What are people looking for

when using a search engine?

[extra]

I know it's impossible to say what people realy want. But I've seen more and more people are comming back to my site. Last 10 days it's been between 20-30% coming back (15-20% earlier). It's a small niche site with only a few thousand visitors each month, but still the growth has been very stable re visitors coming back. Nearby 90% of all customers are from my own country (in Europe).
I want to make a web site that can keep this high numbers of people coming back and raise it even more if that is possible. I don't do any advertising which means all visitors find me through Goggle, Yahoo, MSN,....(+ customers talking about my site).

Are there any official/unoffical site describing what people really want? How many want entertainment? How many want info/documentation? How many want product info? etc, etc.
Is there a smart way I can combine documentation/info with entertainment?

I feel writing a good web page becomes more and more difficult. I have learned a lot by reading WW, but now I feel I have more to learn than ever. And my priority is to make pages people like to read/return to. This means I need to know more about writing, as well as need to know what people are looking for.

Anyone have suggestions for where to find what I'm looking for?

[jdMorgan]

Look at your server logs and 'stats.' The search words and phrases that are searched-for most often will tell you what your site appears to be providing now (in the eyes of search engines). The search words and phrases that are searched for less often --or that you don't rank so well for-- will tell you what other information you might provide for your visitors.

I have written many pages based on "long-tail" searches -- the ones where people enter five or more words because they can't find what they are looking for using short search phrases. These long-tail-phrase pages often rise to become the most popular visitor-attracting pages on the sites.

Just for example, on a site about widgeting competition, one of our most popular pages is about what widget a beginner should buy in order to do well at widgeting. And the page itself explains that due to the nature of widgeting, it does not matter so much what widget they use or buy, but rather whether they take training classes to learn the most modern widgeting techniques so that they can then practice proper techniques and compete in widgeting competitions more successfully.

I got the idea to write this page based on a small but consistent number of queries for "best widget for widgeting competitions" and similar search phrases. It is now the most-popular page on the site among people who are not members of our local widgeting organization.

So don't just look at your stats for your most popular pages -- Dig deep and find out what else you might provide.

Jim

[extra]

Thanks Jim.

I'm sure you are right.
Main problem is that I'm afraid spread the site to much. And writing realy GOOD info/documentation is difficult as every page needs a lot of work as I'm not a specialist in everything (unfortunately). But this might also be the reason why people are coming back to my site and not going to others. I have put a lot of work in every sentence, and I have used a few weeks to write some of the singel pages.

Looking at the statistics and i.e. see how long time they used on each page could give me some information for what have to be done better.

You have indicated a bit different way to attack this goal than I was thinking of myself. Some people are born writers. Other people got altarnative skills. I'm not the born writer, but maybe I could get people back to my site by using other skills. At least it's worth trying.

[jdMorgan]

Well by all means, take a look at the long-tail. You may find many different search phrases, and some of these may describe things that you can write about. I was certainly surprised at some of them when I investigated this aspect of my own sites.

And consider that it may be advantageous to begin the transition from a tightly-focused niche site to an authority site. You might even want to consider asking your users to contribute material, with the understanding that it will be edited for publication, and that payment --if any-- will be on the terms you specify.

Jim

[extra]

"And consider that it may be advantageous to begin the transition from a tightly-focused niche site to an authority site."

Again I'm very pleased that you help me. I have followed discussion at ww for almost 3 years now. Still I'm not sure about all the buzzwords (maybe not that strange as english is not my mother language).
I.e "authity site". I've seen this been used in many discussions, I have read the glossary at top of this page, I have searched for it on the net but only found discussions or description where this word is a part of the message.

Pls help me understand this description. I know what "authority" is in plain english, but not in this setting.

[unvsoldr]

What he means by "authority site" is that many more people will use it for information.

Niche sites cater to small, specific crowds.
Authority sites interest broad ranges of people.

:)

[Tidal2]

I was going to start this 'On a less commercial note' but this could well apply to both commercial and non commercial web sites. Perhaps this is a merely a slightly different viewpoint of what Jim posted above.

Looking through the tail of your search engine gives an opportunity to enhance your content. Did your page supply what the searcher wanted?

If not, and its relevant for the page - research that search term and improve your page. I will however give an example of a search term I decided not to improve.

The page was about the health benefits of oily fish.
After I added a recommended upper limit from the food standards agency for women and girls then I started to get searches for 'oily girls' and I decided not to attempt to improve that one :)

As a question too, would Google consider changes of this sort (if they noticed) manipulating their search results its the tail after all, but strictly speaking ...?