How to apply the address data improvement
The address data improvement workflow allows you to enrich/improve address data in a table. It uses an agentic approach to automatically improve data on countries, cities, streets, etc.
What does it solve?
For any table with address data, i.e., country data, city data, street data etc, respective columns will be validated and enriched or improved where necessary and possible.
This is what it looks like:

Workflow setup
There is a clear workflow that can be followed. Consider the following example:
1. Select file on DQC Platform (from a connector or static file)
2. Set up rules to identify incorrect address data (e.g., missing entries to enrich address data)
Set up a 'not empty' rule for the columns including address data
3. Go to “Issues" and click on “Improve data"
4. Set up workflow
Here, you can simply set up a workflow consisting of "Data input", "Address" and "Preview". A new workflow already contains the "Data input" node. Add "Address" and "Preview" from the Node library by double click or drag-and-drop.
In the next step, the nodes can be specified.
5. Specify nodes
Data input node: By clicking on the pen icon, you can specify the node. Choose the Source first, for example "Issues from the last check". Limit the columns and rows or set a filter if needed.
Address node: Select the columns for which addresses should be enriched (based on the data quality rules defined):
Select all relevant address columns as Lookup columns (Ask AI can suggest them)
Map the columns to the GeoCoder fields
Optionally, add Additional context, for example a company or building name: it enables the POI (Point of Interest) search for fields the geocoder cannot fill

6. Prepare results
Two options to see results: Preview node or CSV download.
For the Preview node: open the node and click "Calculate input sample". It runs the nodes above with sample rows and shows the improved address data including a confidence score.
To get a CSV download with results, click on Run and then on the download button
Notes
Ideal for data cleanup and filling of missing entries
Result table can be downloaded
Running a workflow needs a token budget for your tenant. Without a budget, or when it is used up, the run does not start; the Preview node still works for testing with the first rows
Learn more: Connectors, Duplicate survivor, Rule detail view