Product: Article Marketing Automation (AMAutomation)
Website: AMAutomation.com

In the training videos of AffioBlueprint, Mark Ling recommends joining AMAutomation. It’s a paid service (can try for free) that has two parts:

1. A network of blogs where they’ll post your article. (You can add your blog to this so that you can get content, but it isn’t required to take advantage of the second feature).
   
   
2. A function to ’spin’ your article into different versions, thereby getting around Google’s duplicate content rules.

The first function, posting to a blog, is interesting. The main purpose for doing this is not to get traffic but to get backlinks. Each time someone carries your article on their blog, you get a backlink (assuming that you’ve put links into your article text).

It is similar to when you submit articles to an article directory, but your article is totally anonymous. You don’t put a bio box at all. When someone else decides to carry it on their blog, it looks like they wrote it.

But I’m more interested in the second function of this service, spinning an article.

Here’s what I mean. Suppose you’ve written an article that starts like this:
What you do in AMAutomation is imbed spin tags in your text and add alternate wording. One type of tag lets you specify an alternative sentence, like this:

Recognize the potential here? With the example above, we already have multiple versions of the first two sentences of the article:



If my math is correct, just those few spin tags will generate 12 versions of my sentences.

Then when you submit this, a different variation becomes available to the network. Blog owners can choose to display it or not and there’s no problem with duplicate content if more than one blog carries it.

Unfortunately there is no way to tell which blogs are displaying your article, but you do get a count of how many sites. You also get a count of how many rejected it, so this will tell you if you need to write a better article.

So, all this is well and good, but does it work?

Well, I just signed up for a trial account and forked over $47.00. The site I set up using the AffiloBlueprint methods does not have any traffic from search engines so far. I’m going to start spinning articles, posting them to AMAutomation, and then see if I get backlinks and traffic.

Later…I’ve posted my first article. AMA submits the article to blogs over time so that Google doesn’t see a huge increase of back links overnight.

So here are my results (Day 1 is the day I published it to the network):

Article 1
Day 1: Total blogs who are publishing it: 0. Ranking on Google: n/a.
Day 2: Total blogs who are publishing it: 2. Ranking on Google: n/a.
Day 3: Total blogs who are publishing it: 6. Ranking on Google: n/a.
Day 4: Total blogs who are publishing it: 6. Ranking on Google: n/a.
Day 5: Total blogs who are publishing it: 10. Ranking on Google: n/a.
**Fixed a mistake that caused my pages to not be indexed.
Was also offline for a few days
Day 9: Total blogs who are publishing it: 14. Ranking on Google: n/a.
Day 24: Total blogs who are publishing it: 18. Ranking on Google: n/a.
Day 29: Total blogs who are publishing it: 19. Ranking on Google: n/a.

Article 2
Day 1: Total blogs who are publishing it: 0. Ranking on Google: n/a.
Day 2: Total blogs who are publishing it: 3. Ranking on Google: n/a.
Day 3: Total blogs who are publishing it: 6. Ranking on Google: n/a.
Day 4: Total blogs who are publishing it: 10. Ranking on Google: n/a.
**Fixed a mistake that caused my pages to not be indexed.
Was also offline for a few days
Day 8: Total blogs who are publishing it: 15. Ranking on Google: n/a.
Day 23: Total blogs who are publishing it: 17. Ranking on Google: n/a.
Day 28: Total blogs who are publishing it: 17. Ranking on Google: n/a.

Posted by susb8383 under Misc. & Recommended Products | 2 Comments »

I always though that calculated shipping is a good thing. When the visitor puts in his zip code and gets a shipping quote, it seems to him that he’s being charged the true shipping cost and not some arbitrary amount.

But I know for a fact that I lost at least one sale and possibly more because I used calculated shipping.


I created an eBay auction and I set it up so that it uses a shipping calculator for USPS Priority Mail. I got an irate email from a potential customer asking if I really thought people were going to pay 56.00 for my item. (My item was 35.00 plus shipping).

It turns out there was a major problem with eBay’s shipping calculator. When I went to the auction and put in a California zip code, it came to over 35.00. The same zip code on the USPS site was 11+. The visitor who sent me the email was seeing 26.00 for shipping.

I had noticed the same odd behavior a few weeks ago using Paypal. I clicked on Create a Shipping Label to send an item to Florida via USPS Priority Mail and the price came up as 25.00. The same address on the USPS website to create a ClickNShip label was 11+.

I immediately changed the shipping to a fixed price, but who knows how many orders I lost.

Even if the calculated shipping doesn’t malfunction you could lose a sale. On my Zen Cart site I had a shipping module that calculated Priority Mail costs by using an interface with the USPS shipping server. But I got an email from a potential customer saying he was trying to order but my site couldn’t come up with a shipping rate for his town in California.

I tried myself with his zip code and it worked fine. So I think what happened was the USPS shipping server was down or a connection couldn’t be established when he tried to order (it was close to Christmas).

I immediately emailed him that it should work fine now, but he never returned.

I’ll never again use a calculated shipping module.

Addendum: Well, the case of the mysterious shipping charge is solved. It turned out to be operator error. The dimensions of my box are 12 x 12 x 18. The last time I looked at the USPS requirements, this did NOT constitute what they consider to be a large box (and therefore a higher charge). Their requirements were if the largest girth of the box (in my case 12 + 18 + 12 + 18) plus the length (in my case 12) is greater than 84 inches, it is a large box. (My box adds up to 72 inches).

But somewhere along the line, they added this phrase, "or is bigger than 1 cubic foot (12 x 12 x 12)."

My box is greater than 1 cubic foot. On my website and on eBay’s auction, it asked me for the dimensions and it calculated the charge for a Large Package. But when I went to the USPS site myself, I just selected Package.

It’s interesting to note that I saw a post from another person with the same problem. His box was bigger than 1 cubic foot so the shipping calculator gave him a Large Package price, but when he brought it in person to the post office, they charged him the regular package price.

For me, the solution is to use FedEx or UPS from now on.

And I’m still not going to use calculated shipping again.

Posted by susb8383 under Misc. | No Comments »

I was listing to a teleconference call with Perry Marshall and a guest (sorry, don’t remember who the guest was.)

(Here’s a little tangent: If there’s just a single thing I can recommend to people starting out, it’s to get on Perry’s Marshall’s mailing list. He’s a Google Adwords guru, but even if you’re not using Google Adwords, the general marketing advice I’ve gotten from reading his emails has been invaluable. I recommend signing up for his free 5-day Google Adwords email, which will put you on his list. Then if you ever get an email about a conference call (they’re usually free), SIGN UP FOR IT! The information and advice they share on these calls is absolutely amazing.)

Anyway, this speaker said you’ll be way ahead of the game if you can find out what your potential customer wants. This sounds pretty obvious, but very few people do it.

So my mission is to set up an exit survey using a hover. A hover is like a popup but isn’t blocked by popup detectors because it is part of the page itself. My plan is to only show this survey to people who are leaving my site without buying anything. This should give me valuable insight into why I’m not getting any sales.

Now, before you get too excited about this post, I’ll tell you right up front: it didn’t work. It’s impossible to create a professional-looking survey that pops up when they leave without buying. Why? You’ll have to skip ahead to Step 3 below to find out.

But since I already started documenting the steps, I’m going to leave the rest of this post intact so you can follow along with my reasoning. And also because I’m too lazy to delete the stuff I’ve already written.

So, first thing I did was come up with survey questions. On the discussion with Perry Marshall, the guest recommended these questions:

What were you hoping to find today?
Did this website provide what you needed? Why or why not?
How hard is it to find that?
What caused you to go looking for that today?
Are there any other comments you like to share?

The first thing I did was try to find free hover survey programs. Couldn’t find any. The closest I could find was one that makes the visitor type in a password to take the survey (unless you pay for their premium version).

But I want the visitor to have as few clicks as possible. I want the questions to show up as soon as they try to leave, and I only want them to click a submit button.

So I’m going to build it by hand.


Step One: Create a webpage that has the form

First I created a test webpage with my form on it. To create my survey, I used CgiEcho, a free form handler that comes in the CGI Center of my cpanel.

Here’s my test survey page:




Step Two: Put the Survey into a Hover Box

To do this, I’m going to use a free online hover box creator.

I copied/pasted my test webpage survey into the hover box content. When I ran the preview I noticed that the text of the questions is right up against the edge of the hover box, so I added a div with spacing around it.

Here’s the code (I ***ed out my form action file):
I changed these default settings:
Y target: 30 (otherwise the submit button is below the fold).
Title font: Verdana
Hover ad slide in speed: 800
Content area background color: F8EDEC (to better match the color of my site)

My hover box looks like this:



Step Three: Activating the Hover When Visitor Leaves Without Buying

This is the most crucial, and one of the key reasons I’m using Zen Cart. Since I have full access to my html, I can edit the body tag to make this happen.

Ok, so here’s what I discovered in my attempt to make this work.

My goal was to have this survey “hover” its way in (so pop-up blockers won’t detect it) when the user leaves my site without buying something.

I thought there would be lots of software out there to do this, but I hardly found any. There’s a good reason. It is very problematic to do this correctly.

You Can’t Only Show Survey When Visitor Leaves Site
In general, you need to put something in the body tag of your page which calls your function to show the survey. But, the trick is that you only want to show this to someone when they leave your site, NOT when they leave the page but stay on your site.

How annoying would this be: your visitor is reviewing a product page, then he clicks on your shipping policy and he gets something that says “Why are you leaving without giving us a try?”

I did find a post that tells how to do this somewhat. Here’s an example.



The above example uses the onunload event which fires your code after the visitor has navigated away from the page but before it disappears. You set a variable “exited” to 0 (false) when the user clicks on an internal link so that the hover won’t show; otherwise set it to 1 (true) for external links.

Problem 1: You have to edit every single way the user can get off your page so that you can set the exited variable. You may be able to do this for internal links on a very simple webpage, but what if you have a complex site such as with a shopping cart? The user can click on a product, click on a policy page, click on an image, logo, etc. It’s a lot of work to try to catch all of these instances. And what if the user clicks on the back button, a bookmark, or closes the browser? Some ways a visitor will leave your site, you have no control over.

Problem 2: There is no way to cancel the original unload request. So what I found is that when I tried to do the above for a dhtml hover box, it would flash and then disappear as the page unloaded because it is part of the page itself. There was no way to get the hover survey to stay around long enough for the user to fill out the survey. The above might work ok for a popup because it is a different page and when your original page unloads, the popup is left behind. But I don’t want a popup that will be blocked.

So this led me to the second technique: using onbeforeunload. This is similar to the onunload, but onbeforeunload can be set so that the unload is canceled.

But there are two problems with the onbeforeunload.

Problem 1: Not all browsers support it.

Problem 2: When you use it, it automatically brings up a dialog window. The first line says “Are you sure you want to navigate away from this page?” and the last line says “Press OK to Continue or Cancel to stay on the current page.”

This text cannot be altered in any way. However, you can pass it text that it includes in the middle.

But I think this text will be confusing. Here’s an example:



Have you spotted the problem? The text that can’t be changed (the first and last line) are from the point of view of someone who has initiated an unload. So “Ok” means “Yes, I want to leave the page.” But I’ve asked the visitor to take a survey. So most likely, the visitor will think, “Yes, I’ll take the survey” and will click OK, resulting in the window closing.

And even with the beforeOnUnload, you still have the same problem; having it only fire when someone leaves your site, not leaves your page but stays on your site.

Alas. Well…this was very educational. And now I know why there is no software to do this. (I did actually find one post where a very techie guy said he developed something that did this correctly by creating a hidden window that constantly checks to see if you’re still on the site. If you’re not, it pops up the survey. I wish this guy would market this into a product).

I decided instead to

1. Put in a Live Chat box so it’s easy for someone to ask me a question. I found some free software that I’ll put in another post.
2. Put a really prominent banner at the top asking someone to fill out a survey. If it is clicked on, it will open a new window (but still not a popup.)

But that’s the subject of another post…

Posted by susb8383 under Misc. | No Comments »