|
How many of your visitors are converted to sales? What percentage of your
visitors that land on your landing page purchase your product?
In a prior message we talked about Ad Tracking and how that would
benefit you. Here you're going to learn about a testing method that will blow
your mind. We'll get to that in a minute.
But first, let me refresh your memory about testing. One of the most
important things you can do as a marketer, and everyone is a marketer at some
level, is test. Testing involves changing all the different aspects of your
sales message and finding the combination of headline, offer, body copy,
prices, freebies, etc. that produces the maximum result. The only way you're going to
find out what is your maximized result is by testing. You may be doing well now,
but what if you could increase your sales by 20%, 50% or more? What if it's only
10%, that's money that's going straight to the bottom line. As we demonstrated
in the previous note, even a 1% increase in sales can result in a 300% increase
in you net profit!
One of the tests I ran for another product I sell told me that my clients
were more willing to pay $397 for the product than the $297 I'd been selling it
for for last 5 years. I increased my revenue by 1/3 without doing anything other
than test. Over the 5 years and hundreds of sales, I left thousands of dollars
on table.
One of the challenges with testing is the amount of time it takes to test all
the possible combinations and permutations. You may go through a dozen or more
headlines, several different prices, many offers and so on. In order to
determine the "winner" of a test, you need hundreds of tests and results. This
process can literally take months or even years to achieve the results you want.
It's the right thing to do, but it takes a lot of patience and money.
There's another way of testing that's been used for years in engineering and
production environments. It's been recently introduced to the marketing arena.
This testing method allows you to test multiple variables simultaneously. For
example, you could test six different headlines, three different prices, three
different offers, three different sub headlines, and two different PS's
all at the same time.
What does this mean to you? It means that one test over a much shorter period
of time than it takes with A/B testing you can find the result that optimizes
your results. It also means that may find a combination two variables that may
have not been tried together in A/B testing produces better results. Results you
would have never found through through straight A/B testing.
This method is called "The Taguchi Method". As I noted earlier, Taguchi has
been around for years in engineering and manufacturing environments. It's one of
the reasons that that Japanese car manufactures have been so successful in
producing excellent cars.
It was introduced to the marketing environment by
Dr. James Kowalick.
Originally, he was selling his methodology for $1,000. A while back he released
a Excel spreadsheet into the public domain that performed the calculations that
tell you which elements and what variables within each element produce
statistically significant results. He no longer has this on his web site, but I
have a copy. You can
get a free copy of the
Taguchi Spreadsheet here. I have to admit that it can get somewhat technical. Just take a deep
breath and dig into it. It will be well worth your time.
You can purchase programs that will do Taguchi testing, I've personally used
this one very
successfully, or you can use the Taguchi testing that's free on eCOMpal. I
will tell you that it's more difficult to build your pages with eCOMpal.
Depending on how you set up your tests, it will take an hour or two to set up
your pages. You can test 18 different sales pages (that's the normal number of
pages for Taguchi testing) and we'll keep track of all the clicks and
conversions. You can easily copy the information into Kowalick's spreadsheet to
determine what are the statistically significant parameters and the values you
should use going forward.
Things you can test are Headline, Guarantee, Price, subheadline, first
paragraph, PSs, Free goodies, Fonts, body copy, placement of testimonials, audio
comments, video presentations, etc. Anything you can think of can be
tested. For example, in a test I ran showed that an Arial font headline
outperformed a New Times Roman headline by a substantial margin. For the body
copy it didn't make a substantial difference.
One thing I've discovered with this is that you have to have a
large quantity of clicks/sales in order to be most effective with
this. I'm talking in the thousands of clicks. If you're only
getting a few hundred clicks/sales a month regular A/B testing works
best.
|
