First of all most of the time I know how I am going to process the photo (b&w or color) right before hitting the shutter button. For me this is an easy decision, b&w is the first option for shots where the light is not that great (mid day or any time of the day when the sun is harsh, with heavy contrast and ugly shadows all over the place) or when I am shooting some street shots of people with a bit of a social case touch.
I rarely use the in camera monochrome capabilities, most of the time I rely on my Photoshop skills to achieve the results I want. I only shoot 14bit NEF files, download the files to my PC with a card reader and view the files with ViewNX. My only post processing tool is Photoshop CS4, maybe Lightroom will work in some way too (and even faster) but this is how I started doing things way back and I feel that I can't stop.
Some post processing ideas that might help:
*I start from a properly exposed image, flat looking. Flat means that I underexpose a bit the image when I import it to photoshop and I recover the highlights a bit in the needed areas by using an ellipse selection tool, feathered to 250 pixels and with various brightness settings, as needed. The same goes for the darker areas of the image, feathered selections and an increase in brightness
*after the flat looking image step I make two or more new adjustment layers with the following settings: one with blending mode set to color where I set the saturation -100 (this makes the photo black & white) by double clicking the layer thumbnail from the layers window; the second layer I create has blending mode set to normal and altering the hue settings affects the overall look of the photo. For this to work the layer with -100 saturation has to be on top of all the adjustment layers.
*I apply some further tweaking to the background layer as well, brightness/contrast, channel mixing, curves, it all depends on how the original photo looks like.
*Sometimes I also work with the duotone feature, image / mode/ duotone and combine some several colors for an easy conversion. After I am done with the colors I convert back to RGB from the duotone mode
*I always end my processing with some kind of sharpening filter to obtain that punchy, crisp look. I use either the smart sharpen filter or the unsharp mask, small amount %, hefty selection of pixel radius 25-30px and threshold around 5 levels.
I hope this might help a bit. When I started to learn about b&w conversion I read 3-4 tutorials, experimented a lot and with time I got better and better, using my very own techniques. What's really awesome about postprocessing is that totally different techniques can lead to the same result, finding the best technique to call it your own I think is the hardest part.