After the explicit and specific indicator is chosen, it must be populated with good and reliable data.

There is a trade-off between choosing a good indicator and accompanying it with valid data. Sometimes it is not possible to collect data for an indicator you want to measure, perhaps because the data cannot be extracted from an existing source such as an electronic medical record. Or, data that is available for you to collect may be irrelevant to what you are trying to measure.  

The balance between choosing a well-designed indicator and supporting it with valid data is a challenge. Satisfying both components as best as possible will provide the most value to your system.

Without trustworthy data to feed into it, an indicator is worthless.