Buying Advertising & The Truth About Website Statistics

9 01 2008
651.gif Being the webmaster of a site with over 400 direct advertisers (who are therapists), I often get asked about stats, and currently I don’t have anything set up to deliver stats to my directory’s advertisers. The population I advertise for are therapists who, for the most part, are relatively new to internet marketing. Since I also manage a few individual marketing campaigns for therapists, I have a good idea about the kind of traffic that converts into referrals for them, and the kind of traffic that doesn’t. I even get to see how good my site’s competitors do at sending conversions to therapists. It’s a really unique vantage point, and here’s what I’ve come to learn from it.

The truth about what people are selling you in terms of advertising, is that the bottom line is the price you pay for each conversion. In the case of the therapists who advertise on my website, a “conversion” is a referral for counseling services. With that being said, statistics on visitors and pageviews that people send their advertisers can potentially be meaningless due to the fact that conversions are highly dependent on how targeted the traffic is.

Targeting Traffic

Let’s start with the concept of targeting, “How Targeted is the Traffic?” The truth about traffic, is that you can literally buy traffic for a website through a number of sources, but different traffic will convert into a referral at different rates. A case in point was an experiment I did one time when I bought cheap advertising through a lesser known pay-per-click service. I got a lot of “traffic” from the service, people where clicking on my relevant ads and visiting my website, and the cost was less than what you would pay for clicks through Google or Yahoo! Sounds good right?

It wasn’t good at all. The reason being, that the traffic had nearly a 100% “bounce rate;” in other words, those people hit the homepage of my website and left from there. They weren’t at all interested in what my site had to offer. Would you want to advertise with a site that got that kind of traffic? There are a lot of ways that companies can fail to deliver conversions to their advertisers and poor targeting is one. Another common mistake is by not managing pay-per-click campaigns properly.

Case in point: See the attached screenshot for a Google search I did for “massage therapists in New York.”
massage-therapists.gif
The circled ad on the right is for a relationship therapist who hasn’t added the negative keyword “massage” to his pay-per-click campaign (which would keep him from showing up for this search). Mistakes such as these are common, and can cost advertisers a lot of money because consumers don’t always realize that the ad isn’t for them until after they click on it.

My purpose in bringing up targeting is to demonstrate that 100 visitors, doesn’t always equal “100 visitors.” I wish I could stop there, but there’s more. Another reason “100 visitors” doesn’t always equal “100 visitors” is that there are different ways of defining a visitor to a website, and the same goes for PageViews! Unfortunately, most stats packages define these concepts in different ways that can result in significant differences amongst the stats they report.

Differences Among Stats Packages

Some of the most common methods website owners use these days to look at their stats are log file analyzers that take raw information from the web server’s logs and translate it into useful tables and charts (like AW Stats and Webalizer), and tracking tools like Google Analytics which use javascript to track your site’s visitors. While all have their advantages and disadvantages, the numbers from each can be very different. Since Google Analytics uses a javascript, only visitors with web browsers that handle javascript will be tracked, so those numbers can be more conservative. On the opposite end of the spectrum, log file analyzers like Webalizer and AW stats, will frequently count bots that are spidering your site for different reasons as “unique visitors,” which can make those statistics grossly exagerated. There are a lot of other stats packages out there, but these free ones are commonly used and sufficient for most websites. The important thing to be aware of though, is that the manner in which a stats package defines a visitor or pageview can vary significantly amongst them.

These differences also hold true for ad serving software. One of the most common tools people use to send stats to their advertisers is an auto-generated report that comes from the software they use to serve ads on their website. A lot of online ad sales are done on a “CPM” or “cost per mil” basis, with a “mil” being 1000 ad impressions. In order to estimate the kind of value you will get from such advertising, one needs to know how that advertiser defines an “impression.” Some ad serving software will show a new impression to a visitor who has been on the same web page for an extended period of time. Typically, this happens when the banner cycles through to another advertiser then back to your banner, but not always. Is that something you want getting counted twice? Do you have a choice? Probably not. What about if the visitor refreshes the web page?

Confused yet? Me too. That’s why I take what people tell me about their website’s stats with a grain of salt when I buy advertising. It’s a trial and error process for me that I have learned to evaluate effectively over time.

The Bottom Line

The bottom line is, you need to “level the playing field” by evaluating your own stats. You can easily compare different advertising by looking at the referral stats for your own website. Many therapists I talk to have websites and don’t realize that they already have a stats package installed – that their designer or web host just needs to provide them access to it. Once there, you should be able to look under a “Referring Sites” category to see where your websites visitors are coming from. Compare the amount you spend with how many visitors you get from each advertiser.

Track your conversions as well. For therapists in particular, if you have a “contact form” on your website, you can track conversions by entering a “Goal” (the URL of your “Thank You for Contacting Me” page) into Google Analytics. With this information you should be able to easily see which sites are sending traffic to you that results in a referral. Conversion tracking is an absolute necessity for any serious marketing campaign; it can mean the difference between “making thousands” or “spending thousands.”

But what about phone calls? How can I track those?

First of all, if you have a website for your services, it is important to have your phone number in a prominent place (near the top, and BIG). The easier it is for people to contact you, the more likely your website will be to convert visitors into referrals. Keep a spreadsheet handy to record new referrals you get, and politely ask, “Do you mind me asking how you found out about my services?” If they say “your website,” also ask if they found your site through a search engine or another website. Sometimes people will be able to recall, and the information can be very useful in determining if your advertising dollars are being well spent. This information can also be gathered through new patient intake forms.

The internet is one of the best ways to advertise for many different professions, products, and services. For therapists in particular, the demographics of internet users are perfect for getting quality referrals. Your money can go a lot further on the web than through traditional means such as yellow page or print advertising. With that being said, this form of advertising is relatively new compared to traditional means, so many are not familiar with the concepts involved. Hopefully my explanations will help.

Actions

Informations

8 responses to “Buying Advertising & The Truth About Website Statistics”

9 01 2008
Niall Doherty (12:47:01) :

Interesting insight. I didn’t realize the limitations of Google Analytics. Thanks for sharing.

11 01 2008
Sergey Rusak (03:25:41) :

As for me, I trust only naturalk search engine results. I know that every 500 – 1000 people who finds me through Google become my client and pays me from $500 to $10 000 for SEO services.
Once my post went viral and was shared by thousands people. Over 15 000 unique visitors visited my site in less than 6 hours and I got no clients from this useless traffic which came from Digg, StumbleUpon, Reddit, Delicious etc…

11 01 2008
Chris H. (07:56:17) :

Yeah social media sites seem to be known for that…

12 11 2008
Estes Therapy (18:58:00) :

You definately know the market and how to convert visitors into hits. I advertise on your therapy site and find that I get a lot of traffic from it. You know your work and what helps! Thanks for the internet help!

14 03 2009
Webrecsol Optimizer (03:24:09) :

it is a interesting topic and social media is a way of web site increase traffic..

14 03 2009
Web SEO (03:28:10) :

when i saw this site then i rode start site.nice sharing of internet marketing related..

27 04 2009
Billigflug (07:20:05) :

Excellent read and useful information.
Because of the financial crisis, I will lose my job in few weeks. So Intended to have a home based business and to launch a new web shop. That’s why I was looking for websites with high trafic rates to advertise my merchandise, which is italian shoes. Thus, refered to the Google Analytics stats and chosen 3 websites in order to contact them later. But I have not really payed attention to the bounce rate thing before.
SEO seems to be more complicated than I have expected.

23 05 2011
Should I trust stats from therapist directories? | New Orleans Internet Marketing (09:41:10) :

[…] do you know who use the Alexa toolbar? If you have the time, read my earlier article about “buying internet advertising and the truth behind website statistics.” The Family & Marriage Counseling Directory (http://family-marriage-counseling.com) has […]

Leave a comment

You can use these tags : <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>