Research Before Starting New Sites

[cameron]

I was just wondering what everyone else does to research before starting a new site. For the past couple of days I've been trying to find a topic.

I've been using the wordtracker free trial, searching in google and checking the pr and quality of the top listed sites, and looking through dmoz for suitable places to get ranked.

[thebillionaire]

try tribal fusion, click on the category you may want to start it will tell you the gender that views that category the amount of money user viewing that category have, and there age. In some they tell you how many % make 100k in a year so the higher the percent the more popular the category is.

[Chris]

I usually come up with an idea and then I see if it is viable.

I check Google and look at current top ranked sites, can I compete with them?

I check wordtracker and see what kind of traffic potential there is.

I consider sources of content and if it'll be worth it.

[incka]

I prefer to use the www.digitalpoint.com/tools/suggestion to find key phrases.

Then I look at google competiton.

Then I look at affiliate programs and tribal fusion marketplace, existing sites revenues, etc.

[Mike]

Yep, I just suddenly think of an idea, check google, then check digitalpoint.

[shim]

I come up with some ideas.

Narrow it down to one idea.

Try to find an 'angle' for that idea.

Check Google for competition.

Find content sources.

[cameron]

Chris do you have a wordtracker membership, or do you just use the trial? I had subscribed for a week in January and I liked it a lot. I'm very low on cash right now though so I'm not sure if I want to go for it this time around.

[intelliot]

i usually come up with ideas myself. if i think it is a good idea, that's more than what any tools can tell me :) almost any idea can be successful if it is something you're interested in or passionate about.

[Mike]

I had subscribed for a week in January and I liked it a lot. I'm very low on cash right now though so I'm not sure if I want to go for it this time around.

Are there many more features in the paid version?

[cameron]

The trail just gives results from MSN. With the paid version you can get results from Google, Yahoo, a few other search engines, DMOZ, Yahoo Directory, and some pay per click search engines. You can also store your results, have more than 30 words in your basket, and it brings up more related terms.

I decided to just get it for a day again. Used it for about 5 hours last night. I haven't found anything great, but it helped me decided against doing some ideas I had.

[Anat]

I also bought the one day subscription a couple of months ago. I ran a lot of ideas with it and now have them all stored in a little file for future use. Of course, as time goes on and the market changes, those results will become less relevant.

IMO the wordtracker tool is very helpful when it comes to focusing on particular keywords or phrases. When you look for the field, it's still a lot to do with intuition and what's interesting for you, as well finding the right affiliations, ad networks and sources of content.

[Mike]

So the results the free trial displays, are supposedly less than you could get altogether?

[incka]

It can't be from MSN. I was/am top of msn for msn emotions and I'm not getting even over 100 clicks a day.

[Mike]

Yea, when msn showed looksmart results i was second, for "msn emotions". I was only getting 250 uniques a day, so it can't be from them. Someone else told me it was altavista?

[incka]

Perhaps it is proportional. Say MSN gets 1% of searches then times its results by 100 for the demo.

[ozgression]

Someone else told me it was altavista?

Yes, I believe it is AltaVista.
________
iolite portable vaporizer (http://iolitevaporizer.net/)

[cameron]

I have lost faith in Wordtracker. I came up with "embarcadero old skateboarding pictures" as a keyphrase to target. They estimated 200 searches or more per day. I added the phrase to my site, got ranked first, and I have yet to receive a single visitor from it. This was using the paid version a few days ago. Their whole system is based strongly off the estimated number of searches per day, and if they can be off by more than 200 then there isn't much value to the tool.

[intelliot]

I think it is virtually impossible, technicially, to determine the actual number of searches for any keyphrase. Whatever algo they use to come up with their numbers can't be quite accurate -- unless they owned Google, Yahoo and MSN.