Indicators which are defined with both numerators and denominators are typically more useful for analysis. Because they are proportions, they are comparable across time and space, which is very important since units of analysis and comparison, such as districts, vary in size and change over time. A district with population of 1000 people may have fewer cases of a given disease than a district with a population of 10,000. However, the incidence values of a given disease will be comparable between the two districts because of the use of the respective populations for each district.
Indicators thus allow comparison, and are the prime tool for data analysis. DHIS2 should provide relevant indicators for analysis for all health programs, not just the raw data. Most report modules in DHIS2 support both data elements and indicators and you can also combine these in custom reports.