How can Taguchi testing improve my landing page conversion rates?

Taguchi testing can multiply landing page conversion rates

”Test, test, test.” That’s what catalogers and other direct marketers tell each other at all the conventions. But many of them do very little, if any, testing.

Meanwhile, the big consumer mailers - Publisher’s Clearinghouse, Phillips Publishing - continue to test, and test, and test.

Smaller and medium-size companies often say they have no budget, no time, or a list too small for testing to pay off.

Even big marketers often run simple A/B splits. One variable at a time.

Split testing is also less common in print space advertising. Most magazines and newspapers do not offer or encourage it.

But thanks to the Internet, some of the lost vigor in testing is coming back in the form of a method called Taguchi testing.

Okay, so you use Taguchi testing. You know all about it. Nothing to see here. Anyway, on the slight chance that you pick up an idea or two, read on…

For newcomers to Taguchi, you could be in for the biggest thing to happen to Internet marketing this decade.

To be fair, this discussion will not be a mathematical dissertation.

This will be a general overview.

Testing Many Variables at Once

Taguchi testing is a system of testing many variables at once on a landing page or some other online direct response promotion. It is inexpensive and fairly quick.

David Bullock is president of Results Squared. This company offers Taguchi testing to direct marketers. Typically, he says, they run 3 pre-heads, 6 headlines, 3 subheads, 3 salutations, 3 lead paragraphs, 3 visuals, 3 guarantees, 3 calls to action.

Of course, you can test other items too. Bonus items. Prices. Bullet copy. Bullock often tests traffic source (organic search, Google Adwords, e-mail) and a factor he calls predisposition to purchase.

Predisposition to purchase is how much the prospect already believes in the offer before he or she comes to the landing page. For example, loyal readers sent by a joint venture partner’s e-mail are more ready to believe the offer than those who arrive via a keyword search.

In brief, here’s how it works: Taguchi testing software shows each visitor a unique combination of elements being tested on the landing page.

Then results are tabulated and analyzed. And at the end, you get a report that shows which version of each of the elements tested did the best. Best headline. Best lead paragraph. Best visual.

Taguchi Testing Would Beat an A/B Split

Taguchi testing would beat an A/B split any day of the week.

The conversion rate on the landing page goes up a notch every time you optimize a variable. Best headline. Bump. Best lead paragraph. Bump.

And here’s the magic: when you put all the versions that came out best in the final landing page, your conversion rate can double, triple, even quadruple.

Before You Test

Before you test, you need a complete text of the landing page. Plus every variable, all the different versions you want to test. The six headlines, three subheads, and so on.

So you take the copy, give it to a Taguchi testing company, and then the company runs the test.

In the old days, a direct mail test took weeks, maybe a couple of months to get valid results. And most of the time it was just two options. Two packages. Two prices. Two headlines.

However, when you do it online you can work a lot faster. David Bullock says he can do a full test and verify the results in 3 to 5 weeks, depending on the traffic and conversion rates.

The number of sales or responses (not traffic or how many you mailed) is what makes a test statistically valid. How much traffic you need depends on your conversion rate.