Forum Moderators: DixonJones
What is commonly used to term "hits"?
I know a little bit about logging, and understand the issues and problems surrounding cache, how to identify unique users, etc.
For example, in one sample week my site has
400 000 successful requests
62 000 pages served (including static htm and also php from my forums)
6 000 distinct hosts served.
2.5GB of data served
I regularly see claims of x number of hits from other websites, and I rather suspect that they are counting sucessful requests. I've always felt that I should quote pages served as a more realistic figure. I'm now wondering whether I'm going my site a dis-service by claiming pages served instead of requests.
Sure, it's only an ego thing, but I want to make it clear that I'm the heavyweight in the (admittedly small) niche area I'm in. :)
Do you think these stats are worth persuing for some kind of advertising? Are they respectably large? My site is a hobby site and (currently) has no commercial affiliations.
[edited by: KakenBetaal at 11:15 am (utc) on Oct. 4, 2002]
As you know a hit is every request to the server, so if a page has 50 images then 51 hits are made despite the fact that only one actual page visit has taken place.
If you think in terms of page views and repeat visitors that is better than standard hits. I understand your difficulty in saying page views when other people are saying hits.
I suppose you could just claim hits and be ready to explain yourself when it comes up. I would suggest that a potential advertiser would be on the ball and this sort of "scam" would not work in most cases. Might catch a few lesser informed individuals though.
Difficult situation.
A client recently contacted me regarding a paid ad on a site that was claiming a good number of hits per month. At the bottom of every page this guy had 300 invisible 1x1 graphics. He was indeed getting a lot of "hits" but actual visitors were a bit more scarce. Needless to say he used a stat program that didn't provide uniques or page views.
[edited by: digitalghost at 11:15 am (utc) on Oct. 4, 2002]
What is commonly used to term "hits"?
Well, there's a lot of confusion related to the term. Technically hits ARE requests on the server - both pages, images etc.
I wouldn't recommend that you use hits/request at all (except when monitoring server load) - the numbers are useless. Eg, if the frontpage on your site has 1 image a pageview generates 2 hits. If your competitor's site has 25 images on the frontpage a pageview generates 26 requests. In both cases it's only 1 user viewing 1 page.
Regarding your numbers: If you're happy with them... brag! ;)
1. Are counting all downloads of images, scripts, pages, css, js, sometimes even includes, and much paraphanalia
2. Are amateurs and dont know what they are talking about
3. Are overestimating!
Best way is to declare Page views and unique servers. Both are imperfect but many times better than "hits".
Even break it down by your top pages so people know whether this is over 100 pages or 10,000! eg Home Page: 17,000 page views.. News page: 28,000 page views etc...
If i was trying to determine the popularity of a site i would also want to know how many page views were by search engine or any other sort of non-human robot. I know for a fact that 10-15% of our page views to our established sites are robots, and around 30% or over for newer sites.
"Hits" is a term scammers use to boost their stats. People declare they get "5,000 hits a day", and you check the site, it isn't in Google, isn't in Yahoo!, isn't in any search or directory. A lot of scammers.
Here's a question I really hope somebody could answer for me.
What are the top types of "Direct Request" visits?
Bookmarks is one, and sometimes clicking on a link in a email (OE) counts as a direct request, what else is there?
I get more and more direct requests every month and I'm trying to figure it out. From the sheer number I'm discounting Bookmarks.
Thanks for your help in advance.
I get more and more direct requests every month and I'm trying to figure it out.
As far as I can see, having discussed it with many people, there are a lot of pet theories, but no real answer. For example
As far as I can see none of the explanations I have been offered actually account for the volume of visits from such direct sources.
Boasts about traffic have their place, but I think boasts work best when you quote the "unique visitors" stats, which seems most credible and understated.
But I too have a question. If a site has 1000 unique visitors per day, how many are real people visiting, and how many are not (spiders, nimda probes, etc).
Surfers who arrived at a page other than the front page, then moved to the front page (never really bought that one)
The only way that would make sense is if they arrived on the sub-page, and instead of looking for a home link, edited the text in the address bar and hit Go/Ok/Enter.
But then, you would still have a hit for the sub-page, and where do you account for that at? If it's got no referrer then what? You got a page hit that didn't show up in your logs? Come on... (i know not impossible, but improbable)
not impossible, but improbable
I agree entirely, which is why I put
never really bought that one
The only way that would make sense is if they arrived on the sub-page, and instead of looking for a home link, edited the text in the address bar and hit Go/Ok/Enter.
I do this myself sometimes because it's faster. It also happens quite a lot at my own site since it's based entirely in a sub-directory. There's no end to the number of people who manually decide to take a look at the root directory just out of curiosity. Unfortunately there's nothing in there to see...
Yup- thats the distinct impression anyone who knows their stuff will get if anyone talks about hits. Stick to Page Views and Uniques. If you are trying to attract advertisers then you have to stick to what is relevant to them.
