How do you know if your SEO efforts are really working? Are people finding what they’re looking for on your website? Here is the system I use on my sites to keep track of what people expect to find.
From the outside …
If you don’t have access to server logs, or even if you do, go to http://www.bbclone.de and get bbclone, the php-based web counter on steroids. It provides only the relevant details of your website’s traffic. It will take you a couple of hours to install it on your server, but it is truly a great program, and it’s free.
Here is part of a bbclone detail screen for http://www.MeDiCaLeSe.org. The whole screen is too big to be shown here.
Once you’ve installed bbclone, if you click on any of the underlined numbers in the column just to the left of the operating system on the detail screen, it will give you a report of exactly which pages the visitor saw and how long the visit lasted.
If you hover over any of the underlined referer entries on the right, you will see the search terms that were used to find your website, or the url of the page that links to yours.
… and from the inside
You already have some kind of site search, right? Take the next step and keep a log of what people are looking for. These few lines of code, which are a part of my site search engine, are all it takes to maintain a log file on the server. Permissions on the search log need to be set to 766.
$uri = $_SERVER['REQUEST_URI'];
$ip = $_SERVER['REMOTE_ADDR'];
$dtime = date('r');
$entryline = "$dtime - IP: $ip | $uri ";
$fp = fopen("searched.txt", "a");
fputs($fp, $entryline);
fputs($fp, chr(13));
fclose($fp);
?>MeDiCaLeSe is a reference site for medical transcriptionists and other healthcare workers who need to be able to find medical terms quickly. Everyone is welcome to search the database, but it isn’t necessarily useful to people who are not medical professionals.
Every now and then I read through the search logs to find things that should be in the database, but are not. Even searches for terms that don’t exist can still yield information. Here’s an example.
Wed, 31 Dec 2008 14:47:47 -0700 - IP: 000.000.000.000 | /sitesearch.php?sw=bacterium&w2=prorpyana&xw=Bacterium prorpyana? I didn’t think I’d ever heard of it. First I combed through my database, and I saw nothing remotely like that already listed. And I couldn’t find Bacterium prorpyana on the internet, but I did find an interesting article about the newly discovered pathogen Mycobacterium lepromatosis, which causes DLL, or diffuse lepromatous leprosy. Yes, it’s just as horrendous as it sounds … and these two terms should absolutely be included in a database for medical transcriptionists, who seem to get most of their information from the internet these days.
Recent Comments