Adwords is all about keywords. You need to work on growing a list of keywords/phrases that visitors you want would be typing in. Bid on those. Watch the logs after you start getting traffic, you'll see what people are typing in to find your site. (Lather, rinse, repeat)
I have my Adwords account set at $1 or $2 per day, that's OK.
New concept: CPC = cost per click. Make sure you're happy with what you're paying for hits. You don't have to bid on things that cost too much, if you can find 20 other keywords that you can pay less for and generate a similar amount of traffic.
[edited by: Slade at 3:42 am (utc) on May 11, 2007]
You can use the keyword tool in your adWords account for suggestions, but of course that's no guarantee those words will work for you. You have to try them out, see what works and what doesn't. A good sense of your customers, how they think, what they are looking for, will give you a good place to start.
$1/day = $30/month, actually that's where I started. You won't get many clicks, your budget will get eaten up quickly, but it's a good way to learn to be frugal from the start and begin testing the waters.
Something that took me a while to understand that applies to your keywords question, think TARGETING. "Movies" is an extremely broad keyword and can apply to various topics - movie reviews, buying movies, movie plot, movie prop, etc. Each of the preceding are slightly better than just "movie." You will find the better targeted your search terms are, the more clicks you will receive and the less those clicks will cost you (most times.)
Apply this to your ads too. You will get a better sense of what "works" by creating LOTS of ads with very few keywords in them. Lots of keywords in one ad to try and catch everything actually dilutes your results, and confuses things. Very few of my ads exceed 3-5 keywords, and most of those are variations - example, scifi movie, scifi movies, sci-fi movie, sci-fi movies, science fiction movie, science fiction movie (those are still very broad, just examples.)
Of course as you dig around you'll find it's a lot more complex than that and I'm by no means any expert, but after a few months the light bulbs start coming on. :-)
Here's a hypothetical situation. Through traffic analysis you determine that you earn $1.00 for each visitor that comes to the site. You also know that the COGS is 40%, your overhead is around 25%, and you'd like to see a profit of 15% as well. That leaves 20% left over for marketing. In other words, any time you can buy an on-topic click for $0.20 or less, you should be meeting your financial goals. As far as cash flow allows, buy as many clicks at this price as you can and carefully measure the conversion OF EACH KEYWORD to ensure they are meeting your projections.
