Price and Prejudice: Using Logic to Guide Offer Tests
Besides sophisticated personal-ization, subtle price and offer tests are two of the most difficult elements to track in the Who's Mailing What! Archive. Frankly, we just don't see that many offer tests, other than those for obvious elements like premiums. Hopefully that's related to collecting mail from roughly the same group of correspondents every month, and not a reflection of how much offer testing actually occurs in direct mail. But it might be the latter, especially if you're talking to Grant Johnson, president of direct marketing agency Johnson Direct, in Brookfield, Wisc. "The facts: We all know that lists/media and offers have the greatest influence on resultscombined they account for 65 percent to 70 percent of an effort's power," says Johnson. "The problem: ... Most marketers spend too little [time] on lists and even less [effort] on the testing of offers. They plunge into expensive, new creative, convinced that a fresh look or tone will increase their ROI. They've been proven wrong a thousand times. There are no shortcuts to direct marketing results ... thoughtful, solid and constant offer testing is the only way to get there," he says.
And, points out Steve McKenzie, executive vice president, Troposphere Directed Media, an integrated marketing consultancy in Chicago, "You can only tweak a package so much with creative ideas."
That said, McKenzie also cautions against over-testing offers. All sound testing plans are centered on the search for truth, meaning you need to develop a sequence for testing offers based on which tests have the most potential to help you achieve your acquisition, retention and/or profitability goals.
"Offer testing should be a regular part of any mail plan," says Gary Hennerberg, founder of direct marketing consultancy Hennerberg Group Inc. in Colleyville, Texas. "But offer testing can be challenging, requiring constant programming of order entry systems, sourcing new items, or reconfiguring an existing offer. By all means, if a prospecting control starts to fatigue, testing new offers can be one way to pump up response."
The time to conduct some solid price testing is before your offer reaches the control stage, advises McKenzie. It's far easier to start with higher price points and then roll them back in the testing phase; once you hit rollout, more people seeand takeyour lower-priced offers, which makes it hard to establish a higher price later on. And, he explains, trying to renew these segments at higher price points can induce sticker shock in customers.
Unless it's a real hardship for a marketer to stage continual offer testing, Johnson urges companies to be proactive. "The leading direct marketers know that testing is the foundation of the profession and should be practiced continuously," says Johnson. "If you wait to test [when] your control begins to tire, turn out the lights and say goodnight! You've wasted valuable timea chance to increase responserates, discover and stay current about what works best, and gather vital information about your customers or donors and you may well find yourself having to start all over again tomorrow."
With so much at stake, what types of offers have these experts seen work well, and thus are worth a test?
Hennerberg finds that, "Free items that are logical extensions of the core product being sold are usually strongest." One such example is the free pound of gourmet coffee promised to prospects who buy a fruitcake, a control offer Hennerberg developed for Collin Street Bakery. Another example of the related premiumand one that has been leveraged to the extremeis the free coffeemaker Gevalia Kaffe gives to prospects who agree to join its mail-order coffee club on a trial basis. Over the decades (yes, decades!), Gevalia Kaffe has tested different styles of coffeemakers to keep up with kitchen decor trends, but the link between the product being sold and the premium has remained tight.
Hennerberg also has had success repositioning old offers in a new way that gives these promotions strength againa reminder to revisit offers that worked well in the past. Or, you could turn a fresh eye to test offers that fell just shy of beating the control and might be better suited to current market conditions.
Johnson encourages marketers to think beyond the standard offer structures to improve response and profits. For example, instead of offering a discount on tickets to a music festival, Johnson's firm came up with the idea of selling seat licensesor the right to purchase ticketsthat provided the customer with exclusive event privileges. This approach contributed to a 21 percent lift in revenue for the event and helped the promoter exceed revenue budgeted.
Both Hennerberg and Johnson noted that information-based premiums tend to be low in cost, but can be high in perceived value if you properly merchandise this component in the offer and sell prospects on its usefulness.
The Who's Mailing What! Archive has been tracking all kinds of premium testing in the past year or two, suggesting that merchandise continues to have star power in an offer. When including a premium in an offer, McKenzie cautions marketers to be sure the premium is not the sole reason for response, or these prospects won't convert or renew well down the roadat least not without continued investment in premiums to coax future response.
What about straight price tests? Circulation expert Don Nicholas once told me that high-volume publishers like Nutrition Action Health Letter could increase profits with as small a drop (or a lift) as a penny in price.
Lois Boyle, president of catalog consultancy J. Schmid & Associates in Mission, Kan., pointed out in an article printed in the September 2004 issue of Target Marketing ("Don't Leave Money on the Table") that consumers generally do not recognize the difference in cost between items with price points ending in .50 and .99; order upon order, that 49 cent difference could turn into quite a boost in profits.
"Probably the lowest cost offers are the most overlooked. Sometimes customers want nothing more than 'upgrades,' such as an inbound phone line staffed with the most experienced customer service people or a Web site that streamlines their ability to order quickly and easily," states Hennerberg.
Another aspect of offer testing to consider is the element of choice. Whereas the direct mail philosophy typically has been to reduce prospects' choices, that might not be such a strong rule any longer. Today's consumers want offers tailored to their desires, says McKenzie, so offering a choice of premiums or a way to customize a premium could be a good testing idea.
Johnson finds that not enough marketers test the power of a well-positioned guarantee. "Think about it. When you promise, 'Your satisfaction is 100% guaranteed, or your money back ... ' or 'If you're not happy with (PRODUCT), simply send it back. The (FREEBIE) is yours to keep FREE!' you've made great offers, and you're not out a red cent," explains Johnson. "It's been proven, the stronger the guarantee and the more frequently it appears, the less it needs to be used."
He also likes to test the doubling up of offers, where the marketer makes one offer early in a mailing and then a second, better offer later in the textas a reward to the reader for continuing to read the message. And unique price tests can give marketers the chance to qualify more prospects for the product; for example, Johnson explains, prospects can pay $50 to start on a $350 product, with their credit card being billed $50 each month for six months.
Can You Pre-test Online?
McKenzie finds that direct mail is so expensive compared to the Internet, it's almost quaint to think that marketers can do their most efficient testing via the mailstream. Of course, he explains, if your market spends little to no time online, you won't gain much useful offer insight online.
American Express, one of McKenzie's clients, has been able to conduct some solid testing of direct mail offers using e-mail to its customer file. He notes, however, that American Express has a significant percentage of e-mail addresses for its customers, as well as permission to contact them this way. Without a large enough quantity, it would be hard to get a statistically accurate representation of your housefile's behavior to offer tests. And, he warns, marketers who go this route have to be careful not to abuse customers' e-mail addresses with constant offer tests. The long-term value of maintaining permission to communicate via e-mail with customers is stronger than the short-term value of using e-mail for offer tests.
Of course, you also need to be thoughtful about what kinds of offers you test in the online environment to be applied to direct mail tests. "You always can pre-test online, but be aware that when dealing with different media lead sources, the same offers may not always apply," says Johnson. "The best offer for online is still free shipping with the order. That may or may not work as well with direct mail and catalog buyers, and certainly, as previously stated, offers must be tested regularly. If you find one that is working exceptionally well, then test the same in other media and find out if an across-the-board offer helps you achieve the ROI you're after."