How to Share Your Data Transformation Steps With a Link

Send a colleague your whole data clean-up as a link. It holds only your steps, so they can reuse it on their own file without receiving your data.

Open the Data Transformation with recipe sharing

Short answer: in Data Utility, build your steps in Data Transformation, click Share recipe link, and send the link. Whoever opens it can load your steps and use them on their own file. The link holds only your steps, never your data.

Why share the steps and not the file

A cleaned file is a result. The steps are the method. If you only send the file, the next export from the source system means starting again, and nobody can see what was done to the data. If you send the steps, a teammate can apply the same clean-up to their own file, check the logic, and reuse it next month.

Sharing the steps is also safer. Your data may contain customer names, emails or financial figures. A recipe link contains none of that.

  • Open Data Transformation and load a file: CSV, Excel, JSON or Parquet (and a second file if you want to join).
  • Add and fill in your steps: trim, filter, join, calculated columns, group and summarize, and more.
  • Click Share recipe link next to the Add a transformation menu. The link is copied to your clipboard.
  • Paste it into an email, a document or a chat message.

What the other person sees

When they open the link, a banner says Someone shared a recipe with you and lists the steps, for example Trim spaces, Filter rows, Join with second file, Calculated column, Group and summarize. They click Use these steps, then load their own file, and the steps are applied to it.

If their file is missing a column that a step needs, a warning names the column and the step, so they can pick another column instead of seeing a confusing error. If the recipe joins two files, they load their own second file.

  • Inside: the steps and their settings. That includes column names and any values you typed into a step, such as a filter value or a replacement text.
  • Not inside: your data, your file names, or anything about who you are.
  • Not stored anywhere: the link is built in your browser and put in the part of the address after the # sign. Browsers do not send that part to a server, so Data Utility never sees it and nothing is saved. Share it only with people who may see those column names.

The steps are written in a compact form and compressed. A typical five-step recipe makes a link of roughly 170 characters including the web address, which is easy to paste into chat. Very long find-and-replace text or long filter values make a link longer, and the app tells you when a link is long.

Share the steps, export the code

A shared recipe also works with code export. After you load the steps, you can export them as DuckDB SQL, Python pandas, PySpark or a dbt model, so a teammate who prefers code gets the same logic in the language they use.

Good ways to use it

  • Give an analyst the exact clean-up for the monthly export instead of a written list of instructions.
  • Hand a data engineer a working recipe to turn into a pipeline step.
  • Keep a standard clean-up for a recurring file in a document or wiki, as a link.
  • Ask for a second pair of eyes: send the link and have a colleague review each step.

Limits to know

  • The other person needs a file whose columns match the names in the steps. If the names differ, they change the column in those steps.
  • A recipe that joins needs a second file from them.
  • There are no accounts, folders or permissions. The link is the recipe, so a link that is lost cannot be recovered. Anyone who has it can open it.
  • It shares transformation steps, not results. Each person runs the steps on their own file.

Open the Data Transformation with recipe sharing

Related tool pages

Also try

How to use the Data Transformation with recipe sharing: step by step

  1. Open Data Transformation and load a file.
  2. Add and fill in your steps.
  3. Click Share recipe link and send the copied link.
  4. The other person opens it, clicks Use these steps, and loads their own file.

When to use the Data Transformation with recipe sharing

  • Give a teammate the exact clean-up for a recurring export
  • Hand a data engineer a working recipe for a pipeline
  • Keep a standard clean-up in a wiki as a link
  • Ask a colleague to review each step

Example

Input: steps: trim, filter, join, calculate, group

Output: datautility.in/data-transformation#r=v2.i442RJJGI6KVYEGj…

Frequently asked questions

How do I share a data transformation with someone?

Build your steps in Data Transformation, click Share recipe link, and send the copied link. The other person opens it, clicks Use these steps, and loads their own file.

Is my data shared when I send the link?

No. The link holds only your steps and their settings. It never contains your data or file names.

Where is the shared recipe stored?

Nowhere. The link is created in your browser and the recipe is inside the link itself, in the part after the # sign, which browsers do not send to any server.

Does the other person need an account?

No. There are no accounts. They only need the link and their own file.

What if their file has different column names?

A warning lists the columns that are missing and the steps that use them. They pick another column in those steps.

Can I share a recipe that joins two files?

Yes. The recipe holds the join step, and the other person loads their own second file.

Why is my share link long?

Long text typed into steps, such as a big find-and-replace list, makes the link longer. A typical five-step recipe makes a link of roughly 170 characters.

Can I also get the code for a shared recipe?

Yes. After loading the steps, use Export recipe as code to get DuckDB SQL, Python pandas, PySpark or a dbt model.

Tip: The link holds your column names and typed values, so share it only with people who may see them.