hi Community
I have a created a workflow that analyses a large transaction data set and categorises and appends the data with additional field descriptions based on a sequential set of rules. Approximately 90% of the transactions can be categorised based on these rules, with the remaining 10% requiring some manual decision making outside of Alteryx. The source transaction data will change and grow over time such that there would be additional manual analysis required throughout the year.
At the end of the workflow, the data that requires manual analysis is split based on a filter function. I then take an extract and add manual columns to the file, but would like there to be a feedback loop that enables me to then add the data that has been manually analysed (10%), back to the automatically categorised data set (the 90%). As new data comes through, i may need to repeat this process but do not want to be reviewing existing data that has been already analysed, again.
I was thinking of a workflow with a join function where the right join becomes a file that is analysed, and then fed back as an input to that join function,.. but this doesn't work well.
I would be grateful for some advice on other alternatives.
thank you
Hi @gavin_chin ,
what exactly does not work well? Is the output not as required? As I understand your question I would handle it the same way.
You could use the Find&Research tool and then filter the ones which do not have a match. This tool may fail with a lot of data coming through the Replace side. it gives usually a warning.
Buzz