The art of losing money in very small amounts – Malcolm Auld’s magazine article
Many people get confused about what is meant by the term direct marketing. They wrongly assume it is limited to the medium of mail. And they often mix up the term direct mail and direct marketing. Direct marketing is a way of marketing, in all media, not just mail. It is any marketing where you communicate directly with individual customers or they respond directly to you.
Or as I define it in my book Direct Marketing Made Easy—finding out what your customers want and delivering it to them profitably, one customer at a time. Another definition for direct marketing, which my UK colleague Drayton Bird first stated to me over a drink or two is “the art of losing money in very small amounts.”
This is quite an apt definition, because one of the unique factors of the discipline of direct marketing is that it is completely measurable. You know what works and with a bit of analysis, why it works.
Consequently, you can test your communications in numerous ways before rolling-out to your complete audience. Or in other words, you can know what is going to work, rather than guess.
By investing in a testing programme you may lose small amounts of money up front, so you make large amounts at the back-end. Hence the definition; “the art of losing money in very small amounts.”
Testing can save and make you money
Sadly, most marketers don’t test their marketing investments. They just create their executions, then blindly send them in the hope they will work. They usually have no idea if the message has any chance of doing what they want it to do—because they haven’t taken the time to find out through testing.
I was once involved in the launch of a new service for Qantas Cargo. As Qantas didn’t know who the decision-makers were for international freight services, a large proportion of the budget went into creating the database via list rental and telemarketing. We discovered that the decision-maker varied enormously from receptionist to managing director.
We created an elaborate 3-D mailing that included samples of the new documentation, a how-to guide, letter, rate card, stickers, contact details and an offer for respondents. We decided to test the mailing prior to rolling out to the complete database. We mailed 1,000 packs around the country and sat back waiting for the flood of responses.
Silence of the dms
But nothing happened. The silence of the telephones was deafening—and you can only blame the postal service for a few days until no reply cards turn up. So we leapt into action and called the prospects who had received the mailing.
It didn’t take long to discover the problem—and justify our test. We learnt that the prospects we wanted to reach didn’t buy their services from a rate card. They negotiated contracts and held tenders for their business. The last thing they’d ever consider buying from was a rate card. When they received our mailing, they opened it and read it with enthusiasm.
They were keen to try the service until they saw the rate card. It showed the standard casual rates—which of course are more expensive than contracted rates. So they threw our carefully crafted mailing in the rubbish bin.
Immediately we set about opening the remaining mailpacks and removing the rate cards. Then we tested the mailing again—this time without the rate cards. The results were much more positive and we successfully rolled-out the campaign.
If we had researched the mailpack with prospects before the campaign we may have picked up the subtle point about the decision-making process. But the testing still saved us from wasting an enormous amount of time and money, and probably a client.
Market research is useful to gain knowledge about ideas before putting them in the market. But remember not to fall into the trap that many do with their research results. Don’t use your research results as a drunk uses a lampost—for support rather than illumination. Use research as a guide, because what people say in a research group and what they do in real life are often two different things.
The real acid test for your communications is to run them in the marketplace and see what your prospects do.
What to test?
Only test the big things—those elements that have the most leverage on your results. These elements include:
Many people get confused about what is meant by the term direct marketing. They wrongly assume it is limited to the medium of mail. And they often mix up the term direct mail and direct marketing. Direct marketing is a way of marketing, in all media, not just mail. It is any marketing where you communicate directly with individual customers or they respond directly to you.There are many different ways you can configure your product or service—standard or deluxe versions, rush services, colour ranges, quality of finishing.
The media you use will have more impact on your results than anything else in your communication. With direct mail, you may want to test rented lists against each other, or against your house list; or a dimensional mailing against a standard size mailing; or maybe follow-up a mailing with a telephone or e-mail message.
You could test different publications, or different locations within a publication. Dell is constantly testing locations within publications and often has numerous advertisements running in the one publication. They keep running advertisements until they no longer pay for themselves—something they can only learn through testing.
Offers can completely change the value of your proposition and have an immediate impact upon sales. Test the wording of your offer. For example, a ‘two for one’ offer might work better than a ‘buy one get one free’ offer.
Within the creative execution, you can test single elements against each other, for example, different headlines, or you can test completely different concepts, ideas and designs.
Get on target with presentation
I was involved in writing a financial services infomercial. We tested three versions of the script using the same offer, but with different benefits. After selecting the best performing execution, which used a male presenter, we then tested the successful script using a female presenter to try to improve results. The male presenter performed better for that particular audience and offer.
Changing the format of your message allows you to reduce the production costs of your message. A standard envelope attracts different print and postage costs to an oversize envelope or 3-D mailing. A half page ad costs less to place than a full page ad and so on. Test a HTML version versus a text version in e-mail messages, or even a postcard against a letter. There are many options, which is why you test.
Timing can be the difference between profit and loss, particularly in business-to-business marketing. One rule of thumb for example in Australia is not to mail in December or January, due to the holiday season. In December, people are busy with Christmas plans and trying to get away on annual holiday. In January, people are on holiday or waiting for colleagues to return before making decisions.
Precision timing
The time of day your advertisements run on air, the day of the week you lodge your mail or distribute your catalogue into letterboxes, the day you run your press ads, or the time of day you send e-mails all impact your results. Why not test your timing to get the best results.
Major events always impact results. The various sporting World Cups, the Olympics, grand finals, festivals and public holidays also affect your results. So do big news events—which are things you cannot predict, but at least you can blame them if your results aren’t good.
The choices you give people to respond are also important. The growth of the internet has meant that websites are becoming the preferred form of response. This is because the responses can be tracked, people can respond 24 hours a day, seven days a week and the respondent does the data entry, so it saves the marketer money.
SMS is also emerging as a popular response device, particularly for competitions. The more options you offer your prospects the better your chances of a response. And remember to give your prospects the option of responding ‘No.” It’s better to know if people don’t want you to contact them, than to contact them and risk damage to your brand.
How big a test?
There are all sorts of mathematical formulas and regression analysis you can do to conduct sophisticated testing—and I won’t be discussing these here. Suffice to say that for most businesses considering testing their marketing communications, about 10 per cent of your audience is always a significant quantity to test. If you only have a small number of customers and prospects, say a few hundred or less, than sometimes it’s just as easy to do a simple split-test of concepts between each 50 per cent of your customer base.
Next time we’ll explore more rules for testing and the types of tests you can conduct.
Remember, nobody can predict the future, in terms of your response rates. If someone says they can tell you what response you will get, they are lying or trying to win your business, or both.
The only way you can know what results you are likely to get is to test. After all “why guess when you can know?”
Malcolm Auld is ceo of Malcolm Auld Direct: malcolm@madmail.com.au