Microbial community analysis at the DNA or RNA level ideally requires the ability to measure the quantitative abundance of hundreds to thousands of gene sequences with the highest possible sensitivity and specificity. Microarrays, when fully developed, are expected to fulfill this need. Current microarray studies, however, focusing on either phylogenetic or functional gene targets or both have not demonstrated the full potential of microarrays for microbial community analysis. This is mainly due to the lack of high throughput validation strategies for complex microbial community samples. Theoretical probe design and microarray fabrication targeting thousands of gene or microbial targets is found to be the easier step. Validation of the array to confirm the identity and abundance of specific target genes or microbial populations has proven to be the more difficult step, especially for complex microbial communities. We have designed and tested several microarrays focusing on both phylogenetic and functional gene targets using an in situ synthesis platform. Results to date indicate that platform-dependent high throughput experimental screening is necessary to select good probes. Although shorter probes, e.g., up to 30-mer may provide better specificity, it is at the cost of sensitivity. While 16S rDNA is critical to provide the phylogenetic anchor, genes related to specific functions provide better strain level resolution. Multiplex amplification of genes related to specific functions is only feasible in small batches due to lack of universal primers. Alternative amplification strategies for gene sequences and/or signals may be a better strategy for targets present at less than 1%. We also present new data analysis approaches for enhanced reliability and specificity of detected signals emanating from low abundance targets. On-chip target amplification and signal amplification strategies are being adopted to further improve the specificity and sensitivity to measure low abundance (0.0001 to 1%) targets.