Navigate
Home Compendium
Exercises
Tidyverse packages
dplyr ggplot2 tidyr stringr lubridate purrr forcats readr broom Tidyverse R for Data Science
Deep dives
dplyr Joins dplyr Window Functions dplyr group_by ggplot2 Themes ggplot2 Facets ggplot2 Color Scales ggplot2 Bar Charts ggplot2 Heatmaps tidyr Pivot tidyr Nest/Unnest Regex
Wrangling & EDA
Data Wrangling Data Cleaning EDA Data Visualization data.table dbplyr / SQL Web Scraping API Calls
Statistics
Hypothesis Testing Linear Regression Logistic Regression Correlation Probability Distributions
Machine Learning
Machine Learning tidymodels Random Forest XGBoost Clustering PCA Cross-Validation
Time Series
Time Series ARIMA
By Industry
Finance Marketing Analytics Healthcare Biostatistics Genomics Sports Analytics A/B Testing
Reporting & Apps
Shiny R Markdown
Levels
R for Beginners R Interview Questions
R r‑statistics.co

tidyr separate_wider_delim() in R: Split Column by Delimiter Into Cols

The separate_wider_delim() function in tidyr 1.3 splits one column into multiple columns based on a delimiter. It is the modern, type-safe replacement for the deprecated separate().

By Selva Prabhakaran  ·  Published May 12, 2026  ·  Last updated May 12, 2026
⚡ Quick Answer
df |> separate_wider_delim(col, delim = "_", names = c("a","b","c"))
df |> separate_wider_delim(col, delim = ",", names_sep = "_")
df |> separate_wider_delim(col, delim = "-", names = c("y","m","d"), too_few = "align_start")
df |> separate_wider_delim(col, delim = "_", names = c("a","b"), too_many = "merge")
df |> tidyr::separate(col, into = c("a","b"), sep = "_") # superseded

Need explanation? Read on for examples and pitfalls.

📊 Is separate_wider_delim() the right tool?
STARTsplit by delimiter into multiple columnsseparate_wider_delim()split by character positionseparate_wider_position()split by regex groupsseparate_wider_regex()split into ROWS not columnsseparate_longer_delim()old code with separateworks but supersededcomplex JSON-like nestingtidyr::unnest_wider()

What separate_wider_delim() does in one sentence

separate_wider_delim(data, cols, delim, names) splits the values of cols by delim and puts the parts into NEW columns named in names. Replaces the older separate() with type-safe column count handling.

Syntax

separate_wider_delim(data, cols, delim, names = NULL, names_sep = NULL, too_few = "error", too_many = "error", cols_remove = TRUE).

Run live
Run live, no install needed. Every R block on this page runs in your browser. Click Run, edit the code, re-run instantly. No setup.
RSplit a date string
library(tidyr) library(dplyr) df <- tibble(date_str = c("2024-01-15","2024-03-20")) df |> separate_wider_delim(date_str, delim = "-", names = c("year","month","day")) #> year month day #> 2024 01 15 #> 2024 03 20

  

  





Tip
separate_wider_delim() ERRORS by default if rows have too few or too many parts after splitting. Pass too_few = "align_start" or too_many = "merge" to control this.



Five common patterns



1. Standard split




  

    
RID into prefix-number

    
df <- tibble(id = c("user_1","user_2","admin_3"))
df |>
  separate_wider_delim(id, delim = "_", names = c("role","num"))

    

      
      
    

    

  

  





2. Handle uneven splits with too_few




  

    
RSome rows have fewer parts

    
df <- tibble(x = c("a-b-c","d-e","f"))
df |>
  separate_wider_delim(x, delim = "-", names = c("p1","p2","p3"), too_few = "align_start")
#> Aligns to the LEFT, fills missing with NA

    

      
      
    

    

  

  





3. Merge extra parts




  

    
RSome rows have too many parts

    
df <- tibble(x = c("a-b","c-d-e"))
df |>
  separate_wider_delim(x, delim = "-", names = c("p1","p2"), too_many = "merge")
#> Merges extra parts into the last column

    

      
      
    

    

  

  





4. Auto-name with names_sep




  

    
RAuto-generated names with prefix

    
df |>
  separate_wider_delim(x, delim = ",", names_sep = "_")
#> Columns x_1, x_2, x_3, ...

    

      
      
    

    

  

  





5. Keep original column




  

    
Rcols_remove = FALSE

    
df |>
  separate_wider_delim(x, delim = "-", names = c("a","b"), cols_remove = FALSE)
#> Both x AND a, b in result

    

      
      
    

    

  

  





Key Insight
separate_wider_delim() is type-safe by default: too_few and too_many are explicit choices. This is the main improvement over separate(), which silently produced inconsistent results.



separate_wider_delim() vs separate() vs str_split







Function
Output
Status






separate_wider_delim()
Multiple columns
Recommended (1.3+)




separate_wider_position()
Multi cols by position
Recommended




separate_wider_regex()
Multi cols by regex
Recommended




tidyr::separate()
Multiple columns
Superseded




stringr::str_split()
List of vectors
Manual workflow







When to use which:



    


  • separate_wider_delim for delimiter-based wider split.
    • 


  • separate (old) only in legacy code.
    • 


  • str_split for vector-level work outside data frames.
    • 





A practical workflow



Use separate_wider_delim for parsing structured strings into columns.




  

    
RInteractive R

    
log_data |>
  separate_wider_delim(
    raw_message,
    delim = " | ",
    names = c("timestamp","level","module","msg"),
    too_many = "merge"
  )

    

      
      
    

    

  

  





Parse log entries into structured columns. too_many = "merge" handles cases where the message itself contains the delimiter.



Common pitfalls



Pitfall 1: too_few = "error" by default. If any row has fewer parts than expected, separate_wider_delim errors. Switch to "align_start" or "align_end" to tolerate.



Pitfall 2: forgetting names. You must provide column names via names (or use names_sep for auto-naming).



Warning
separate_wider_delim() requires tidyr 1.3+ (Jan 2023). Earlier versions only have the superseded separate(). Check version with packageVersion("tidyr").



Try it yourself





Try it: Split a full_name column into first and last. Save to ex_split.



  

    
RYour turn: split full names

    
df <- tibble(full_name = c("Alice Smith","Bob Jones","Carol Lee"))

ex_split <- df |>
  # your code here

ex_split
#> Expected: 3 rows with first and last columns

    

      
      
    

    

  

  





Click to reveal solution



  

    
RSolution

    
ex_split <- df |>
  separate_wider_delim(full_name, delim = " ", names = c("first","last"))

ex_split
#>   first last
#> 1 Alice Smith
#> 2 Bob   Jones
#> 3 Carol Lee

    

      
      
    

    

  

  





Explanation: Split each full_name on " " into first and last columns.










After mastering separate_wider_delim, look at:



    


  • separate_wider_position(): split by character position
    • 


  • separate_wider_regex(): split by regex groups
    • 


  • separate_longer_delim(): split into ROWS instead of columns
    • 


  • unite(): opposite (combine columns into one)
    • 


  • tidyr::separate(): superseded predecessor
    • 





FAQ



What does separate_wider_delim do in tidyr?



It splits one column into multiple new columns based on a delimiter. Replaces the older separate() with type-safe handling of uneven splits.



What is the difference between separate_wider_delim and separate?



separate (old) is superseded. separate_wider_delim has explicit too_few / too_many handling, making errors visible instead of silent.



How do I handle rows with different numbers of parts?



Use too_few = "align_start" or "align_end" for fewer parts; too_many = "merge" to put extras in the last column. Without these, mismatch errors.



Can I keep the original column?



Yes. Pass cols_remove = FALSE to keep the input column alongside the new ones.



What if my delimiter is a regex special character?



Pass it as a literal string. separate_wider_delim treats delim as a literal substring, not regex.



          

        


        
      


      

        

        
© 2016-2026 Selva Prabhakaran.
          This work is licensed under the Creative Commons License.
        

        

          About ·
          RSS