Forum Moderators: buckworks
I can see from my logs that people enter the site and then don't go anywhere, or only go to one page, then leave.
My questions are these -
Is this normal, do I need a greater ratio of browsers to purchases to effect a sale?
Is my site too slow (I think it might be, but that's just me)?
I'm not sure if I can post the url here, so I'll refrain from doing it unless advised.
Thanks.
You are going to get some people that leave quickly simply because you aren't what they are looking for. That could be a result of mistargeting or under-targeting your adwords.
Your pages may be slow. No matter what, if you think you can make your pages faster, you should do that. I read somewhere (though I can't remember where. Might have been Brett) that you lose (I think) 1% for every second over 9 seconds.
Having someone on the site for a long time might not be a good thing either. That can somtimes indicate that the visitor can't find what they want (like checkout).
Your best bet would be to figure your average time. No matter who you are, bad site or good site, you will get some people who leave quickly and some that surf around for a long time. If your average is really short, you have problems. If your average is really long, you have problems.
For instance you sell "new blue widgets" so you are bidding on blue widgets in adwords, however people search for "used blue widgets" click on your advert, and then realizing you don't sell used merchandise, click back and try another site.
I don't worry about those visitors who found me through the search engines -- it didn't cost me anything for them to come visit -- but I do keep a closer eye on how many conversions I'm getting from my various PPC campaigns.
About a month ago, I started using HitsLink, which is a hosted statistics and analysis service, and I've been thoroughly impressed with it. Previously, I could see how many clicks I was getting from the various PPC campaigns, but could never tie them to specific orders, *especially* if the customer bookmarked my site and came back later.
With HitsLink, every order is attributed to its original source and I can see how much money I'm making for each PPC campaign, which will help me focus my efforts -- I can either modify or remove the ones that aren't converting and spend more on the ones that are working. The nice thing is, if a visitor comes back a day or a week or a month later to place an order, it's still attributed to the original source. Apparently they drop some cookies on the visitor's browser that lets them track return visits, etc. - I've heard a few people suggest that they're a "spyware" company but I find their tactics to be completely harmless.
We then want at least 25% to be on the site for the average time it takes for a conversion (The conversion rate isn't 25%, just looking at average visitor times). Overall, setting a goal of 5-10 minute average visitor times I belive is reasonable for PPC campaigns.
There are two places to start to cut down on the very short visits.
The first is the keywords & ad itself. Targeting the keywords to specific offerings, and writing ads that describe your site will only entice people who are looking for something you can offer.
The second is landing pages. Don't send everyone to your homepage. Send them to the most appropriate page for that exact keyword.
It sounds like you might want to read through some posts in the AdWords, PPC, and Overture forums (AdWords is the most active) about targeting.
Do you AdWords ads link to the related section of the site, or just to the homepage?
Like others have said, be careful not to make your keywords too broad as this will make your PPC advertising costly.
For instance, you may be selling an widgets of various colors. If someone enters the phrase yellow widget and your come up in the listings they don't want go to the home page or a page about blue widgets. They asked for yellow and that's what they want.
It may, also, be the case that your page doesn't promote your yellow widget as well as it might. Many top PPC subscribers test landing pages to determine which elements they need to tweak to get the best results. This may be an overwhelming task for a small business, but you can do it just the same.
But the point is if people ask for yellow widgets they want to see yellow widgets.
as I understand it, this means one IP address is used to download the html content, another for this image, another for that image, another for the css file, etc, etc.
some (most?) log trackers will see this person as entering and leaving the site straight away, as his IP changes so quickly. makes a bit of a joke of stats analysis
i don't understand the actual technicalities behind it, just enough to grasp the impact of it :)
(the tracking and logging forum which has some excellent posts on this)
