Simultaneous processing of sample and standard under similar condition gives better analytical precision and accuracy through HPTLC. Additionally, extreme flexibility for selection of stationary phase, mobile phase, developing technique , detection with or without pre or post column derivatization etc are some of the advantages of HPTLC over HPLC , further the choice depends on the outcome expected.
If qualitative analysis, TLC is just fine as one is just looking to see if one component matches the retention time of a standard.
For quantitative analysis, I prefer HPLC since I can make a measurement of the quantity of a compound and also compare the retention of a compound compared to a standard. Note that co-elution on TLC or HPLC doesn't guarantee the compounds are the same.
Sometimes, people just trust HPLC more, sometimes people don't understand that HPTLC can be used for quantitative analysis.
Both techniques are well established and used in qualitative and quantitative studies. My own preference would be HPLC or even better UPLC to get a total picture of the chemical complexity of the sample. Using the higher resolution instrumentation will result in an increase in the analysis cost. Not having to repeat a HPTLC analysis by HPLC or UPLC, I typically go directly to UPLC.
For herbal drugs/ crude plant extracts which have the complex mixture of many chemical constituents of various nature/ polarity, HPTLC being the open system is better than HPLC.
HPTLC is a simple, fast, economic techniques since simultaneous analysis of standards and samples is possible on the same plates. On a 10 ´ 20 cm plate one can spot about 40 samples. It is a powerful technique in fingerprint determination of herbal plants and extensively used in pharmacognosy. Although it has some limitations, number of literatures on determination of herbal effective compounds is increasing todays. Among the main problems that one may encounter in quantitative analysis is instability of the compound in air, problems with volatile compounds, and plate to plate variations in quantitation results. In our research groups we define a correction method to solve variations in quantitative results while standards and samples are not in the same plates. We will submit the paper soon. Of course using internal standard solve the problem.