Divide to Conquer
What do you love vs. Do you love money? - There are two ways to drive to you to a topic your site will cover: is it of interest or is it popular? The best topic is one that's popular, then you have the widest audience. But you also will have the most competition. If there is a topic that you have a passion for, you may be able to speak authoritatively. You may know some minutiae that is totally unknown by the larger population. For example, "R2-D2" the name for the Star Wars robot came from the production house where Star Wars was being worked up-- R2-D2 stood for Reel 2, Disk 2. Can you spin that kind of info into something interesting to people who have a more surface interest in the same topic? If you can, then you want a topic related to something of interest to you.
Entertainment vs. Education - Do you want your topic to come from one of two houses: one that educates or one that entertains? For example: a site about smart investing educates. A site about Bernie Maddoff's excesses entertains. Education is good but the information needs to be sound. If it's analytical data (like stock market numbers), people may be able to find the same data on a more popular site. If you make a great inference out of the data, then your site is worth the visit. Instructables.com does a great job of showing people how to do things. Entertainment is both an easier act to pull off and more fleeting. That said, Youtube.com got to be popular not because of its educational content, but because of all those great videos of cats that can talk.
Where To Dig
FUD - I like to think that people seek what they enjoy, but they are also attracted to what concerns them. The Internet is a private confessor. You can search for a very controversial subject and do it with comparative anonymity. Do you have a bladder issue? Scared you're gay? Looking up divorce lawyers? The Internet is a friend who won't blab about your interests (as long as you clear your browser history ;)). A place to mine for topics is inside of what was labeled "FUD": Fear Uncertainty Doubt. People look for information on what they are afraid of, uncertain of, or what they have doubts about. If you want to dig into the direction of an education or informative site, FUD is fertile ground. Cancer is one of the biggest killers and most people will come in contact with it-- they'll get it or their lives will be changed by it. If you have an interest in writing about cancer research, survival rates or tales of triumph, that's right in the middle of FUD territory and your site could find a lot of interest. On a less morose note, a lot of people worry about finances. A site like FrugalVictoria.com shows people how to save money. It's a terrific topic with one gotcha: if you're poor, you may not shop the bookstore or lay out for the eBook. Knowing your topics is a little bit about knowing their earning potential.
Sunny Unicorns - FUD is about the dark drivers of interest. There are a lot of bright topics. While a coupon clipping site may cater to people without cash, a site about the best golf courses would do just the opposite: it would cater to people with money and it would carry along a lot of affiliate program potential (stand by for upcoming piece on affiliates). People will enjoy a positive site, but such sites fall into more of a constructive or recreational realm. People don't need to go there, so you have to make the content sing: timely, clear and fun.
Be Happy - The world likes positivity, despite everyone's fascination with horror, crime and corruption. If you made a blog about how your life sucks, people may come for a dose schadenfreude, but you will be hard pressed to build a lasting audience. What if you did a blog about how people fail? Fail Blog, Cakewrecks and a lot of others do make a market out of failure because the enjoyable bad examples are all guest stars on the site. Even though it's looking at bad stuff is does it in a way that's fun. This is why ThereIFixedIt.com is helping to power the life support Jay Leno's career by showing up on the Tonight Show; but Stileproject.com and Ogrish.com are in the dark fringes.
Living Topics - Visits are traffic. You can attract one person to do a drive-by; or you can get one person hooked and convinced to come back every hour for years. Your site should be about a topic that evolves and grows over time: a living topic. Breaking news is good, but CNN dabbles in news. You should aim for a specialty. More in-depth looks at a subject are great. Likewise, something that aggregates new information as it becomes available is awesome. I am working on an all-mobile site, www.VicBC.mobi. It has a large contingent of crowd sourced information. Google will frown on regurgitations of old data that is can find elsewhere (more on that topic in an upcoming piece), but it will reward smartly cooked data. If something is new on your site each time a visitor hits it, visitors love it and Google does too.
Where do you start?
We have a podcast about ideas: Ideas where do they come from?. Maybe that can get your gears in motion.
Look at what interests you. Look at what interests your friends and co-workers. At the beginning of this exercise, you'll have to find some sounding boards. If your and your friends are into basketball, getting their take on a new particle physics site may not yield a lot of good feedback. Make a list of topics that you know a little bit about.
Take out the dud topics. If a topic is too obscure, you can write about it, but it may be hard to turn it into your First $10,000. Ask yourself: could 100,000 people worldwide be interested in the topic?
You need to turn out a decent volume of content. The best approach would be to have 1 piece a day popping onto your site and out to your followers. A fallback is one a week and some sites where they have LOTS of new content get a new piece every hour. I am not a cartoonist because I can't come up with hundreds of comic strip ideas, I can come up with 10. That wouldn't be enough to give TheOatmeal.com a run for its money, so I'm not doing a cartoon-a-day site. Maybe you could. Could there be 50 articles on the topic?
Still got nothing? Look at Google Trends for ideas