = {
             let mut builder = PositiveUpdateBuilder::new(update_path, schema, tokenizer_builder);
 
-            builder.update(0, &doc0).unwrap();
-            builder.update(1, &doc1).unwrap();
+            builder.update(DocumentId(0), &doc0).unwrap();
+            builder.update(DocumentId(1), &doc1).unwrap();
 
             builder.build()?
         };
@@ -248,8 +249,8 @@ mod tests {
         database.ingest_update_file(update)?;
         let view = database.view();
 
-        let de_doc0: SimpleDoc = view.retrieve_document(0)?;
-        let de_doc1: SimpleDoc = view.retrieve_document(1)?;
+        let de_doc0: SimpleDoc = view.retrieve_document(DocumentId(0))?;
+        let de_doc1: SimpleDoc = view.retrieve_document(DocumentId(1))?;
 
         assert_eq!(doc0, de_doc0);
         assert_eq!(doc1, de_doc1);
diff --git a/src/database/update/negative/unordered_builder.rs b/src/database/update/negative/unordered_builder.rs
index b73ecd2e3..4278e6974 100644
--- a/src/database/update/negative/unordered_builder.rs
+++ b/src/database/update/negative/unordered_builder.rs
@@ -30,7 +30,7 @@ impl<W: io::Write> UnorderedNegativeBlobBuilder<W> {
 
     pub fn into_inner(mut self) -> io::Result<W> {
         for id in self.doc_ids {
-            self.wrt.write_u64::<NativeEndian>(id)?;
+            self.wrt.write_u64::<NativeEndian>(id.0)?;
         }
         Ok(self.wrt)
     }
diff --git a/src/lib.rs b/src/lib.rs
index 655a234ca..d95dcc2ae 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -11,7 +11,12 @@ pub use rocksdb;
 pub use self::tokenizer::Tokenizer;
 pub use self::common_words::CommonWords;
 
-pub type DocumentId = u64;
+/// Represent an internally generated document unique identifier.
+///
+/// It is used to inform the database the document you want to deserialize.
+/// Helpful for custom ranking.
+#[derive(Debug, Copy, Clone, Eq, PartialEq, PartialOrd, Ord, Hash)]
+pub struct DocumentId(pub u64);
 
 /// This structure represent the position of a word
 /// in a document and its attributes.
diff --git a/src/rank/criterion/sum_of_typos.rs b/src/rank/criterion/sum_of_typos.rs
index a7074bd86..3af339233 100644
--- a/src/rank/criterion/sum_of_typos.rs
+++ b/src/rank/criterion/sum_of_typos.rs
@@ -44,6 +44,8 @@ where D: Deref<Target=DB>
 mod tests {
     use super::*;
 
+    use crate::DocumentId;
+
     // typing: "Geox CEO"
     //
     // doc0: "Geox SpA: CEO and Executive"
@@ -56,7 +58,7 @@ mod tests {
                 Match { query_index: 1, distance: 0, attribute: 0, attribute_index: 2, is_exact: false },
             ];
             Document {
-                id: 0,
+                id: DocumentId(0),
                 matches: matches,
             }
         };
@@ -67,7 +69,7 @@ mod tests {
                 Match { query_index: 1, distance: 0, attribute: 0, attribute_index: 2, is_exact: false },
             ];
             Document {
-                id: 1,
+                id: DocumentId(1),
                 matches: matches,
             }
         };
@@ -89,7 +91,7 @@ mod tests {
                 Match { query_index: 1, distance: 0, attribute: 0, attribute_index: 1, is_exact: false },
             ];
             Document {
-                id: 0,
+                id: DocumentId(0),
                 matches: matches,
             }
         };
@@ -99,7 +101,7 @@ mod tests {
                 Match { query_index: 0, distance: 0, attribute: 0, attribute_index: 0, is_exact: false },
             ];
             Document {
-                id: 1,
+                id: DocumentId(1),
                 matches: matches,
             }
         };
@@ -121,7 +123,7 @@ mod tests {
                 Match { query_index: 1, distance: 1, attribute: 0, attribute_index: 1, is_exact: false },
             ];
             Document {
-                id: 0,
+                id: DocumentId(0),
                 matches: matches,
             }
         };
@@ -131,7 +133,7 @@ mod tests {
                 Match { query_index: 0, distance: 0, attribute: 0, attribute_index: 0, is_exact: false },
             ];
             Document {
-                id: 1,
+                id: DocumentId(1),
                 matches: matches,
             }
         };