Today we're going to walk you through everything you need to know about how Kapiche cleans any data that is mapped to the NUMERICAL and NPS field types during the Schema Mapping stages.
In this guide:
Why Kapiche needs to clean data
We've noticed a number of customers have been experiencing some issues with the way their survey providers export certain types of numerical responses like NPS & CSAT where records would contain both Numerical and Text. For example:
Disagree0
10Agree
0 - Extremely Disagree
10 Excellent
Completely Satisfied10
When this happens, Kapiche doesn't know whether to process these records as Numerical Data or Text data which results in a processing error - ultimately causing Kapiche to ignore these records during the upload stage.
To help solve this problem we have a feature that allows Kapiche to automatically detect and clean any data that is set to the NUMERICAL and NPS field types during the Schema Mapping stages.
This means that if Kapiche detects that a record mapped to the NUMERICAL or NPS field type contains text (like one of the examples above/below) we will automatically remove the text and process the data as intended/expected!
How does the cleaning of the data work?
We clean a numerical field by removing all non-digits from the beginning and from the end of the record/data which includes all punctuation and symbols (eg. $ % . - etc) as well as spaces.
To accommodate negative numbers, a hyphen (AKA minus sign) will be included if:
A hyphen was immediately before the first digit found
and it was either the very first character in the pre-cleaned data or it had a space in front of it
Types of data Kapiche cleans
Here are some more specific examples of the types of data Kapiche will automatically clean for you (providing they're set to the NPS or NUMERICAL field type during the Schema Mapping Stage):
Data Example | Output | Cleanable? | Reason |
|
|  | Removed the text before the 0 |
|
|  | Removed the text after the 0 |
|
|  | No cleaning needed |
|
|  | No cleaning needed (Kapiche handles decimals) |
|
|  | Not cleaned - no digits detected |
|
|  | Not cleaned - multiple digits detected |
|
|  | Removed the text before the 4 and after the 4 |
|
|  | Minus number detected as the |
|
|  | The hyphen was detected as part of the text 'Age' because it was attached to the text without a space before or after it. |
|
|  | The hyphen was detected as part of the text 'years-old' which was then removed because the text appeared at the end of the data and was separated from the digit with a space. |
|
|  | This data would be ignored / failed as it multiple numbers were found. |
|
|  | This data would be ignored / failed as it multiple numbers were found. |
|
|  | Removed the text and symbols before and after the numeric data. |
|
|  | Minus number detected as the |
|
|  | The hyphen was detected as part of the text 'Unlikely' as there wasn't a space after the text and before the 1. |
Questions? 🤔
If you have any questions about Numerical & NPS Field Cleaning (or need some help!) you can get in touch with us any time by hitting the blue chat button to your right 👉