How many characters left?

What’s wrong?

1. It only gives me 255 characters. This is probably because they have chosen to store the text in a fixed-length database field for rapid retrieval. That’s what happens when the usability people and the marketing people get overrruled by some engineer who thinks that this minor efficiency improvement is sufficient reason to cripple the interface.
2. It doesn’t tell me anywhere how many characters I’m allowed. It wasn’t until I submitted my original version that it rejected it and came back and told me there was a 255 character limit.
3. It has no running count of characters used. This has become a standard feature everywhere else. We’re used to it on Twitter and most places that have low character limitations.

So what? Well, it took me three tries to get my feedback accepted. And by the time I did, the glowing testimonial I had for the merchant was gone. No room left.

All the merchant got out of this was my comment that I think they should make their free shipping offer appear more prominently on the page.

What the merchant did not get was my comment that when I factored in free shipping, their price was significantly better than the competitor’s price.

Okay, I just placed the order, so I can’t comment on speed of delivery and all that, but I would say that the shopping experience on US-Mattress.com was close to ideal. It’s easy to navigate, there are no real surprises (except why did they offer to let me upgrade to “standard front door” delivery for $49? What is the delivery I’m getting for free?).

Anyway, the bad part is that I agreed to do the Bizrate survey because I like to reward e-commerce merchants who do it “right”. I arrived at the survey with a good feeling, wanting to leave a great testimonial. But the frustrations of using Bizrate’s system left me feeling, well… frustrated. Of course, I don’t hold US-Mattress responsible, at least not consciously, but that’s the thing about usability problems — often they operate on a sub-conscious level. The good feeling I had upon completing the purchase is now forever associated with the stupid Bizrate survey.

And then, there’s Bizrate’s enticements to get you to take the survey, promising all sorts of free stuff. Obviously, everyone who spends a lot of time on the net knows by now that these are not “rewards”, but affiliate offers from which Bizrate makes additional income, but that’s a whole other story.

I just loved the service on keyword airlines!

Source: Skymall catalog on a recent Delta flight.

I’m sure this testimonial is 100% legitimate. I mean, your typical loving family peppers their speech with “keyword” when they can’t think of the exact word, right? And a company would never reuse a testimonial would they? Of course not!

I was really bored, so I picked up the in-flight gadget catalog and they had this ad for the coolest keyword ever! The family in the testimonial for the keyword made it sound so good, I just had to have one!

The Audience Problem

Like I say, not sure how much my thoughts are worth, since clearly they’re aiming at another audience. As in: I don’t do chat, IM, text messaging or any of that. I have long since trained my friends that I don’t often answer emails the same day I receive them (and long before I heard of Tim Ferris). The only immediate response thing I do is phone and skype and I only give my skype address out to family and a few friends and try to limit that. So it’s a bit hard for me to see how I would participate in Aardvark.

Conceptual level. The Big Idea level….

Here’s where they fail to make the sale to me and once they fail to make this sale, it’s an uphill battle for them to build trust through the rest of the process. The thing that is difficult for me to get around is that in my view there are personal and impersonal channels of communication.

• Personal: email, chat, skype, Facebook personal messaging, Twitter direct messagings. These are all messages from someone to me specifically and nobody else.
• Impersonal: Twitter posts, forums, Facebook wall, etc. These are messages that go from someone to the wide wide world. They’re not to me personally and uniquely.

I try to keep my personal channels free from impersonal messages. I have spamcatcher email addresses I use for things that blur the line, such as newsletters, mailing lists, signing up for accounts with BestBuy, Amazon and such. It strikes me that Aardvark is trying to use a personal channel (chat, email) to deliver an impersonal message. Yes, it is personalized — I only get messages that are supposed to be appropriate to me — but not personal, that is only to me. So that’s an adoption hurdle for me just as a concept.

The Registration Problem

They could overcome the personal/impersonal problem by using the registration process to allay fears and make the sale, but in my opinion, they do the opposite. Aardvark actually asks for quite a bit of information just to get started. I’m always skeptical of that and if I’m going to give away a lot of personal information about where I live and what I like, information that marketers will kill for (or worse yet, pay for). To give away all that information, it needs to meet one of two conditions, and preferably both:

• I need it. I may be a little uneasy about a site, but they have something that I absolutely need. I can’t do without it or I don’t want to do without it. They’re asking for personal details, but they’re offering something of great value.
• I trust them. There are a few sites that I trust implicitly with my information. I don’t give Amazon more than I have to, and they have only my spam catcher email address, but over the years they’ve built up great trust by not abusing my information. Often not-for-profits ask me to trust them because they have a great mission and are inherently good. Just like the government, if you catch my meaning. And if you don’t, that is to say that the government has been a poor steward of my privacy lately.

Typically, when I sign up for a new service that I don’t necessarily trust, I start by giving a spam catcher address and often a fake name (and almost always a fake birth date). If they want personally identifiable information,they need to build my trust either before, during or after the registration process.

I actually went all the way through the Aardvark registration process because I was asked by a friend to test it. I found it much too intrusive for a site that I had never heard of and knew little about. They have detailed information on how it works in theory, but nothing at all on what happens with my data, who can see it, and what control I would have over contact from people I know and don’t know.

An example

And then there are parts that I didn’t do anyway, even if invited by my friend…. Example: in general, I block all Facebook apps. I find all those snowballs fights, mafia, pirate stuff absurd and just a distraction to keeping in touch with family and friends. And I don’t collect Facebook friends. I try to keep it a personal channel as much as possible. If you we don’t have personal history together, you’re not on my Facebook list. When Aardvark offers to connect to Facebook, it’s still not clear to me exactly what’s going to happen, how it’s going to show up on Facebook, what my friends will see, and what exactly my benefit is. Ideally, exactly next to the Facebook connect button there should be a "what’s this?" or "how this works" link to a video that shows how it shows up in Facebook, what my friends will see, what benefits it offers and what hassles, if any, it imposes on my life. For me Facebook is a semi-personal channel and and I don’t want to annoy my friends and family that I keep in touch with via Facebook. Before I connect other data, I need to know that it won’t annoy my friends or affect my reputation.

A Broken Interface Erodes Trust

If I start setup, I can’t get to the welcome/home page any more or at least I couldn’t figure out how. It always brings me to the last spot I was in during setup like a pitbull that won’t let go. Clicking on the Aardvark at the top should always take me to the home page (a web interface standard that must not be broken), but it took me to the Facebook Connect page. So I’m not on Facebook (though I am) and it took me to the Add Categories page. But do I want to add categories? Again, are my categories and demographic info being shared with marketers? This type of behavior once again erodes trust. It makes the user feel trapped.

A recommendation

Think about every possible hesitation and catch me exactly at my hesitation point, like the suggestion to have an explanation about effects on privacy and such right next to the Facebook Connect button. I know of marketers who say they get much higher conversions when they have a popup link to their privacy policy right on the registration or order form, for example. That would help a lot.

Aardvark needs to think a bit more about the registration process if they want easy adoption beyond social networking true believers: what trust and social proof barriers might people perceive, figure out what the choke points are by keeping track of exactly where people abandon the process, figure out why, and take steps to fix it.

Online, trust is everything. In person, we have the idea that if something goes truly bad, we can go down to the business or local animal shelter or whatever and picket, protest, call the police, walk in with a lawyer. It doesn’t mean I trust those businesses. They often ask for a phone number at transaction time and I simply say no. But I do have the assurance that I can come down and find these people.

Trust is harder to build online and must be cultivated carefully and persistently at every possible occasion. There is no such thing as paying to much attention to building trust, and Aardvark needs to pay more attention.

I was looking to buy a copy of Microsoft Office Home and Student Edition. I looked at my favorite places: Tiger Direct, Amazon, BestBuy and decided to have a look at EBay, though I’m always a little leery of EBay. It had what looked like a legitimate copy from a legitimate seller. It was a UK seller, with almost 500 sales and over a 98% positive rating. The description said the item was a retail copy, not OEM, which was important to me because the retail version can be legally installed on three machines, but the OEM version is only for one machine. Everything looked good, but I was still skeptical. Finally, one thing tipped the balance: it was guaranteed by Paypal Buyer Protection. I’m a fan of Paypal, so it seemed like a good bet.

A day later, I got notification that my software had shipped. The email was funny — several characters that wouldn’t render. Uh oh. So I clicked the tracking info and it was shipping from Shanghai! Okay, could be a drop shipper, but it wasn’t looking good. I logged into my EBay account and all trace of the purchase had been wiped clean. The seller account was gone. All information about my purchase was gone. Double uh oh.

So I call EBay. They can’t help because the seller has been kicked out and they won’t arbitrate at this point. They tell me I will have to wait until the item arrives and then take it up with Paypal if there’s a problem.

The item arrives and it’s a recordable DVD with a handwritten product code on it. I call Paypal and they will not cover it until I return it. I’m livid. This person is an obvious criminal without a let to stand on. If I had put it on my credit card, I would cancel my charge and end of story. With Paypal, not only do I need to return the item, I need to return it with tracking. As it turns out, the cheapest way of tracking the item will only track it out of the US, but not to the seller’s door. To do that, I would need to spend more than the cost of a the software at Amazon.

Alas, it’s a sunk costs question. No way to recoup my money. So I went back to Amazon and bought it there. Brand new, retail version, three licesnes that validated with Microsoft. Only about $5 more than the Chinese ripoff artist.

Lessons learned:

• Never buy software on EBay. There’s just no way to tell unless it’s a merchant you know.
• Avoid Paypal for online purchases and use a real credit card with reasonable dispute policies
• Scammers are making brands more powerful and making it hard for small merchants to make a living on the web.

This is the third part of the series on Weapons of Influence, based on Robert Cialdini’s book Influence, Part 1 discussed the principle of reciprocation; Part 2 covered the principle of commitment and consistency.

Social Proof

Birds of a feather, flock together. Or so we’ve been told. In other words, we like to be around people like us and we like to be like them. We also look to others for cues as to how to act. This explains a famous incident in New York where 38 witnesses heard a woman’s scream for help and did nothing. The failure of others to respond is a clue to us that the situation is not serious, we don’t need to respond either. Unfortunately, that cue, often reliable, cost the woman her life in this case. It turns out that you’re much better off if one person sees you being attacked or sees smoke coming from under a door than if a crowd of people do. If you do find yourself in a bad situation and there’s a crowd, use the word "help", look someone in the eye, point at that person and say "You, sir, in the red shirt, please help." General pleas to a crowd tend to go unanswered until one person responds, then the social proof works in the other direction and others will jump in to help.

The effects of social proof go beyond what I would have guessed. For example, in the months after a highly publicized suicide, the rate of airplane and automobile fatalities goes up significantly. This has been observed over long periods in large numbers and with numerous controls. Furthermore, if it’s a murder-suicide, it is more likely for multi-passenger airplane and multiple vehicle automobile fatalities to occur. If it’s a simple suicide, it correlated with single-victim crashes. After adding in numerous controls to the data, researchers were forced to conclude that these increased crashes were secret suicides. In a similar vein, after a heavyweight boxing championship bout, murder rates briefly rise around the country.

Back to how this is typically employed specifically to influence you, Cialdini looks at laugh tracks. Everyone polled says he or she hates canned laughter soundtracks on television shows. And yet, research shows that even though the canned laughter is obviously fake and we say we hate it, we find shows funnier if they include a laugh track because our subconscious mind can’t escape the fact that "others" are laughing so it must be funny. Similarly testimonials, even when it is obviously not a "natural" unsolicited testimonial, influences our decision to buy (and someday I have to tell the story of the Hansen’s soda lady trying to elicit a testimonial from me. I didn’t end up famous).

There are some crazy variations on the influence of social proof. For example, in a study by Kimberlee Weaver of Virginia Polytechnic University, the researchers created two conditions: one where people heard several people express a given opinion once each and one where one person expressed an opinion several times. It turns out, that in both cases, respondents judged opinions to be popular based on the number of times they had heard the opinion, but did not adjust for the fact that in some cases it was actually an opinion expressed repeatedly by one person [1]. So strangely, we sometimes perceive social proof when what we’re really seeing is one persistent loudmouth. That’s a good lesson if you really want to get something done in your community, but it’s not so good if someone just has a big enough budget to broadcast that message at you 12 times per day.

How to Say No. First, remind yourself that the testimonial you’re seeing is quite possibly faked. Large numbers of review sites on the web are laden with fake reviews. If you don’t have a reason to trust the testimonial, don’t. Second, don’t assume that if a lot of people are doing something, they must have information that you don’t (that’s not a person in a diabetic coma who needs help, but a drunk sleeping it off in the gutter). I must say, I don’t think Cialdini has read Kierkegaard, because he grappled with this question over 150 years ago with his famous ruminations on Abraham preparing to sacrifice Isaac, knowing that this was a solitary and unjustifiable act. Kierkegaard believed that only "the crowd" could have executed Christ and that if each person had had to face Christ alone, one at a time, he would never have been crucified. Kierkegaard argued that truth is subjectivity, not objectivity and that when you side with the crowd, you cannot know if your decisions are moral
or true. It doesn’t make them immoral, it’s just that you don’t know. So Kierkegaard’s philosophy implies a simple question: what would I do if I had to make this decision with nobody else around? What would I do if everyone else were doing the opposite of what they are doing now? In other words, if everyone wanted to honor Christ, would I still vote to crucify? If everyone was rushing around in a panic because there was smoke coming from under a door, would I calmly walk past? Are my actions conditioned by the crowd, or by my sense of what I should do in this situation? I think Cialdini finds it impractical to pose such questions every minute of the day and that may be why he doesn’t invoke Kierkegaard. I cite Kierkegaard because I like to pretend I’ve actually read Kierkegaard instead of just heard about him on Jeopardy. Don’t tell Alex Trebeck.

[1]"Everyone Agrees", but Melinda Warner, Scientific American Mind, Aug/Sept 2007 (vol. 18, no. 4), p. 13.

This is the second part of the series on Weapons of Influence, based on Robert Cialdini’s book Influence, Part 1 discussed the principle of reciprocation.

Commitment and Consistency

We like to be consistent and honor our commitments. As with reciprocity, under normal social circumstances, these are good traits, but they can be used against us. The famous experiment in this vein was the one where a research accomplice goes to the beach, lays down a blanket and puts out some personal items, including a radio. The accomplice then goes away. A few minutes later, another accomplice comes up and "steals" the radio. The experiment varies between two conditions. In one condition, the original accomplice gets up and leaves without saying anything. In the other condition, that orginal person specifically asks someone to watch his or her stuff. In the first case, four in twenty times the second accomplice could "steal" the radio without challenge. In the second condition, the "thief" was stopped and challenged nineteen out of twenty times. So the challenge rate went from 20% to 95% (p. 59). In other words, people had an overwhelming desire to be consistent with
their prior commitment.

These techniques can be remarkably subtle. For example, when a telemarketer calls it makes a huge difference whether that person says "How are you feeling tonight?" and gets an answer or says "I hope you’re feeling well tonight". The difference is that in the first case, the target has committed publicly to having a good evening (because the response is typically Fine, thanks"). Having publicly committed to doing "fine", it is very hard for the target to shirk on giving money to the earthquake victims in wherever who are so unfine and so in need of help. In the case where the caller simply says "I hope you’re feeling well this evening" no such commitment was extracted and the response rate was less than half (15 vs 33 percent) what it was when the caller asks a question.

Or how about this one. Toy companies advertise items in the runup to Christmas that they have no intention of stocking in sufficient numbers to meet demand. The unwitting parent commits to the present for the pleading child. Since the gift isn’t available, dad buys something of equivalent value for Christmas. But two months later, well there’s that item miraculously on sale. Dad goes and buys it because he feels a commitment to his kid. The toy companies know this and use this technique to prop up sales in January and February.

Companies use essay contests to make you feel good about them. Something as simple as copying out a message in your own handwriting can make you want to follow through on all the nice things you’ve said about that company. Public utilities have gotten people to save lots of energy simply by getting people to commit to saving energy.

It turns out that internalizing the commitment is key. When utilities hold a contest and say that those who save a certain amount of energy will be recognized, people do cut down on energy usage. But when they then call to renege and say the contest is cancelled, it turns out that energy usage falls even further. It seems that this is because people are actually less motivated when they feel they’re doing it for external reasons, and more motivated when they feel they are doing it for themselves. Being the kind of person who likes to be energy efficient is more powerful than being the kind of person who will reduce electricity usage in order to save $5.

There’s one further usage of the commitment and consistency principle that is worth noting. I’ve been taken in by subtler forms of this on many occasions. It’s called the lowball. Essentially, your sleazy used car salesman offers you a price that he knows he can’t really honor. You agree. When it comes time to sign the final papers, the accounting department finds an arithematic "error" or the sales manager notes an "error" in the trade-in value and the salesman sheepishly fesses up to his "mistake". But here’s the thing, you’ve committed to buying the car, and at this point it’s psychologically hard to turn that around, and you buy it anyway, rather than going down the street to where they have it for real at the initial price the salesman offered. The subtler version that takes me in is deciding to buy something that’s on sale. But then I drag my feet and miss the sale. Two months later, I miraculously own it anyway.

How to Say No. When we get caught in these situations, we typically know in our gut that we’ve been had. If the spidey sense starts tingling, ask yourself a simple question: "Knowing what I know now, would I make the same decision as if I had not committed myself?" So in other words, ask yourself, "If I had not shaken the hand of the sleazy used car salesman on the deal, would I buy the car at this price?" "If I had never seen that Marmot Precip jacket on sale, would I still be buying it today at full price?" The key here is to focus in on how you feel in the microsecond after you pose the question, before the rationalizations kick in. That is when you’re most honest with yourself or, as Cialdini says (p. 110): "Accumulating psychological evidence indicates that we experience our feelings toward something a split second before we can intellectualize it."

So here’s a rundown on the most potent weapons of influence.

Reciprocation

Someone offers me a gift and I want to offer a gift back. Someone does me a favor and I want to return it. That’s a good thing. When my neighbor Bruce offers to take me to the Fresno airport, two hours away, that makes me feel not just a bit obligated, but also like I must be a a pretty good person because why else would he offer? So that’s an obvious gain for me. But when he needs to go to Fresno and pick up his car and I help him out, I not only feel I’ve repaid my obligation, but I again feel like I must be a prett good person, because I’m doing something nice for a friend. I don’t think this is some atypical egomania, but rather pretty normal, though 99% of the time subconscious. We like to do for others and have others do for us and that makes our lives richer. So far so good.

But savvy salesmen and fund raisers will use that against us. The salesman offers us a free gift and we feel like we need to reciprocate by helping him out. So we buy something. You would be surprised how effective this is. When the Disabled American Veterans asked for money, they got an 18% response rate. When they sent out the same appeal with the "free gift" of address labels, they got a 35% response rate (p. 31). The Hare Krishna’s used giving a flower away to similar effect.

Even more subtle, someone asks something onerous of us: "Would you commit to spending one night per week at the homeless shelter for two years?" We say no, and the person then makes a concession: "Could you help out for three hours on Wednesday night?" Studies show that the concession tends to make us feel obligated and the compliance rate soars compared to a direct request to help out for three hours on Wednesday night. Another version is where you go in to buy a camera lens, pool table or an appliance and say you’re looking for a basic version. The salesman starts you off with the most expensive, premium version just to show you what real quality is. Then he backs off to the mid-priced version to show you something more in line with what you want. In this scenario, which invokes both the reciprocal concession and the contrast rule ($500 now seems cheap compared to the $5000 lens). This has commonly been shown to dramatically increase sales (p. 47).

So the truly savvy ask for something they never really expect to get anyway, and then "settle" for what is really their true goal in the first place.

How to Say No. Recognize these practices for what they are — neither gracious nor nefarious, but simple sales techniques employed for the purpose of making a profit. Consider the "free gift" like any other form of advertising. Just because a company spent 26 million dollars to show you an ad during the Super Bowl doesn’t obligate you to anything. Nor does a free home inspection or a free trial of hand lotion. It’s just business. Ask yourself, what would I do if I had not received the free home inspection from them? You would call around and get prices around town, and so you still should.

You’ve been receiving killer deals on great SmartWool gear. Now it’s time to return the favor. We need you, our pros, to share the wool with the world. The submitters of the TEN BEST Testiomonials (with photos) will each win a pair of Spring Gloves!

I recently wrote about how Dan Ariely says that you must not mix social norms and market norms. Within social norms, reciprocation is important, but you have to be careful not to turn it into a transaction. Once you’ve said “If you do this for me, I’ll do this for you,” you’ve essentially entered into a market exchange. Now that you’ve entered into an exchange, the question you naturally ask, is “Is this a good deal for me?” My first reaction to this come-on was “No.”

Why? The Smartwool appeal starts out by explicitly trying not just to obligate me, but by literally saying: “We did this for you, and now you have to do do this for us and this is how much we’ll pay you.” So they set it up as an exchange. But that’s a deal made after the fact — the original deal was they give pros great deals and pros who love the stuff naturally tell their clients that they are the most comfortable articles of outdoor clothing they’ve every worn (which they are). Then they get to my “pay” and when I saw the offer from Smartwool, my first thought was “All I get for giving one of the ten best Testiomonials (with photos!) is a pair of wool gloves?” I mean, the best writing skills out of thousands of people is worth a $30 pair of gloves?

Turns out the gloves sell for $70, but it was too late by the time I read that. For not much more money, they could have offered any one product except the Banff jacket and their worst case payout would have been a $130 product. I personally would have taken a $70 or $90 shirt. More importantly, though, it was just wrong-headed to approach it this way.

What should they have done? I would propose something like this:

Do you love Smartwool products? We need our Smartwool pros to go to our website and leave testimonials and tell the world how great Smartwool products are. Please take a minute and go to our website and leave a testimonial.

Does that work better? It could be hipper or funnier, but I think this is an improvement because if I break it down, here’s how I read it:

• “love Smartwool” — you don’t want anyone else leaving testimonials anyway, right?
• “We need our Smartwool pros” — i.e. we have a relationship, we’ve done something for you; invoke the reciprocity principle, but without turning it into a commercial exchange.
• “Please” — remember, this is a social norm, not a market norm, so we ask for a favor, we don’t offer a bribe.
• “take a minute” — this will be easy. We’re not asking for much considering that your one of our pros.
• “go to our website and leave a testimonial” — tell people what you want them to do. It’s the best way to get them to do it!

Maybe I flatter myself to think I can do better. They have, after all, gotten a lot of testimonials, but they didn’t get mine. Assuming I’m right, it saddens me to see an amazing company like Smartwool shoot themselves in the foot like that. I give testimonials all the time to Amazon Marketplace sellers simply for the asking, and I don’t love those products. I love Smartwool products and would happily give them a testimonial, but I couldn’t help but think “Is that all they’re offering?”

And by the way, go buy some Smartwool stuff. If you do outdoor sports, I’m confident it will be your favorite shirt, socks or whatever. It really is that good and I’m happy to say so here without any reward whatsoever.

AOL or Apple customer? Click for full size

[caption id="attachment_177" align="alignleft" width="150" caption="AOL or Apple Customer - fill in the right blank. Click image to view full size"][/caption]

Especially with the AOL triangle thing looking like an arrow, to me this looks like one blank for AOL customers and one for Apple customers. It’s even worse when your login fails, because then it draws an arrow between the AOL logo and the blank, which just emphasizes the connection between AOL and the bottom blank. One has to wonder, has Apple tested the usability on this at all? I thought it was just me, but I showed it to Theresa and she had the same reaction as me.

By the way, one blank is for username and the other for password. Of course, this is a pretty standard interface, but as more sites (like my banks and credit cards) now have two or three-step logins, so you enter your username, then you get a verification image, then you enter your password. I assumed it was like that here.

I had a friend who worked at Disney World in the late 1970s at the complaint desk. Do you know what rule of thumb they used was for how long she should spend responding to each complaining customer? Answer: as much time as it took until the customer was happy. “But what about the people lining up behind them?” she asked. Answer: they’ll just be that much madder when they get to you and they’ll take even more time, but don’t send anyone away angry. If they’re lining up, it’s our job to get more people out here. It’s your job to make sure that the person who is upset, goes away happy.

So when I ask for support, it’s up to me to decided when my issue is closed. Strangely, I used to find it odd, even slightly annoying, that the tech reps at JaguarPC, my web host, never closed tickets, even when I said, okay, I’m fine, close the ticket. It’s always up to me to close the ticket and I’m not even sure the reps have the authority to close your ticket. As near as I can tell, only you do. Foolishly, I used to see that as a chore. Now I realize that it is a privilege and a best practice. That how it should be done. How does the Joni Mitchell song go?

Don’t it always seem to go
That you don’t know what you’ve got til it’s gone

Social norms and market norms are separate and you must not mix them. Social norms prevail in social situations. For example, if two friends go out skiing and one friend gives the others some pointers just for fun, that’s a social situation and social norms prevail. The instructor would find it absurd to be given cash at the end of it, but might feel slighted if the student didn’t invite him to his Super Bowl party. If a person goes out and hires a professional ski instructor, the ski school requires the student to pay full freight, but the instructor has no expectation of being invited to the student’s party.

In most of our lives, it’s clear which realm were in. However, in Predictably Irrational, which I’ve mentioned before, Dan Ariely shows the danger of mixing these two realms, because if you do, market norms typically win, though by themselves social norms can have a bigger effect. For example, when they pay people to do tasks, the people who are paid tend to perform more work in a given amount of time as their pay increases. But those who do the work as volunteers actually do more work than any of the paid subjects (see pages 70-75). People love to help other people and in the social realm we work for the good feeling that we get from doing something for someone. This is so powerful, in fact, that research shows that giving to others makes us happier than does buying something for ourselves.

You mess this up if you tell Aunt Marge how much your gift bottle of wine cost. Even if she knows it’s cheap or expensive, even if she knows the exact dollar worth of the wine, it fits within the context of social norms until the price is explicitly mentioned. But then, no matter what the price, it fits within market norms.

Companies mess this up all the time by trying to ingratiate themselves, pretend you have a relationship, you and the company are friends. But the second they hit you with a late fee and refuse to budge, the second they tell you that they have policies and can’t treat you differently than everyone else, they have violated the social norm and entered the realm of market norm. If it has always been a market relationship, that presents no problem. But if you’ve been courted like a friend, like your relationship is personal, like you won’t be treated like everyone else, the abrupt reentry into the realm governed by market norms feels like a betrayal. You end up having stronger negative feelings toward the company than you do towards companies for whom they never had any warm fuzzy feelings. It’s like the difference between hailing a cab and, upon reaching the destination, being asked to pay the fare. No problem. But if you ask a friend for a ride to the airport and at the destination you’re asked to pay “just half” of the cab fare because “we’re friends”. It’s a stab in the back and when companies act this way, they should be prepared to have accounts closed and to see virulent blog posts and horrid word of mouth publicity.

Ariely puts it thus:

If you’re a company… you can’t have it both ways. You can’t treat your customers like family one moment and then treat them impersonally — or even as a nuisance or a competitor — a moment later when this becomes more convenient or profitable.

Personally, I never become “friends” with companies, only with people. So no matter how much I respect a business, I don’t buy t-shirts with their logo and I don’t put their stickers on my car. So I’m disloyal, but I’m safe. But what about all those people who not only buy ice cream, but buy a Ben and Jerry’s t-shirt, that is they pay for the right to wear advertising?

Of course, I can be bought cheap. If I don’t hate your company and there’s a free t-shirt in it…

We as humans tend to key on contrast and judge value by the relationship of one thing to another. If we can find a comparable, we always do. The way Starbucks got us to buy $4 cups of coffee (er, you, anyway, since I have never bought a coffee a Starbucks, but I have bought a double chocolate cream frappucino) was to make the experience difficult to compare to Dunkin Donuts. Euro-style tables, funny names, funky music, soft lighting, all contributed to an ambiance sufficiently different to make the comparison difficult. Tough economic times, have made people more willing to see coffee as coffee and refuse to pay for the experience (that and, of course, the fact that the Starbucks experience has become mundane itself, just like Dunkin’ Donuts).

We all know that from personal experience, but I have been seeing it a lot more clearly since reading Dan Ariely’s fun book Predictably Irrational: The Hidden Forces That Shape Our Decisions and the interesting, though a bit more stodgy Robert Cialdini book Influence: The Psychology of Persuasion. So here’s where it gets interesting. Savvy marketers know that we judge value by contrast and relationship. So the Economist offers subscriptions for the following rates (or did when Ariely did his study):

1. $59 for the online-only subscription.
2. $125 for the print-only subscription.
3. $125 for the print and online subscription combined.

What’s going on there? Why even bother to offer option 2? Simple. It isn’t clear which is the better deal between $59 for the online subscription or $125 for the print subscription, but there’s no question which is the better deal between the print-only and the print and online option. Because of that and because those two are obviously comparable — different offers at the same cost — we key in on those two options. When Ariely showed the offer to MBA students at MIT, only 16% went for the online-only subscription, none went for the print-only option and a whopping 84% signed on for the combo. The deal was too good to pass up. But, and this is where it gets really really interesting, what if you eliminate the print-only subscription? After all, not a single person wants it anyway, so it’s not really an important part of the offer, right? Well when he offered only two choices, the online version and the combo (options 1 and 3 in other words) to MBA students, with no “decoy” offer, 68% opted for the internet-only option. So in other words, by focusing the comparison on the $125 option, they shifted from a measly 32% willing to pony up $125 to a whopping 84%. That’s the power of contrast! We are just not wired as humans to think in absolutes, which is usually a good shortcut as historically, evolutionarily (and in most life-threatening situations) we have very few choices and choosing quickly has advantages. In the modern marketplace, though, it’s a different story.

Cialdini has all sorts of examples where the contrast principle is used to influence our decisions. Brunswick pool tables instructed salesmen to start by showing the most expensive pool tables “just to see what the high-end features are” and then bring people down the price ladder. Result: a big increase in the amount people were willing to spend because the mid-range tables now seemed cheap. Some clothing retailer figured out that if a man comes in to buy a suit, always sell the suit first and the accessories second. After making the big purchase, what’s another $20 for a tie? But if they choose the tie first, they’ll go for the $10 tie instead.

This is also why discounts, coupons, MSRPs on cars that nobody pays, and “$97 value, yours for only $27″ work even if nobody in the history of humanity would consider paying $97 for the piece of junk that really isn’t even worth $27. Even though in our rational mind we know with certitude that the list prices are absurd and nobody pays them, they anchor us on high prices and we compare the sales price to the high price put in our mind because we are wired to compare. This is so subtle and so powerful that if you simply ask people what the last two digits of their social security number are, this will actually influence how much they are willing to pay for something later. Those with higher numbers are actually willing to pay more because the higher number is still stuck in their mind and that provides the mental anchor at that moment. In the absence of a meaningful comparison, they are simply comparing the last two numbers they have heard and that makes a price seem reasonable or unreasonable depending on what has become set as their anchor.

So as a consumer, you need to really think about what comparisons you make implicitly, without thinking about it. And as a merchant, of course, you need to think about what comparisons your customer is making.

I usually refrain from nasty notes. In most cases, I figure I’m better off encouraging continued good service with a letter of praise than sending a nasty letter in a vain attempt to change a place that gives bad service. When I do send a note of thanks by email, where responding is simple, I expect a response. Some companies do a great job. A friend wrote a note to say that Formula 409 had been really effective at cleaning a problem mess and he received a thank you and coupons for all kinds of free product.

So I recently wrote to a company praising their tech support (see below for a copy of the letter). I had some trouble installing a new DSL modem and getting connected. I wanted the tech rep’s supervisor to know that I got some of the best support I’d ever gotten, but there was no email contact for tech support, general contact, or a simple “comments and complaints” address. So I sent it to the only department that had an email address on the website: sales. Now, you would think if anyone knew the importance of customer service it would be the sales department.

To my surprise, I didn’t get any response. How hard would it have been for someone in sales to say “Thanks for contacting us. I’m glad we could help you out. I’ll see that your comments get forwarded to the tech support people”.

So I have two questions:

1. Why didn’t sales respond?
2. Why isn’t there a testimonials@zyxel.com email address?

The sales department’s failure to respond wiped out much of the goodwill created by the amazing technical support rep. If I had sent a complaint, I bet I would have gotten a response, but someone who sends a complaint may have already decided to hate you forever. Recovering the good will of that customer will be hard. If you do it right, that person will be loyal forever, but doing it right takes a lot of work. When someone writes you a letter of praise, though, it’s more like dating. That person is saying “I think I would like a long-term relationship with you.” You don’t have that loyalty yet, though. Think of it as a test, like a first date. Is there someone there? Is there a good response? If not, you’ve squandered that opening the customer offered. That’s what the Zoom modem people did to me.

If you want people to say good things about you, you should make it possible. So why don’t you have an email address for testimonials? You have “support”, and “sales” which should handle problems and complaints. Who do you have assigned to handle praise?

I am not writing for a sales question, but because there was no general contact email address on the contact page.

I just wanted to ask you to forward this to management and especially management in support to say that I was really pleased with the support that I got today on my Zoom 5615A modem. I’ve been tearing my hair out for two days going to a WiFi hotspot and searching the internet, calling my ISP, and made an inquiry to Zoom which got me connected, but then I ran into troubles when I rebooted and couldn’t get back where I was.

This afternoon, I could tell that though the tech was running through a step-by-step protocol, he seemed to know a lot more about modems than just what was written in his protocol. So I asked him a few questions, got him to explain some things so I could better explain my goals and my situation. Deviating far from the script and giving me the more fundamental knowledge I needed to frame my question correctly, he really sorted me out. Everything works now — modem, router, the whole deal. I would also commend him for his patience as I went through this screen and that screen and changed over to another computer.
[other identifying details snipped]

Without hesitation any child can tell you that it’s please. But in fact there is another magic word. Consider this study reported in Robert Cialdini’s book Influence: The Psychology of Persuasion (p. 4). People standing in line were asked in three different ways whether or not the person asking the question could cut in line. Here is the question, followed by the response rate in each case.

Frankly, I find it sort of surprising that 94% of the people said yes to because I’m in a rush which is barely a reason ("I’m in a rush because my plane leaves in one hour and I need to get this copied before I get to the airport" is a reason). But the amazing thing is that there’s such a huge difference between no because at all and one that adds no information whatsoever (obviously the person wanted to make copies).

It occurred to me that I should subtitle pages "Please read this because I wrote it" as in "The Magic Word (please read this because I wrote it)."

So I go stand in line and about 40 minutes after my arrival at the airport, I get a ticket agent who says she won’t check me in because they can’t check anyone in within 20 minutes of departure. It is exactly 20 minutes from the announced departure time which, by the way, was moved up by 10 minutes since I checked the website 45 minutes earlier. Finally, after getting a bit hot, she called to “reopen” the gate and get me on the flight. As I gather my things, she yells “Run sir!” I am still gathering some stuff (this is a total of about 26 seconds by the way) and she yells “Sir! You have to run now!” So I run. I offend a woman by cutting in line a bit during the security check. I get there and the gate is closed. Why? Because the aircraft is not even in the airport yet!

We finally load and the gate closes at 5:55. It was 5:00 when she told me she would not allow me on the plane because it was within twenty minutes of departure. As it turns out, the flight wouldn’t even land at the airport until 5:30. Why isn’t the customer service terminal connected to a database that updates in real time or near real time? In an era when Wal-Mart can forecast, not just record, how much it will sell of any given item hour-by-hour through any given day, how can US Airways tell a customer that he doesn’t have time to get to the gate when the plane is still thirty minutes from landing (and again, this is a small airport and you can get to any gate in five minutes).

Then when I get there, I find that the 4:55 flight to La Guardia is now scheduled for 5:20. Of course it is 5:20 and the plane is nowhere to be seen. Whatever. That’s not the bad part. The bad part is that the 12:24 flight to La Guardia is scheduled to leave at 5:49 (I think they may also need a lesson in significant digits). I can’t describe how mad I would be to see the 4:55 flight leave before me if I were on the 12:24. But it gets better. When I get on the plane, approximately 1/3 of the seats are empty. Why didn’t they just fill it with people from the 12:24? I can only guess that it’s because once again their database is not up to the task and they didn’t have a count of seats available.

So then I get to La Guardia and things go reasonably smoothly to get on my next plane, despite delays. My boarding pass says that I’m in row 9D. The guy in front of me is in 5F. The only thing is, the plane only has three rows, labelled A, B and C. After everyone decides that the obvious solution is to take any seat, since this is obviously completely messed up, the flight attendant gets on the intercom to explain the seating situation. Row A is A, Row D is Row B and Row F is Row C. Makes perfect sense. Somewhere there was a software problem. Since that rule meant that everything mapped perfectly and there was no conflict between someone assigned to seat 9F and someone assigned to seat 9C, presumably the computer knew all along that there were only three rows, A, D and F of course.

Of course, all that data exists. It’s all available in some systems, but not in others. As a result, customers are mad and confused. Customer service reps are harried and yelled at. But that’s okay, because the airlines have so many customers and are making so much money, they can afford to piss off customers. Or maybe not.

I’ve been a vegetarian since 1982 if I remember correctly, which is hard what with my brain fried from malnutrition from not eating dead cows and all. Anyway, back then going to restaurants as a vegetarian was a disappointing affair even in lefty liberal bastions like Burlington, Vermont, where I lived at the time. Burlington was on the verge of electing a socialist mayor and had already made minor local celebrities out of two hippies named Ben and Jerry, but it was still quite a challenge to find a vegetarian meal in a restaurant. When my brother took me out to eat and I had to settle for a baked potatoe for dinner, he declared me a cheap date.

Since then, it’s become much easier to find a vegetarian meal, especially in California, but for about the first time in my twelve years in California, I ran up against the wall. We picked up my hungry in-laws at Fresno airport to whisk them off to Yosemite and decided to eat at the rather fancy, supposedly award-winning Steak and Anchor restaurant at the Piccadilly Airport Hotel.
Once seated, it turned out there was not a single vegetarian item on the menu. I mentioned that to the waiter and he said, unhelpfully, that I could look through the menu for ingredients I liked, and ask the chef for something made from those ingredients. Since there were hardly any vegetarian ingredients mentioned, that didn’t get me very far. In similar situations in the past, the waiter has typically offered to go talk to the chef to get a list of options. Anyway, so what they served me was an attrocious salad.

When the restaurant manager came by to ask how everything was, I said politely that it was "fine" and then thought better of it. He would probably want to know how I really felt, right? So I called him back and said I was actually quite disappointed that there wasn’t a single vegetarian option on the menu. He said they used to have one on the menu, but it sold poorly, and therefore they removed it. At that point I had to explain to him that his math for counting vegetarians was wrong.

How the restaurant counts vegetarians

I’m guessing they figure the value of vegetarians like this:

• Add up the number of times a customer orders the vegetarian dish.
• Compare it to the number of times customers order each of the other dishes.
• If you’re planning to cut the three worst-selling dishes and the vegetarian dish is the second worst-selling, it gets cut from the menu.

How they should count vegetarians

But here’s the problem with that: vegetarians do not typically eat alone and they do not typically dine only with other vegetarians. They will not, however, go to a restaurant that has nothing to offer vegetarians. So if you lose the vegetarian’s order, you lose the whole group. So the proper way for a restaurant manager to count vegetarian meals is like this:

• Add up the number of times customers order the vegetarian dish.
• Multiply that number times the size of your average group.
• Now use that number to rank order the importance of offering a vegetarian meal on your menu, because that’s the true income that vegetarian meal represents.

The best estimates are that in 2008, 3.2 percent of Americans are vegetarians, according to research conducted by Harris Interactive and commissioned by Vegetarian Times. Additionally, 10 percent "say they largely follow a vegetarian-inclined diet." Let’s take the real vegetarians only though. That means that only about one in thirty customers should be a vegetarian. But let’s say that the average groups size is three. That means that the chance of there being one vegetarian in the group is 9.3% (see Notes below). But here’s the killer. Occasionally you have groups of 10, which is presumably a nice, high-profit group.
That group has a roughly 30% chance (30.1% actually) of having a vegetarian who will not want to patronize a restaurant with no veggie items. So you could actually lose out on almost 1/3 of all large groups. As group size increases, the chance of taking the vegetarian menu into account increases and your chance of seeing that group in your restaurant goes down.

Most restaurant managers get it. This one did not (though he was real friendly and nice). So here’s the end result of that: I will never go to that restaurant again. It no longer matters if they add a vegetarian item or not. It’s too late. They’ve lost me as a customer. I’m not being petty or vindictive, it’s just that I will never bother to check. I’ll try the Holiday Inn next time and if they have no veggie meals, I’ll just give up on the airport dining, but why would I take the time to go back on the off chance they’ve added something for me? It’s just not a good investment of my time. They’ve also lost as customers every hungry passenger I will ever pick up at the airport, whether vegetarian or not.

So Who Cares? I Don’t Serve Food

Fine. You can do without vegetarians, but that’s just one example of how a small group can cost you a lot of customers. In a similar vein, my mother-in-law is hearing-impaired. If a restaurant is particularly noisy, we don’t go back. Of course this can apply to any type of business. So you have to ask yourself — who can you afford to alienate and how much would it really cost you to minimally accomodate that group?

Notes
Some people assume that you figure cumulative probabilities by adding, so if the chance of any one person begin vegetarian is 3.2%, the chance that a group of ten has at least on vegetarian is 32%. Obviously that doesn’t work, because if you had 100 people the chance would be 320%, which is not possible. Actually, if x is the probability that any one person is a vegetarian and there are n persons in the group:
probability as percentage = 100 * (1 – (1-x)^n)
Got it?