diff --git a/.gitignore b/.gitignore
index c1a76ff..8757073 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,4 +1,5 @@
 /target
 */tmp
 scratch.rs
-data_prep
\ No newline at end of file
+data_prep
+*.csv
\ No newline at end of file
diff --git a/Cargo.lock b/Cargo.lock
index a8b5987..341af2b 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -288,6 +288,27 @@ dependencies = [
  "cfg-if",
 ]
 
+[[package]]
+name = "csv"
+version = "1.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac574ff4d437a7b5ad237ef331c17ccca63c46479e5b5453eb8e10bb99a759fe"
+dependencies = [
+ "csv-core",
+ "itoa",
+ "ryu",
+ "serde",
+]
+
+[[package]]
+name = "csv-core"
+version = "0.1.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5efa2b3d7902f4b634a20cae3c9c4e6209dc4779feb6863329607560143efa70"
+dependencies = [
+ "memchr",
+]
+
 [[package]]
 name = "either"
 version = "1.9.0"
@@ -661,6 +682,7 @@ dependencies = [
  "clap",
  "crfsuite",
  "criterion",
+ "csv",
  "lazy_static",
  "unicode-normalization",
  "xml-rs",
diff --git a/Cargo.toml b/Cargo.toml
index ebb1fa8..c13a049 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -11,6 +11,7 @@ unicode-normalization = "0.1.22"
 clap = { version = "4.4.6", features = ["derive"] }
 crfsuite = "0.3.1"
 xml-rs = "0.8.19"
+csv = "1.3.0"
 
 [dev-dependencies]
 criterion = "0.5.1"
diff --git a/src/lib.rs b/src/lib.rs
index d78472d..ed48626 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -10,33 +10,35 @@ pub mod train;
 
 use abbreviations::{DIRECTIONALS, STREET_NAMES};
 
-pub enum Tag {
-    AddressNumberPrefix,
-    AddressNumber,
-    AddressNumberSuffix,
-    StreetNamePreModifier,
-    StreetNamePreDirectional,
-    StreetNamePreType,
-    StreetName,
-    StreetNamePostType,
-    StreetNamePostDirectional,
-    SubaddressType,
-    SubaddressIdentifier,
-    BuildingName,
-    OccupancyType,
-    OccupancyIdentifier,
-    CornerOf,
-    LandmarkName,
-    PlaceName,
-    StateName,
-    ZipCode,
-    USPSBoxType,
-    USPSBoxID,
-    USPSBoxGroupType,
-    USPSBoxGroupID,
-    IntersectionSeparator,
-    Recipient,
-    NotAddress,
+lazy_static! {
+    pub static ref TAGS: [&'static str; 26] = [
+        "AddressNumberPrefix",
+        "AddressNumber",
+        "AddressNumberSuffix",
+        "StreetNamePreModifier",
+        "StreetNamePreDirectional",
+        "StreetNamePreType",
+        "StreetName",
+        "StreetNamePostType",
+        "StreetNamePostDirectional",
+        "SubaddressType",
+        "SubaddressIdentifier",
+        "BuildingName",
+        "OccupancyType",
+        "OccupancyIdentifier",
+        "CornerOf",
+        "LandmarkName",
+        "PlaceName",
+        "StateName",
+        "ZipCode",
+        "USPSBoxType",
+        "USPSBoxID",
+        "USPSBoxGroupType",
+        "USPSBoxGroupID",
+        "IntersectionSeparator",
+        "Recipient",
+        "NotAddress",
+    ];
 }
 
 lazy_static! {
@@ -55,6 +57,20 @@ pub fn parse(address: &str) -> Vec<(String, String)> {
     zip_tokens_and_tags(tokens, tags)
 }
 
+pub fn parse_addresses(addresses: Vec<&str>) -> Vec<Vec<(String, String)>> {
+    addresses
+        .iter() // .iter is 42% faster than .par_iter()
+        .map(|address| parse(address))
+        .collect()
+}
+
+pub fn parse_addresses_from_txt(file_path: &str) -> Vec<Vec<(String, String)>> {
+    let raw_data = std::fs::read_to_string(file_path).unwrap();
+    let data: Vec<&str> = raw_data.lines().collect();
+
+    parse_addresses(data)
+}
+
 pub fn zip_tokens_and_tags(tokens: Vec<String>, tags: Vec<String>) -> Vec<(String, String)> {
     tokens.into_iter().zip(tags.into_iter()).collect()
 }
diff --git a/src/main.rs b/src/main.rs
index a3aed34..e49f361 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -1,6 +1,6 @@
 use clap::Parser;
-use us_addrs::parse;
 use us_addrs::train::train_model;
+use us_addrs::{parse, parse_addresses_from_txt, TAGS};
 
 // use std::path::PathBuf;
 
@@ -8,6 +8,7 @@ use us_addrs::train::train_model;
 enum USAddrsCli {
     Train(TrainArgs),
     Parse(ParseArgs),
+    ParseFile(ParseFileArgs),
 }
 
 #[derive(Parser)]
@@ -22,12 +23,43 @@ struct ParseArgs {
     address: String,
 }
 
+#[derive(Parser)]
+struct ParseFileArgs {
+    #[clap(short, long)]
+    file_path: String,
+    export_path: String,
+}
+
 fn main() {
     match USAddrsCli::parse() {
         USAddrsCli::Train(args) => match train_model(&args.export_path) {
             Ok(()) => println!("Trained model"),
             Err(e) => println!("Error training model: {}", e),
         },
+        USAddrsCli::ParseFile(args) => {
+            let parsed_addresses = parse_addresses_from_txt(&args.file_path);
+            // write as CSV with Tags as columns
+            let mut wtr = csv::Writer::from_path(&args.export_path).unwrap();
+
+            wtr.write_record(TAGS.iter()).unwrap();
+
+            for tagged_address in parsed_addresses {
+                let mut record = Vec::new();
+
+                for tag in TAGS.iter() {
+                    if let Some((token, _)) = tagged_address
+                        .iter()
+                        .find(|&(_, token_tag)| *token_tag == *tag)
+                    {
+                        record.push(token.to_string());
+                    } else {
+                        record.push("".to_string());
+                    }
+                }
+                wtr.write_record(&record).unwrap();
+            }
+            wtr.flush().unwrap();
+        }
         USAddrsCli::Parse(args) => {
             let parsed = parse(&args.address);
             println!("{:?}", parsed);