If numbers can make you look smart, it should stand to reason that big numbers can make you look super smart.
But then, looks can deceive.
Don’t get me wrong. I hardly need enumerate the virtues of big data. It reveals behaviors, preferences, and predilections on a mass scale. All of which are useful.
But note the “mass” in “mass scale.” Then, recall that a strong brand is, by design, not for everyone. It is important to guard against letting big data homogenize at the expense of niche opportunities, and against letting it keep you from winning by zigging while the rest of the world zags. Unless yours is a financial institution like all others, do not let big data lull you into assuming that your clients are like all others.
While big data is useful for revealing how we behave, it is less so for predicting how, given the chance, we might behave. Big data did not predict, nor could it have, that consumers would willfully stand in line to overpay for coffee, text more than talk, or, for that matter, buy shoes—something that data revealed people had to buy in-person—online.
That is why wise marketers take care to round out big data with other, smaller, more specific data. And to test small before going big.
On the technical side, beware applying the same logic to big as to smaller data. Take, for instance, p-value, defined (loosely) as the probability that the null hypothesis is wrong. The bigger the data, the greater the danger of false positives. As Minitab’s Patrick Runkel explains, a too-large sample can appear to magnify numbers you would rightly ignore in a smaller sample. And as neurologist Steven Novella explains, “The p-value tells us the probability of the data given the null hypothesis, but what we really want to know is the probability of the hypothesis given the data. We can’t reverse the logic of p-values simply because we want to.”
Big data can offer an alluring but dangerous out for the lazy marketer. (Perish the thought that lazy marketers rank among this blog’s readers. I’m taking about, you know, those other marketers.) Embrace it. Stay current with it. Mine it for every gem. But don’t assume the thinking has been done.
It’s time for Section 215 of the Patriot Act to expire
It was once common practice never to descend into a coal mine without a canary. Canaries are more sensitive than humans to methane and carbon monoxide, which are odorless, colorless, and deadly. If the canary expired, it meant you needed to hightail it out of the mine before you expired, too.
The practice ended early in the 20th century. But now, in passive-aggressive response to Section 215 of the Patriot Act, a namesake practice known as a “warrant canary” has arisen.
Section 215 lets the FBI make any person or entity fork over on demand pretty much anything—printed matter, computer files, data, online search histories, phone records, purchases, reading habits, underwear, you-name-it. The FBI can exercise this privilege without having to show grounds or probable cause.
Moreover, if you’re issued an order under Section 215, it is illegal for you to say so.
Still, it wasn’t long before a few astute, rebellious souls noticed that Section 215 neglected to make it illegal to say that you haven’t been issued an FBI order. Suddenly statements began appearing like this one, from Apple’s 2013 Transparency report:
“Apple has never received an order under Section 215 of the USA Patriot Act. We would expect to challenge such an order if served on us.”
That statement is a warrant canary. The idea is that if the statement changes or disappears, we might reasonably infer that the warrant canary has died—that is, that an order has been issued. That is why not a few people were concerned when one year later Apple changed the statement to this:
“To date, Apple has not received any orders for bulk data.”
The qualifier “bulk” may portend a deceased canary in the form of Apple’s having received orders for specific data.
The FBI, it seems, is not a fan of loopholes. Nor does having named the loophole in question after a sweet, colorful finch appear to have appeased them. As I write, the FBI is pursuing legal action to disallow warrant canaries. Not just Apple, but the likes of Twitter, Google, others, and of course the ACLU, are opposing the FBI on this one.
I bring it up at this time for two reasons. First is that the issue has been heating up of late. Second, Section 215 of the Patriot Act is due to expire on June 1 of this year.
I hope that our elected leaders will stay their hands and let Section 215 pass into oblivion. I am no lawyer, and call me a rebel if you must, but I think that Section 215 rather flies in the face of the spirit of the Fourth Amendment:
The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no warrants shall issue, but upon probable cause, supported by oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.
Or, to put it even more succinctly: Without cause and due process, government officials cannot barge into your personal place and dig around willy-nilly just to see what they find.
Watch for this issue to escalate as June 1 draws nearer. Until then, let’s hope that the ACLU, Apple, et al can continue to—I can’t resist saying it—give Section 215 the bird.
Last week, a friend of mine, also in marketing, posted this after fighting his way to live chat on Adobe’s website:
“Adobe live chat is immensely helpful—once you get there. First you must prove your worthiness. You must navigate a maze, undergo a colonoscopy, pass a written exam on String Theory, run a gauntlet (blindfolded), sacrifice an unblemished lamb (hard to find this time of year), balance a tree stump on your left index finger, don ceremonial garb, fight off seven angry gorillas (lowland), flawlessly execute 100 consecutive jumping jacks, roll in mud, swear so as to make Quentin Tarantino blush, prove you weren’t born in Kenya, and hold your breath until you expire. Only then will Adobe provide you the chat link.”
Not exactly the sort of PR any company needs. It illustrates a paradox: as more customers demand digital services, the more you’d better be prepared to back them with live, easy-to-reach humans.
Most companies get the part about “live” and “humans.” Some, however, seem to have trouble with the “easy-to-reach” part.
No one knows better than I that ones and zeros have advantages over live bodies. Ones and zeros don’t expect a paycheck, go home at 5, take weekends off, expect overtime, unionize, gossip, call in sick, take vacations, request family leave, whine, stretch 15-minute breaks to 45 minutes, take personal calls when they should be working, crack their knuckles when their neighbor is trying to concentrate, waste time, or show up for work hung over.
So at first blush it might appear cost-efficient to hire fewer live bodies and drive customers to the likes of FAQs and forums.
In reality, if you make it too hard to reach a live body, you will create frustration. If you create enough frustration, you will lose customers. You needn’t lose too many customers before you find that hiring a few extra bodies would have cost you less.
While all companies are well advised to clear away obstacles between customers and live help, I venture to say that it is even more important for financial institutions, where questions and problems tend to be urgent and time-sensitive.
The best practice is to offer live help, and offer it early. When a customer clicks around the same feature one too many times or pauses for too long, provide a popup that says something like, “Need help? Click here to chat with a live representative.”
DIY resources are all well and good, but customers shouldn’t have to navigate an obstacle course before being given a link to live assistance. It is best to allay frustration before it has a chance to arise, much less escalate.
From the moment the U.S. Constitution gave Congress the power to establish post offices, marketers went to work learning how to make direct mail pay.
Over time, direct mail marketers began noting techniques that seemed to work most often and on a consistent basis. These became must-know guidelines for anyone in the business. It turns out that many of those guidelines apply just as much in email as they do in direct mail. Here are a few of my favorites. (See last week’s post for direct mail practices not to use.
The envelope, please. In direct mail, an envelope has one job: to get opened. A headline on an envelope can work wonders to that end. The email equivalent is the subject line. But be careful: many phrases that work in direct mail will likely be intercepted by a spam filter if you try them with email. For instance, “FREE” works on envelopes, but in subject lines it signals spam.
Let’s get personal. In mail and email alike, letters have power. The most successful ones are laid out like letters (albeit with liberties taken here and there), display a signature, and use a personal, “me-to-you” tone—not “our company to our customers.”
P.S. Though a stylebook may advise against using “P.S.” on a business letter, direct marketers have demonstrated that that’s where most people start reading. Use a P.S. to tease with information that will pull readers into the body of your email. Provided, that is, your P.S. lands above the fold. Short of that, try a Johnson Box. In fact, it’s not a bad idea to do both.
Headlines. Stylebooks advise against headlines, too. Use them anyway. A good headline increases readership.
Subheads. Short subheads in bold type between groups of paragraphs break things up and help pull readers along. (So do in-line bolds, like the ones I’m using in this post.)
Keep paragraphs short. Big blocks of type are daunting and hard to read.
Include a call to action. Any sales training course will rightly tell you that you must urge customers to take a desired action. Here, email has the advantage. While direct mail relies on coupons, postpaid cards, URLs, and toll-free numbers, email needs only a link, a few encouraging words, and “click here.”
While you’re at it, include lots of easily-noticed calls to action. Ever studied a direct mail letter? Toll-free numbers and addresses show up on the top and/or bottom of every page as well as in body copy. The idea is to make it easy to order at any time. Likewise, email should display easy-to-spot links at the top, in body copy, in side columns, right above the “fold,” and at the bottom.
Incentive offer. Most prospects who do not take immediate action will ultimately not take action at all. The cure? A reward—a freebie—for responding within a limited time. Warning: in every company, there is always someone who objects to incentive offers as “unprofessional.” Nonsense. Properly worded, there is nothing unseemly about thanking customers for prompt action by sending them a gift. But “there is nothing unseemly” understates. It is absolutely vital is more like it. The right incentive offer will increase responses by two, three, four, or more times.
We who do email marketing are heirs of an earlier profession known as “direct mail.” It is still practiced, profitably I might add, by holdouts. Perhaps you know some. Or perhaps you read about direct mail in a marketing history book.
Though the ubiquity of email is relatively new, its successful use grows largely out of techniques that direct mail marketers have been developing ever since Section 8 of the United States Constitution gave Congress the power to establish post offices and made Benjamin Franklin the nation’s first postmaster.
The national postal service was up and running by the early 19th Century. Marketers lost no time figuring out how to profit from it. Mail let them sell over vast distances without having to establish stores in every market. They could communicate with large audiences. When audiences responded, their actions were measurable.
That should sound familiar. It is essentially what we do today, albeit faster, with email.
Early direct mail merchants fast sorted techniques that generated results from those that fell flat. Today, it turns out that many of the resultant do’s and don’ts are as apt in zeros and ones as they were in envelopes.
Not that we should emulate everything early direct mail marketers did.
Here are a few things that you and I had better not do. I’ll save direct mail’s more positive legacy for Part 2.
Though plenty of early direct marketers had scruples, a fair share did not. One early mail-order marketer offered “a genuine, U.S. Government-approved engraving of Abraham Lincoln” for the low, low price of just five cents. For their nickel, dupes received a penny. Another scammer offered a sure-fire roach killer. What arrived in the mail were two blocks of wood with instructions to place the roach on Block A and smash it with Block B. Yet another offered unmarried women (in the days that “spinsterhood” was deemed a cause for concern) a guaranteed method for making “a good impression.” For their money, hopeful brides-to-be received printed instructions to sit on a pan of bread dough.
In time the U.S. Postal Service cracked down.
But since you’re reading this blog, I assume you’re a legit marketer, above lawbreaking. So let’s look at techniques that are perfectly legal by mail, but either illegal or at least inadvisable when it comes to email.
It’s legal for direct marketers to mass-mail “blind envelopes,” that is, with no return address. Since most people have difficulty discarding any envelope unopened, imagine the difficulty they have discarding an unidentifiable one. It’s also legal to make an envelope look like official mail from a government agency or other authority, provided you do not explicitly state that it is. I happen to think that misleading envelopes are underhanded and inappropriate. But they are legal and, unfortunately, effective.
Hiding or falsifying your identity in an email blast can be effective, too. In the U.S., however, it’s not just underhanded: it violates the CAN SPAM Act of 2003. That’s why you don’t see informed, legitimate, domestic companies try it, and you don’t see uninformed, legitimate, domestic companies try it more than once.
Direct mail has always been and still is an effective prospecting tool. A skilled analyst can overlay and process mailing lists to increase the odds of success of a cold mailing. There is nothing illegal, nor, I might add, underhanded about it. Showing up uninvited in a mailbox is not intrusive, whereas showing up uninvited on a smart phone, tablet, or computer is.
There’s nothing illegal about cold prospecting with an email blast, but it’s not advisable. For one thing, as I said, it’s intrusive. For another, email marketing tends not to be a terribly effective prospecting tool. Email works better on a permission basis, that is, when customers authorize you to ping them. That aside, cold email blasts can result in your being intercepted by spam filters, consigned to junk files, and, worst of all, blacklisted by email servers that will thenceforth block all of your emails, permission-based or not.
It’s generally better to use email for customer retention, growth, and cross-selling. For finding new customers, look to the likes of banners, referrals, social media ads, and pop-ups to secure permission to email. And, see above: Direct mail has always been and still is an effective prospecting tool. Many an online ball gets rolling by use of snail mail.
A final don’t. The only limit to how often you should pepper a customer by mail is determined by diminishing returns. You can send mail as often as it continues to pay. When it comes to email, people—and many servers—have a threshold. Ping too often, and you’ll be consigned to spam or junk. That is why services like Constant Contact won’t let you send blast after blast. And they won’t let you blast to the non-opted-in or to prospect lists, period. They like remaining in business.
Next week: The Good.