Wednesday, January 31, 2007

Testing Flash with GWO

An interesting side effect in the GWO with alternate Flash files is the lack of the message, "Click to activate and use this control" that appears in every embedded Flash file since Microsoft lost its patent infringement case concerning the "embed" tag.

For the past few months, every Flash file requires that this action be taken. There is JavaScript available from Macromedia (Adobe) to circumvent this problem, but it seems to prevent the GWO from working. I resorted to removing this Javascript from the page, which then let the Google Optimizer find the section code, so I proceeded with setting up the test.

However, once the test was running, this message did not appear with the "alternate" Flash files displayed by the GWO. The message only appeared in the "original" Flash image. The alternate Flash files operate the same way that they used to - without that stupid message popping up. I don't know if this is a bug or not, but it sure is nice.

In any event, if you treat a Flash file like it is just like any other image on a page, you should have no problems setting up a test.

Just put the section tags around the "object" tags that define the Flash file, and then you can create a new Flash file with a different name for the alternate portion of the test. Put both Flash versions on your server, and then you can begin your test. You can alter the content of the Flash files after the test is running, but it can make for some confusing results since there is no way to "reset" the test counters at the moment, even if you pause the test (this may change in the future).

If you create your entire website in Flash, I'm not positive that the above will work, but if you embed the flash file in a static page that resides on your server, it will work fine.

Labels:

Monday, January 29, 2007

Goodbye Newspapers...

I feel bad for our paperboy as he leaves the last issue at our doorstep. Joey knows him by name, and she knows all of his older brothers by name as well, because they have been delivering the newspaper to her for a long time now. I was once a paperboy, and although I once delivered the newspaper to a notorious gangster, I know how it feels when a customer cancels (even though he was put in jail).

She canceled the newspaper subscription not long after the newspaper announced that they were laying off 20% of their workforce, and were planning to create a "media center" that would expand their online presence (read: money-making capacity).Right after that announcement, the newspaper itself shrank in thickness to less than half of what it had been. Even the full-color photos seemed to lose their luster...

So, after more than 20 years as a newspaper subscriber - she cancelled her subscription. And, when she called to cancel, they told her all about the new features in their wonderful new "media center" but she had to subscribe in order to get them. No incentive there, so she cancelled, and I don't blame her.

Although, I must confess that I do still feel bad for our paperboy...

Labels:

Thursday, January 25, 2007

Blogger's Custom Domain Bug

There appears to be a small bug in Blogger's Custom Domain feature that only allows a user to set either a www version or a non-www version of the domain. So, until they get it fixed, I'm going to leave this domain set to the non-www version because it is a lot shorter to type in, and that was the whole point of my snapping up brianm.com when it became available.

Sorry for any confusion this may cause you, and I only hope that the Blogger team releases a fix soon...

Tuesday, January 23, 2007

Sorry about the domain problem...

Please be patient, I'm having a little trouble with my registrar and the new custom domain service from Blogger...

Update 01/24/07 - If you put www in front of brianm.com, you will see a "Page Not Found" error. It takes a few days to propagate the change through the Internet, so I am waiting for the fix to take place.

I appreciate your patience, and hope this doesn't cause you too many problems...

Friday, January 19, 2007

End of Yahoo Answers?

Consumer Reports is retracting an article on baby car seats after its study came under attack by a number of "experts" who claim that CR exaggerated the results of the crash tests.

Why is there such a furor over this news that it makes headlines around the world and gets a featured spot on National Public Radio? At least Consumer Reports tries its best to accurately and independently test the products that it reviews.

Yet, there is no monitoring of the "experts" in Yahoo answers. In fact, anybody can provide an answer, and then readers get to "vote" on which answer is the best. Then the original questioner gets to choose which one is the best answer. So, if I want to make myself an "expert" I can get a few friends to vote for my answer, and presto! I'm the "expert!"

I'm afraid that this is a lawsuit that is just waiting to happen. Someday, somebody will take the advice of a complete stranger and then do something really stupid, and then sue Yahoo because it provided the "answer."

When that eventually happens (and it most certainly will), Yahoo will shut down its answers program (like Google did not too long ago), and then move on to the next "big" thing in hopes of monetizing its SERPs.

The "Answers" program a nice idea, but it is bound to fail, and it won't fail because the answers are not relevant. It will fail because there is no way to independently evaluate the advice of the "experts" who provide the answers.

Labels:

Wednesday, January 17, 2007

Viral marketing

"Word of Mouth" advertising, or "Viral Marketing" is the best possible marketing because a happy customer will brag to all of their friends about what a great decision they made to purchase a product. They are proud of that decision, and they will tell everybody they know or meet about how smart they were to make that purchase. It's all about bragging rights, and not at all about the product itself. And, it really doesn't matter if an item is over-priced as long as they are happy with it, since expensive items tend to take on an importance of their own - just look at the diamond industry.

However, an unhappy customer will warn everyone within their circle of influence about making the same mistake. They may be embarrassed about making a dumb decision so they might not brag as much about their mistake, but they will definitely scare other potential customers away whenever the opportunity arises. Even if they are initially embarrassed about their bad decision, they eventually overcome this embarrassment by becoming an "expert" on the product and will then scare everybody away who will listen to their "expert" advice.

In the days before the Internet, this word-of-mouth advertising was limited to small, isolated circles of influence of roughly 10 or 20 people, so an unhappy customer had very little impact on potential sales. For a marketer, there were always new circles of people to approach, and the chance of negative news leaping from one circle to another was negligible. This limited amount of contact with "experts" brought rise to organizations like the BBB and Consumer Reports magazine, which became substitute experts on products that they reviewed.

The Internet has changed all of that because that same word-of-mouth advertising can now easily reach 20,000 people or more in an extremely short time. Also, one negative comment will stay on the web for many, many years and will continue to scare people away. This is part of the "social networking" aspect of the Internet, and all of the search engines are looking at ways of making social networking even easier for people to apply.

A person who owns a product becomes an "expert" on that product, so his advice becomes a valuable resource. In the days before the Internet, you would ask an "expert" that you knew for their advice about something you were thinking of buying. If you didn't have access to any expert advice, you might be much more willing to take a risk or be persuaded by a salesperson into making that buying decision.

However, "experts" are now available all over the Internet in an instant, and they are more than willing to leave their advice in chat rooms, forums, and in blogs (the "blogosphere" doubles in size every six months), etc. These experts get a thrill from seeing their opinion in print, and they are often rewarded by the very sites that invite them to leave their "expert" opinion. There is also a proliferation of "expert" web sites (like epinions.com and complaints.com) cropping up that are supported through advertising links.

"Expert" web sites are rising in popularity, and the "social network" of experts via personalized search, tagging, and blogging are expanding word-of-mouth advertising in a major way. You can make the argument that these so-called-experts know nothing, but you will never have the chance to make that argument with potential customers who do their research on the Internet before they are willing to give you their personal information. They will quietly click away, and we will never even know anything about them. You can see them as "dropped shopping carts" in your server logs, but there is no way to know why they have abandoned their carts and left the site.

This problem will continue to grow as the Internet expands and more people learn how to take advantage of it. You can push down a lot of bad press on the web, but eventually, other links will rise up. It's like trying to stuff an inflatable life-raft back into its case. Unless you let all of the air out, pushing down in one spot just makes another spot rise up. The only way that you can stop this is to "let the air out" of all the bad press on the web.

And, I'm afraid that the only way to do this is through an "old-fashioned" approach where the "customer is always right" even when the customer is wrong. In other words - if you treat your customers badly, you will spend all of your time trying to counter the bad press that appears on the web. But if you treat your customers right, you will breed an army of positive "experts" who will market your products for you...

Labels:

Thursday, January 11, 2007

Gold SEM Advice

OK, here's a bit of SEM advice that is worth its weight in gold: When copywriting for the web, the 1st person narrative is 3-4 times more effective than the 3rd person narrative.

What do I mean? Consider the two following phrases. The first example is written in "third" person, which is a traditional "marketing" approach:

"WidgetMaker is the finest builder of custom widgets in the world, and Widgetmaker sells its widget in a direct-to-the-consumer program that saves the consumer money."

However, the following example is written in "first" person, and it can be 200% more effective in convincing somebody to "convert" or take the action that you want:

"We are the finest builder of custom widgets in the world, and we sell our widgets directly to you in a program that saves you money."

Do you see the difference? The first example talks about the "consumer" while the second example talks directly to the reader, as though you were actually having a conversation with that reader.

This may seem too simple, but it can make the difference between a 2.5% conversion rate, and a 15% conversion rate.

I am still pondering the psychology behind it because a traditional marketing approach stresses the third person, but all of the testing I have done so far shows a significant increase in the conversion rate, just by changing the content to a "first" person, or direct conversational approach.

Labels:

Wednesday, January 10, 2007

New Search Engines every day

It seems like there is a new search engine cropping up every day, but they all are lacking one basic feature: a webmaster interface!

Just look at the leaders in the search industry, and what they offer to site owners:

Google - webmaster tools, sitemaps, URL removal tool, submit feeds.

Yahoo - Site Explorer, submit feeds.

MSN - submit 1 URL.

Ask - nothing.

Cranky - nothing.

Even Google could improve their communication with web site owners, but at least they are way ahead of the rest. I must admit that I'm starting to like Google more and more, even though they sometimes do things that make me scratch my head.

But in comparison to the rest of the search engines, Google is trying to work with site owners. It's no wonder that Matt Cutts of Google gets mobbed like a "rock star" whenever he goes to a conference. He is the most vocal and communicative of all the search engine representatives.

I just wish that Danny Sullivan's idea of a "Matt Pass" would catch on. I could use one right about now...

Tuesday, January 09, 2007

Need More SEO / SEM advice?

Here is a monster list of SEO/SEM blogs:

http://www.toprankblog.com/search-marketing-blogs/

Who knows, maybe you'll see my blog listed there in a few weeks...

Monday, January 08, 2007

SEO versus SEM

SEO is getting your web site to be visible in the search engines.

SEM is getting a visitor to your website to take the action you desire.

To me, it's all a puzzle that needs to be solved.

I must admit that I'm getting a little bit jaded by SEO, since I've been doing it for so many years and been so successful at it. So, please forgive me if I drift off that subject and spend more time on SEM. But, when I see the conversion rate on a page go from 2.5% to 15%, I tend to get a little excited and I start thinking more about SEM. After all, a 2.5% conversion rate would require 6 times as many visitors to equal the same number of customers as a 15% conversion rate.

But, what happens when the conversion rate does not change? Or, when it actually declines? That's when I get even more interested, because there are no tools out there that will help me fix the problem. Instead, I have to rely on my own brain to figure out that puzzle.

The Google Website Optimizer (GWO) sometimes shows disappointing results, and the decline in the conversion rate can be pronounced. I suppose that I could blame Google and say that the GWO sucks, but it really is just a tool that is showing me the results of my SEM efforts, even when the conversion rate declines.

So, now what? How do I increase the conversion rate when the GWO shows a decline? I ask myself this question:

What is the searcher looking for?

I admit that this question causes a lot of speculation on my part, but once I have a theory in mind, I need to test it and that's where the GWO really shines. I create another test, using my best performing combination as the baseline (or "Original") and try some new combinations to see if the conversion rate improves. It's a very slow process (I've seen grass grow faster), but the payoff is definitely worth it, and it's much, much faster than other methods.

Of course, it always helps to have a great tool when solving a puzzle...

Tuesday, January 02, 2007

Unofficial Website Optimizer Expert

I seem to have become the Google Website Optimizer expert on SEW, since there is currently no way to communicate with Google about this, and I am out here testing it madly. I must admit to being a huge fan since it has helped me to dramatically improve the conversion rate of visitors. I currently have four tests running on one site, and will leave them in place a little longer, but the following data has surfaced as a result so far:

1st person seems to be taking the lead in convincing people to convert. The traditional marketing approach to copywriting leans toward 3rd person, so you should use the company name and avoid things like "our" or "we" etc. However, I can see a 200% increase in conversion when I simply make 1st person changes to existing text...

The other unexpected increase comes from image variations, which are the easiest to set up in a test. It is surprising how much a change in an image can affect the conversion rate, and even if you see a downward trend, you can learn a lot from the test.

Tip - There is currently no way to restart a test once it is stopped. However, the only thing you need to start a new test is the utmxkey number (which you get when you create the new test) and you can copy that new utmxkey number over the existing utmxkey code number on the pages.

So, if you have any question, I will be happy to respond either here or over on the thread at SEW:

http://forums.searchenginewatch.com/showthread.php?threadid=15480

Update - 01/11/2007: Image testing is now much easier because you can just swap the image on your own server. In other words, if you create two images, such as, "image_1.jpg" (original) and "image_2.jpg" (variation 1) then see that "image_2.jpg" converts much better, there is no need to stop the test - you can just copy it over to make it "image_1.jpg" and create a new "image_2.jpg" and then watch the results with creating a new test. Just remember the date that you made the change so you can factor that in the results...

Labels: