Most bloggers and webmasters who do keyword research aim to find opportunities where the search term is fairly popular, but not fiercely contested. The idea is that it's no use trying to rank well for 'san diego hotels' because some very big players target that keyword and it's going to be very hard to rank well for it, and it's no use putting all your time and effort into ranking well for 'cheap dog friendly san diego hotels' because capturing the top spot won't do you much good. Very few people will search for that phrase.
But what if you don't focus all your efforts on that very specific keyword? What if you do the minimum amount of work required to rank well for that phrase (if it's not very competitive it won't take much work), then do the same for 30 more like it? Each one may not bring in much traffic alone, but between them they could supply a healthy stream of visitors. The total amount of work required to rank well for 30 easy keywords could still be a lot less than what you'd have to put in to rank well for one very strong keyword.
That's the principle behind 'long tail' keywords. They aren't popular but they are easy to rank for, and if you get enough of them, they can bring in plenty of visitors.
It's not just a clever way of boosting the number of arrivals either. It's very clear what someone searching for 'cheap dog friendly san diego hotels' wants. If you can provide a page that contains that specific information, you'll have a good chance of making that visitor very happy. They'll be far more likely to click on an affiliate link, sign up to your newsletter, or even just spend a lot of time on your site than someone who arrived through a more general keyword and didn't find exactly what they were after.
Blogs are ideal for targeting long tail keywords. You have the freedom to write each post on a very specific topic and can add as many new, diverse pages as you like- something that can make sites with a more static architecture look cluttered and untidy.
You don't need to focus your whole strategy around long tail keywords, but just choosing half a dozen to experiment with can be a worthwhile activity. Pick a few relevant, low competition and relatively low traffic phrases and see what it takes to get your blog ranking well for them. Compare the amount of work required with the traffic you get back (remembering that long tail traffic can be very high quality) and you might find that these keywords deliver a pretty good return on a very small investment.