Tropical Storm Hanna has mostly passed through the area, but it's still pretty gusty outside, so power failures may still take the site down.
One of my other servers will need a manual kick, so supporting the site
via credit card is currently not possible.
This article was originally written when Google was a "research" search engine! Still relevant though!.
Are all Clickthroughs created equal? Some thoughts for those
interested in advertising on the 'Net
This article will be of most interest to people running commercial sites, but people running
personal homepages may find it amusing. As with all my opinion pages, your comments are appreciated.
It seems that most webmasters are obsessed about three things; first, their
rank in the search engines, the number of people who visit their site,
and, well, the size of certain unmentionable parts of their... well, you get the idea.
To give you a taste of how cool the Secret Net Tools are (you have to
contribute to get access to all of them), I've enabled your access to several of
the simpler ones. Enjoy!
Believe it or not, 11.13% of the people who use SelfPromotion.com support the site!
I will argue, however, that the raw number of people visiting your site is almost always the wrong
way to judge how well you are doing. In the rest of this article, I'll take the perspective of a
webmaster whose site sells things because, well, that's what I am. For me, the important thing is not
how many people visit SelfPromotion.com, but how many I can suck... I mean, persuade into contributing
some money.
Now, your "metric" for success may be different from mine, but in all cases, it won't be the number
of clickthroughs you get, but rather, the number of clickthroughs that end up with the end-result you want.
It is easy to get clickthroughs. You can buy them by the megabyte by purchasing banner advertising,
or by spending a lot of time or money to get a high search-engine ranking, and so on. What you
have to determine is, "how much is each clickthrough costing me?", and "am I getting my money's worth?"
Let's say you buy advertising on a major site for $1.50 per clickthrough.
If 10% of the clickthroughs result in a new customer/consumer/user for your site, then $1.50 cost-per-click
becomes $15 per new customer. If only 1% do, then it's $150 per
customer. $150 per sale may be wonderful if you are selling luxury cars, and a total disaster if you're selling bubblegum (unless you sell it by the truckload).
Bottom line, if spending $40 on promotion generates $60 in profit from the clickthroughs, great. If it
generates $20, you're in deep doo-doo. You've got to have some way of determining if a particular investment in
promotion, be it banner ads, search engine ranking, etc., is paying off for you. Fortunately, in many cases, you
can get a pretty good measurement of these things.
If there's one thing I've learned, it is that while all men are created equal,
all clickthroughs definitely are not. In rough
decreasing order of quality:
Online Newsletters and Mailing Lists
Links on other sites with topics related to your site
Categorized Indexes (Yahoo, NewHoo, etc)
Links on other sites (general)
Search Engines
Banner Ads
Free For All Pages
So, weirdly enough, I would much rather get 50 clickthroughs from
people who found me on Yahoo than 10000 Free-For-All clickthroughs. The former are looking to promote their
website; the latter are bored out of their minds so much they have nothing to do but cruise FFA sites.
This was one of the interesting lessons I learned from running the FFA submitter Secret Net
Tool. It got me hundreds of clickthroughs and very little income. If I had used a commercial service
that does FFA submissions, I would have wasted my money.
On to some practical suggestions for optimizing your return on investment from online marketing.
First, clearly the best investment around is getting yourself properly listed with all the indexes. But
since you're already here, you've probably got that well in hand.
If you are going to target search engines with banner campaigns and ranking campaigns campaigns, it makes sense
to do some research to make sure you are targeting the right things. At the time this article was originally written
(ie: back in the dark ages!),
a good place to cheaply test things out was
Goto.com (which then became
Overture, and finally Yahoo Search Marketing -- that shows you how long ago this page was originally written!).
Some people object to the idea of selling the top positions in search returns, but I find them a
wonderful resource. Rather than mess around trying to capture lots keywords on all the search engines, sites like
Yahoo Search Marketing and Google AdWords
lets you quickly test which keywords work, and how well they work.
For example, by doing a few simple tests I quickly found out that by far
the most important keyword for me was "add url", for example. I was surprised. I thought it would be something
like "register site" or "promote website". Nope. 80% of the clickthroughs came from "Add URL". What I also
found out, equally important, was whether or not it was worth the expenditure of time and effort to try and
capture a high position for that and other keywords.
For example, if keyword A gets 2 times as many clickthroughs as keyword B, but B has a clickthrough-to-sale
conversion rate 3 times
better than A, B is the keyword you want to target; those clickthroughs are more valuable to you. And in my
case, while "Add URL" generated more clickthroughs, the conversion rate on more targeted keywords like
"Register Website" was higher. So when I buy pay-per-click ads, I dont't target "Add URL";
I get more for my money (but less hits) with "Register Website".
If you are paying high rates for banner advertising, this kind of information is crucial to determining when you want your ad shown.
I personally would never pay on a per-banner-displayed ("CPM") basis, btw; only on a per-clickthrough or per-conversion basis. And don't
let the sites snow you and say they only do CPM deals. If your budget is big enough, they'll do
cost-per-clickthrough deals as long as you sign a non-disclosure in blood and promise your firstborn if you blab.
These days, however, most people should avoid banner ads, and just use PPC if you want to pay for traffic -- it's much easier to see if you're getting your money's worth.
Bottom line, if you can set up a situation where an expenditure of $X brings in $X+1 in profit, you're "in like flynn," as
they used to say. If you can't, then you'd better hope your boss buys the old "we're investing in branding"
line, else you're "out the door." The great thing about the net is that, even more so in some cases than junk
mail, one can quantify things that in traditional advertising were just wild-assed guesses. But the flip-side is
that this puts the campaign-designer's feet to the fire; there's much less "fog of war."
Me, I try to make sure that every ad I pay for generates more income than it costs, and that they all
have my incredibly memorable domain name mentioned on them. That way, any "branding mindshare" is gravy.
If you would like to reprint this article in your online or paper
newsletter, please contact me for permission.
This site was developed on a Macintosh, programmed in
WebSiphon, and served
by WebStar. The author,
on those exceedingly rare occasions when he does think, indeed thinks differently.
