MarketingExperiments Blog: Research-driven optimization, testing, and marketing ideas

In a recent Wall Street Journal article, Lee Gomes describes his search through 2.27 gigabytes of search data from AOL.

According to his research, the single word which appeared most frequently in searches on AOL over the period covered was “free”. The second most popular word was “now”.

It’s curious how some words and short phrases become clichés when over-used, while others, like Free, retain their meaning and punch. It seems that however often we see “Free”, whether on product packaging, TV ads or the web…we still find the word compelling.

However, simply including the word in your headline is not enough to maximize conversions.

Currently we are working on our next webinar…which happens to be free…and the topic is “Optimizing Free Trial Offers”.

We will be sharing test data collected from tests with two of our research partners. In both cases we show how a free offer page can be optimized to deliver far better results, without having to change the offer itself.

In fact, in both cases, conversions were almost doubled simply by making careful changes to the design and copy.

The webinar is being held on Wednesday, August 23rd at 4:00PM EST. You can sign up here.

We are working right now on our next free teleconference call, scheduled for August 9th at 4:00PM EST.

We’ll be presenting a merchandising calendar for the last five months of this year, with particular emphasis on the last eight weeks before Christmas.

In preparing the calendar, we asked ourselves just how well prepared online merchants are when it comes to the major retail buying season later in the year.

- While every online retailer knows that the eight weeks before Christmas is the busiest buying period of the year, are they truly prepared…right down to individual product promotions?

- Are fast growing companies certain that their technologies and platform are ready to handle the holiday spike in sales?

- Will they have sufficient customer service people trained and ready to serve by the time the first backorder or product availability problem occurs?

- Are marketing groups preparing promotions and PPC campaigns to make the most of this essential time of the year?

Our own research indicates that while a handful of online retailers have become quite sophisticated at merchandising for the holidays, many other companies have yet to put in place a marketing blueprint and merchandising calendar specifically designed for this perennial opportunity.

In this brief we will help you to develop such a blueprint and calendar, customized to your specific industry and target markets, and prepare you for what should be your most profitable period of the year.

If this topic interests you, please sign up for the MEC Merchandising Calendar for the Fall and Holiday Season 2006 webinar.

We are working right now on our next teleconference clinic. It’s on the topic of web site metrics.

As we study the various PPC reporting tools, and a variety of web analytics services, it becomes more and more apparent that inaccuracies are not uncommon.

For instance, if you track a PPC campaign through both the PPC engine reporting tool and your own web analytics program or server logs, you will likely see some inconsistencies.

This can become a significant problem if you base future marketing decisions on flawed data.

In addition, we will be looking at the problem of being overwhelmed by too many metrics. Analytics services are differentiating themselves by offering more and more features and reports.

We address this problem by stripping away all the metrics you don’t need, and focusing just on those that are fundamental to your business model.

In other words, we are isolating the metrics that matter.

If you would like to take part in this free teleconference call, please register here.

During the call you will be able to download a simple dashboard tool to help you identify key metrics for your own business.

As I mentioned a couple of posts back, we are putting the final touches to the content for tomorrow’s teleconference call, Essential Metrics for Online Marketers.

Late yesterday we added some new data for the call, based on an experience we have had with the metrics from a third-party reporting tool.

On one line of data, if you attend the call, you will see that one of the figures reported to us was off by 900%.

The most disturbing element here is that we were already preparing changes to the site based on the figures from the third party.

Had we not decided to double-check the figures independently, from the raw data, we would have gone ahead and made some very damaging changes to the site.

The moral of this story is that you always need to verify data from third-party sources.

We will be covering all this in full detail in the call tomorrow. We have also prepared an Essential Metrics tool which you will be invited to download at no cost.

Sign up to reserve your place on the call here…

As you may already know, we have been busy working on a series of certification programs.

The first one, which started in April – and we’ll be running it again, starting in June – is the Foundations of Online Testing course.

During the first course we found that the topic people wanted help with most was determining the validity of test results.

Right now we are working Dr. David Reiley, Professor of Economics for the University of Arizona, to develop a theorem for calculating validity with a small sample size that is based on making a prediction prior to running a test.

We believe this will allow marketers to move more quickly as they test, and to make sound decisions with less information.

In the interim, we have put together a PDF file which takes you through the steps to quickly calculate test validity.

Feel free to download your own copy of Determining if a Data Sample is Statistically Valid.

What are eco-ads? They spring from marketing strategies, often within large corporations, which address the need to present companies as being friendly towards the environment.

Companies recognize that many of their customers are concerned about the environment, and want to be seen as making positive moves in the right direction.

A number of examples of these eco-ads are featured in a recent article in The New York Times.

What is the cautionary element here?

In a nutshell, if you say it, you need to do it. In other words, you can’t just write an ad about how you are taking steps to protect the environment, you have to actually take those steps.

This issue of transparency is addressed in a lot more detail in Dr. Flint McGlaughlin’s essay on Transparent Marketing.

If you don’t walk the talk, someone will catch you. This is exactly what happened when Starbucks publicly promised that anyone walking into any Starbucks coffee shop would be served Fair Trade Coffee on request.

The announcement was made to please supporters of the Fair Trade Coffee movement. Unfortunately, not all the store managers were told about this. So when people asked for their Fair Trade Coffee, in many cases none was available.

And before you could say, “Darn that Internet”, the Starbucks Challenge had been created by a couple of bloggers, and hundreds of people around the world walked into Starbucks shops and deliberately asked for Fair Trade Coffee. They then posted their experiences online for everyone to read.

The moral of the story is that promises relating to the environment generate a lot of passions and expectations.

Few people will argue against the benefits of large companies supporting environmental issues. And eco-ads are part of that.

But you need to follow the path of Transparent Marketing. And you have to keep your promises.