Let’s pick on the Visitor Loyalty graph, which has always bothered me. It looked like this:

Google's Old View of Loyal Users

And now, the new-and-”improved” graph looks like this:

Google's New View of User Loyalty

The only thing Google improved here is changing the name from Visitor Loyalty to Frequency & Recency. The concept of “loyalty” is relative to your users, their goals, and the context of their usage, so frequency may not even be a good indicator of loyalty. For example, a loyal Google user may visit google.com more frequently than a loyal Amazon user visits amazon.com. However, a loyal Google user probably views less pages per visit than a loyal Amazon user. That does not mean that either user is less dedicated to the site, they just fulfill their needs in different ways.

So, the name change is good, but what has bothered me most about the old graph still hasn’t been fixed. My concern is that it’s virtually useless to anyone who spends time with it, and misleading to anyone who only glances at it.

Apples Meet Oranges

If you look down the left side of the graph, you’ll notice that the scale is all wonky. It goes from 8 visitors to the range 9-14. This weird incrementing creates what appears to be a bump in the data around 9-14, when it is probably a smooth curve if you map it out correctly. If that’s not confusing enough, each range is different. Another way to express the increments they are laying out would be: 1, 2, 3, 4, 5, 6, 7, 8, next 5, next 10, next 25, next 50, next 100, everything else. So, while the bump at 201+ may cause an initial wave of euphoria, you soon realize that the 159 visits in that bump are probably spread evenly between 201 and 360.

(Before you ask, I’ve checked across all of the high-volume and low-volume sites we monitor for our clients. The wonkification of the scale doesn’t change.)

I get what they are trying to do. Loyal users will show up in clusters around one of these increments, but when the increments continue to adjust along the span of data, it’s impossible to compare one chunk to the next. What seems to be bumps are probably nothing, but there’s no way to be sure. There cannot be a generic benchmark for frequency that applies to all sites. Trying to genericize a graph using arbitrary increments is a disservice to the data and only misleads your users.

So, how would YOU do it, Phil?

Well… the first thing I would do is make it a line graph to generate a smooth curve with a consistent scale from 0 to X, where X is the most number of visits recorded. That’s easy to do and Google Analytics uses line graphs elsewhere, so we know they have the graphing capability. It would look kinda like this:

A Useful Visitor Frequency Chart

OK, Phil, why is THIS better?

It’s better for me, but maybe not for everyone. It depends on what you need to do with the graph.

I was once asked by a client to briefly explain how I find odd things in their analytic data before anyone else. There are many different reasons to be looking at analytics reports. I play a troubleshooting role on that team, so my answer was simple: “I follow the trend line and look for aberrations. If I find one, I dig into the data to find the root cause. Then I call you.” The chart above wouldn’t set off any alarms, but the chart below would cause me to dig:

Oooh, Piece of Candy!

That peak would make anyone wonder what’s going on, but with the bar charts that Google uses, that peak would end up buried inside one of the wonky ranges.

But that’s only how I would use it for a role I play on a specific team for a site that has unique needs. That’s the weakness of the Google Analytics interface that I was hoping they would fix. I want to be able to customize their charts like I can in Excel. Sadly, the data visualization tools they offer in the new interface aren’t enough. Even more sadly, when I export the data from the wonky chart, it retains the bizarro scale. So even if I wanted to use my own charting software, the data is garbage. To quote my film editing professor, “garbage in, garbage out.”

What Would Tufte Do? (WWTD)

When I encounter poorly communicated data, I ask myself, “What would Tufte Do?” Edward Tufte has authored several books on data visualization.  While his work has sparked much debate, I believe he would have a lot to say about the use of bar charts for something like Loyalty or Engagement.

You see, even though the charts presented by Google Analytics are missing the chartjunk Tufte has railed against, they rarely communicate data in an appropriate manner. A bar graph is absolutely the wrong tool to use to analyze this set of information. To select the appropriate mapping of this data, you first have to ask yourself, “What is the user going to do with this data?”

For something like engagement (which Visitor Frequency hints at), you may want to know:

• What are the behaviors of an engaged user?
• Are we losing, gaining, or keeping a consistent level of engaged users?

Both of these would require completely different visualizations of data, but neither is served by the out-of-the-box graphs presented by Google. My gripe about the out-of-the-box graphs in Google Analytics is that I often can’t figure out what they are trying to tell me, and they don’t let me customize the graphs to meet my needs.

We are entering a period where strong data visualization is getting a lot of attention. It’s more than making pretty pictures, it’s making images that provide value and answer questions. Google Analytics is a valuable tool for an unbeatable price, but when a new version is launched and the greatest value you’ve added is a UI facelift, you’re kind of missing the mark.

As those of us who were there know, and perhaps much to the chagrin of those who weren’t, there’s a lot of chatter about SXSW on the social networks – a lot of it along the lines of…

“I’m having the best time ever!” or “I’m having lunch with @pixelgeek24 and we’re going to conquer the world! We’re ninjas/rockstars/superheroes! W00t!”

That’s all fine and good, and generally aligns with the typical self-promotion and non-wave-making that you’ll see clustered around conferences. We’re all guilty of it, and for a lot of folks that’s the whole point of engaging there in the first place.

But there’s a special kind of walking on eggshells that seems to happen around SXSW – because I found a totally different undercurrent of sentiment lurking in the physical realm that isn’t nearly as represented in the Twittersphere: being in Austin with a lot of like-minded folk has its definite positives and potential, but the conference itself is decidedly “iffy” (note: the health track held in the Hilton Garden Hotel was the one shining, glorious exception). In fact, I talked with several folk who probably wouldn’t want to go on record saying it, but who said that they’re seriously considering next time just coming to Austin during SXSW Interactive and not even registering for the conference (actually, if you read between the lines on Twitter, it was easy to see just how many were regularly out gallivanting during sessions already this time around).

Love Austin...But The Jury Is Still Out On SXSW Interactive

Why might that be? Well, here are my top 5 reasons:

1. Picking a valuable session is like trying to make your way through a minefield with your shoes tied together…after downing a 5^th of Jack. Aside from the sheer tyranny of choice triggered by having to select one of 20 or more sessions occurring during each time slot, they’re also spread out all over the city – putting even more weight on making the right choice, and making it all the more painful when you choose…poorly. That happens to be an extremely easy thing to do – ranging from something that is probably decent but just isn’t as relevant to you as you thought it might be from the description, to someone with the presentation skills of a handball and/or really weak content. Give me a single-track powerhouse like An Event Apart any day, with vetted speakers and solid content – I know I’m going to walk away with value. The Russian Roulette at SXSW is brutal.
2. This minefield has resulted in a culture completely accepting of getting up in the middle of talks and walking out, and then barging into other ones already in-progress. I guess I can’t blame people for this, and I must confess I walked out of a few sessions myself. But I was completely floored by the audacity of people walking into sessions I was genuinely interested in 45 minutes late…sitting down for 5 minutes…and then getting up and leaving again. In almost all of my sessions I endured a never-ending stream of KA-CLICK…KA-CLACK! – the doors opening and closing relentlessly. Again, it’s a symptom of #1, and while the symptom was treated at some sessions by volunteers trying to minimize the impact by holding the doors, it’s not getting at the disease.
3. The whole thing is mind-numbingly commercial. You know when you’re walking down the strip in Vegas and there’s those people slapping postcards and pamphlets against their hands to get your attention, and when you make eye contact they accost you? Same deal, only with techy tchotchkes and other nerdy stuff of questionable value. Absolutely everything is sponsored and plastered with a logo…and that’s just a drag.
4. The “must see” sessions that you know are going to be cool/interesting are so over-attended due to the size of the conference that it’s not even remotely enjoyable. I can’t think of too many reasons I’d submit to standing in a line that stretches for two blocks (and maybe for this reason I am getting old). But I’m definitely not doing it (again) to then be crammed into a tiny tent where people are packed shoulder to shoulder like an underage frat party, or to have the doors be closed in front of me because the venue is at capacity. If they curated the sessions better (while still keeping the important democratic element so it’s not all “celebrity” speakers) they could increase the size of the rooms and improve the experience greatly. And hey, while we’re at it, subtract a day, cut the session count in half, and see how that separates the wheat from the chaff.
5. Bigger does NOT equal better. 25K+ people feels like being on a college campus when classes are changing. There’s nothing intimate, communal, or special about the way you feel racing from one session to the next, trying to find a place for lunch without a monster wait, or battling for a free power outlet.  At least not the way the conference is run now. No wonder people elect to just step out of the flow and do it on their own terms (jeez, Seth Godin even does that at TED) – the conversations I had with folks outside of sessions were certainly the most valuable part of the experience…but I still genuinely like to be enlightened by a killer session.

OK, whew…I got that off my chest. I would have liked to see an honest review of SXSW Interactive before going because the experience wasn’t really in line with the chatter. Maybe this year was the turning point and it wasn’t always this way. But as I finished writing this I was informed that I’m not the only one saying some of these things…so I’m hoping the organizers are taking note and thinking long and hard about how to preserve the experience if they choose to keep growing – because it’s breaking.

Pssst – you’ve got our contact info if you want to fix that.

P.S. I’ll be posting a follow-up soon on the fantastic health track and the few outstanding sessions I did attend.

Full disclaimer: after almost 20 years of attending and speaking at professional conferences, I’ve become more of a fan of smaller conferences. In fact, I’m a huge fan of single-track conferences like AEA or something of the TEDx variety. While SXSW Interactive has grown into a megaconference; the SXSW reputation, recommendations from our pals, my film background and a longtime desire to visit Austin convinced me that it’s worth checking out.

I’m still hopeful, but I have to say I’m disappointed so far. It would be kind to say the SXSW online experience has been difficult. Apologists may defend it, but c’mon.

After registration, create your schedule by re-registering on the schedule site, which flips you over to SXSocial to register there. Got it?

I could enumerate the issues users encounter as they navigate registering, paying, finding a hotel, re-registering for SXSocial and picking sessions to attend – but if you’re going to SXSW, I suspect you know what I’m talking about. If you’re not attending, well, let’s just say they don’t follow the principle of “design it so it just works.” It’s more the principle of “eventually, they’ll figure it out.”

But maybe it doesn’t have to just work. SXSW Interactive is successful enough to book the entire Austin Convention Center and every hotel in town. Maybe the online issues are because it is so huge. That’s a lot of information to manage. But isn’t that enough justification to take the time to design a more effective online experience? At the risk of upsetting the SXSW Gods I hesitate to point out that the word “INTERACTIVE” is in your title. From where I’m sitting, it doesn’t feel like any of the decision makers said, “The service the site provides is the beginning of the SXSW experience. It needs to be integrated and seamless.”

If it’s so huge and so successful, can’t a few more dollars be invested in some thoughtful online experience design? Wouldn’t that be in line with many of the tenets proposed by the interactive sessions?

I’m still looking forward to the conference and checking out Austin. If you’re going to be there, let us know. If not, we’ll do a few reports from the road.

LEVEL 1: FAST FOOD

At its most basic, it appears as “would you like to Super Size that for just a dollar more?” – it’s a pretty logical question in the workflow, doesn’t take much time, and requires only a simple “yes” or “no” to get past it. Fair enough. I have no doubt that this little phrase has proven extremely lucrative.

LEVEL 1.5: CHAIN CAFÉS

A slightly evolved variant is the “can I get you something to eat with that?” that you’ll now experience at Starbucks, Seattle’s Best, etc. It irks me a tad that I have to field this question anytime I just want a simple coffee drink, but maybe I’m in the minority there (I’d be willing to bet the contrary, though). It’s a slight interruption, one question long, and I guess it could be argued that it’s reasonable given the workflow.

LEVEL 2: RITA’S WATER ICE

It becomes a little more irritating as the queries veer further from the likely goal in pursuit of the upsell. Ever go up to a Rita’s Water Ice on a scorching hot day, order up a cool, refreshing treat, and then be asked “would you like a hot pretzel with that?” Hm. Actually, when I started sweating through my shirt and pulled over for some refreshment, I hadn’t really considered a warm hunk of bread. Now I’m used to it of course, but the first few times it definitely threw me. Yes, in the grand scheme it’s a small hurdle between me and my cherry gelato, but OK, now I know Rita’s has pretzels. And you do too.

LEVEL 3: BEST BUY, APPLE, CAR DEALERSHIPS

Let’s ratchet this up a notch. Go buy some electronics from Best Buy or Apple, and wait for the significant upsell at checkout for a warranty or “Apple Care” protection plan. And how about buying a car? After you’ve made peace with the monthly payment you’ll be shelling out for the next five years, you need to meet with the head of sales or someone similar and endure the endless waterfall of warranty combinations that could ratchet it up hundreds of dollars…and the “hm…should I?” question is sometimes literally phrased as a question of life or death. I always feel assaulted by this process. One reason is that at no point can you simply end the transaction – you’re forced to hear the explanation of each option, and it’s time intensive.

JUST PLAIN ANNOYING: THE USPS

And finally, the upsell experience that got me thinking about writing this in the first place: the U.S. Post Office. Every time I go into any local post office, the experience goes something like this:

1. Stand in line and clearly listen to every person in front of me endure the script I will be hit with when I reach the counter.
2. Approach the counter with an oversized envelope to ship.
3. Be asked if my envelope contains anything liquid, fragile, perishable, or potentially hazardous. It’s federal law, so even though it’s annoying and I said “no, it’s only a DVD” after “liquid”, and then “no, no, no” afterward, I get it.
4. Be told that it can arrive by 9:00 AM tomorrow morning for $32, though I already said I just wanted whatever was cheapest. Repeat this, and perhaps hear one or two more options before reluctant acceptance that I’ll just be going with Priority Mail for $1.32, thanks.
5. Be asked if I want confirmation or tracking. I’ll give them this one, sometimes I do.
6. Be asked if I want any stamps. I’ll give them this one too, sometimes I do.
7. Be asked if I want any packing materials. Now this is getting a little silly. I saw the packing materials display, and if I wanted them I would have grabbed them. I endure.
8. Be asked if I want to open a P.O. Box. OPEN A P.O. BOX. Every. Single. Time I’m in there. Who in their right mind figured they should add this to the script these poor clerks have to recite all day long? FOR WHOM IS A P.O. BOX AN IMPULSE BUY?
9. You get the picture. Depending on the length of the line in front of you, this process can eat up your lunch hour in a jiffy.

Now think of your online business transactions. Most self-respecting e-commerce retailers wouldn’t think of forcing their customers to endure experiences like those described above (to be sure, there are definitely still exceptions). They know that if they employ annoying upsell tactics they’ll be skewered by the design community and their users will flee for alternatives. Teams of goal oriented, user-focused designers (the kind that are often employed to design digital experiences and too often not for real-world ones) would never allow this kind of workflow interruption and non-sequitor questioning/clicking on the way to goal completion.

For instance, on Amazon the upsell is EVERYWHERE, but it’s off to the side and there if you’re interested. On Amazon, you don’t see your final cost and THEN get solicited for additional warranties and protection packages to jack up that price (though a default to Free Shipping when it’s available would be a nice touch, guys). On Amazon, you get recommendations based on what is in your cart and what you’ve purchased before, not random recommendations from a vast list of products simply because they exist. These are just a few examples of how digital experience design has now evolved to the point where it can have positive impacts back to poor experience design in the real world. We’re not going to get away from upsell, and I’m not suggesting we should – but like most things, there’s a right way and a wrong way.

Try To Escape Without Being Upsold One Of These Bad Boys

My wife recently visited the store at the Willow Grove Mall in PA (for what she says is the last time), and stated that she wanted to buy a small gift for no more than $10. She was shown a specific tea and told that she couldn’t buy it in a quantity less than $18 worth (this is loose tea we’re talking about here so I don’t quite get that, but we’ll let it slide). She said that was over her limit, and made some other inquiries to try to find something for $10. She then had to endure a volley of being directed toward kettles $20+ and other expensive add-ons until after enough restatements of the $10 limit the clerk took it upon herself to just start bagging the $18 worth of tea as if it was a done deal. Finally, exhausted by the exchange, my wife stated “you know, this is why I don’t like to shop here – I always feel like nobody’s listening to me. It’s like they have their own agenda, and I always walk out feeling bad about spending more than I had intended.” From what I’ve read online and plenty of anecdotal evidence, this is a pretty textbook experience – heck, the company’s ethics are even disputed on their main Wikipedia page.

Our story had a little extra kick to it, though. At this point the clerk shot back “fine, see ya. BYE!” She then refused to give my wife her name, yelled at the other sales people daring them to reveal her name, and apparently said that she was the manager though the guy giving out samples at the front of the store quietly confided that she wasn’t. For two days afterward my wife tried to call Teavana to discuss this and got steady busy signals. How’s that for service? For more first-hand evidence of bizarro behavior like this in the Teavana ranks (from the very top down) and the culture that fuels it, check this out. For those who might say “it’s your fault if you let yourself be suckered by them”, take note – they WOULDN’T settle for the $10 sale.

When Bad Things Happen To Good Teas

It comes down to this: when you’re in the retail game, your strategy should be to provide value so that people will continue to buy from you and also refer you to others – that’s a no-brainer and pretty universal. Of course you also want them buy as much as possible, and there’s nothing wrong with “upselling” at a fundamental level. But upsell has to deliver on value, which traditionally is triggered in your customers’ minds because you’re providing the highest quality or most innovative products (think Apple), superior customer service/experiences (think Nordstrom), or you’re highly operationally efficient (think McDonalds) – or some carefully considered combination of the three.

But here’s the rub, and it’s where Teavana fails in a big way: the effective way to upsell is to present the opportunities to customers in a way that they clearly perceive the additional value and then make an active and informed decision to purchase. In other words, if you trick people into buying more than they had intended through sneaky practices, they’re not going to walk away feeling like they splurged a bit and got something they wanted in return – once they realize what just happened to them, they’re going to feel swindled.

What’s really a shame here is that Teavana has some fantastic tea. I recognize that traditional tea has yet to really go mainstream in the U.S. and comes nowhere close to sales of coffee, soft drinks, etc., so the business of selling it requires being creative. But when customers tell stories like my wife’s and hear from others “I know, whenever I leave there I just feel dirty”, you would think that might give business owners some pause. Put simply, if your sales strategy revolves around underhanded upselling, you’ve got a bigger problem than sales.

However, thanks to U.S. Airways (and I suppose to the poor practice of overbooking by the airline industry in general), those plans were dashed. The culprit: multiple severely overbooked flights in a row, no takers for a consensual bump with a future voucher, and general mayhem and confusion on the part of the terminal employees. Of course I could give you a detailed blow-by-blow, but my own particular experience isn’t really the point – we’ve all been through it, so you more or less know the horror story. What is more important are some of the overarching things I thought about as it was all unfolding that would have either avoided the situation altogether, would have made it more bearable, or can serve as warning signs when designing experiences elsewhere.

1. Don’t Make Your Customers Suffer For Your Own Business/Industry Shortcomings
Look, as a business person I know that it is common practice for airlines to overbook flights, and I know there are algorithms to determine by how much they should oversell to generally have full planes without too many people getting hosed (like I did). I also know that the organization takes a hit that really adds up when there are unsold seats on flights. However, as a passenger with plans, I DON’T CARE about any of that. It’s not, and it shouldn’t be, my problem – I have no sympathy for any of that when I’m standing there watching my plane take off and I’m still on the ground. I’ve gotten used to not getting meals anymore and having to pay for everything I consume or “borrow” on the flight, but when the fundamental thing I’ve paid for – transport itself – is denied to me based on something that in no way is my fault, it simply feels like a swindle. I can’t think of another industry that gets away with behavior quite like this…why does flying always feel like a game of roulette? Why do we willingly continue to pay for something that we know (weather aside, even) might not be worth the paper it is printed on once we’ve packed our luggage, taken time off, gotten ourselves to the airport, and paid for parking? Why are we always pretty shocked and bemused when it goes smoothly and we end up where we’re supposed to be on time? I realize we’re not talking about something as simple as running a lemonade stand here, but as consumers we’re also paying really good money for this service…why is all of this emotional strain being put on the customers, for something that the business can’t effectively figure out? The industry needs to find more effective ways to manage schedules, cut costs and/or maximize efficiencies, to make sure that we actually get what we pay for and can be confident in our purchase. Without that, you’re certainly not going to increase the amount of people lining up (and emptying their pockets) to say “thank you sir, may I have another?”

2. Don’t Burden Expensive and Important Transactions With a “Luck Factor”
Our flight was oversold by 12 tickets, and we were numbers 8 and 9 – having checked in at the airport a little less than 2 hours before the (domestic) flight. Some people have said to me, “well, you should have checked-in online during the day”. Well, it just wasn’t that kind of day – we’ve all had ‘em. Regardless, that is airline apologetics, and we shouldn’t be in the business of giving them an “out” – would you say that to your grandmother who doesn’t really get what “online” even means? It isn’t OK that we’ve learned the rules to their flawed game so therefore we can exploit it at the expense of those who haven’t – that is just unfair. A ticket purchased in July shouldn’t require another mad-dash of confirmation on the day of the flight three months later to make sure it’s still valid.

3. Set The Rules, and Stick To Them
I wasn’t happy about the oversold situation but I wanted to be clear about the rules, so I asked. “You said we’re numbers 8 and 9 on that list, right? So if 9 seats free up, we’ll be on that flight? Right?” I was told that was correct, by the attendant who was being actively badgered by a woman who announced that she was 12th on the list and HAD to be at work that evening in Nashville. So explain to me then how that squeaky wheel got on the flight and we were still standing in the terminal? Because the rules were broken. Look, I know it can’t be easy to listen to that one (or those few) nut-job(s) jabber on about how much more important the flight is for them than for anyone else. But rewarding that behavior not only destroys your organization’s credibility in anything they “officially” say or do, it reinforces those behaviors in the future (and therefore makes your life even more difficult) – because in this instance you can bet you’ll read that little “tip” on a travel blog.

4. If You Can’t Avoid It, Have A Very Clear And Well-Communicated Contingency Plan
OK, we’ve already established that it’d be great if the need for the whole ugly practice of overbooking could be squashed somehow. However, in the end we’re talking about an industry that is really hurting, so we may already be at wit’s end as far as that is concerned. OK, but how to mitigate that then? U.S. Airways chooses to keep this practice as quiet and ill-explained to its passengers as possible in hopes that it’ll only affect a small percentage of them and things will generally just clunk along as they do now. Interestingly enough, as a result it is not only completely baffling and frustrating to passengers when they get bumped, but also to U.S. Airways employees – who obviously don’t know what the correct next-steps are, what the lines of communication should be (witness us being sent to 4 different desks, having to re-tell the exact same story 4 times to people who had no idea what we were talking about and generally treated us like we were criminals trying to steal something from them), or how to maintain their composure in the midst of what is no doubt a frustrating experience for all.

Until Next Time, Nashville

What it all boils down to is this: we missed an important family visit and a concert. The airline owes us more than just the price of the flight refunded, but at least we did finally receive that several days later (I wasn’t too confident even that would happen). I was ticked, but even more ticked when I thought about things that other people have missed as a result of these practices (or, at best, just bafflingly terrible experience design). I posted my plight on Facebook and heard from people who had missed weddings, funerals, and more. This isn’t buying a TV, bringing it home, and finding out it doesn’t work – this is denying people visits with their friends, family, and attendance at major life events. You cannot possibly provide me with an excuse good enough for that.

…which is a darn good thing because below that little “inline message follows” indicator? Yeah, you guessed it: there was no message following (although even if there was, that still might not get the point across (by itself) to some users).

1. Skip The Tech-Speak, Threepio
So my first gripe is that this is a techy and jargon-heavy message that wouldn’t make sense to a lot of people who, based on the appearance of this message and lack of any kind of context, might either a) think it was spam and delete it or b) not have any idea what it was trying to convey and in both instances might very likely have no idea that the message hadn’t actually been received by the intended recipient. We all know that universities frequently deal with time-sensitive and confidential information, so wouldn’t it be nice to let the sender know what had happened in no uncertain terms? What is a permnent delivery error? What exactly does that mean? Something like this could mean the difference in an application being received on time or something even more serious. It’s unacceptable for that to be at the mercy of poor interaction design.

2. Beyond the Message, Look at the Vehicle
In some ways though, even the poor messaging is beside the point because the sender of the message should NEVER be the recipient of an e-mail like this. Why? Because THEY CAN’T DO ANYTHING ABOUT IT. If I send a message to someone whose mailbox is over its size limit, well, that’s on them – it’s not MY fault – why am I made to feel like I’m being shut out with little more than what is essentially a “tough luck, kid”? What if I don’t have any other way of contacting them? What do I do now? If I’m crafty enough I might make a call to Temple to find a phone number or a name of someone else in the department I might be able to contact…wasting my own time as well as those who have to route my call (or another e-mail to an inbox that isn’t over its size limit).

3. Mail Quotas? Seriously?
OK, Temple is a large organization and I guess I understand mail quotas on some level, particularly if students are on the same server and they’re sending large movie/sound files around, etc. But for business users to be hamstrung by those same quotas feels incredibly inefficient. Regardless, here’s a newsflash: as a user; I don’t really care what your technical issues are – the experience is all I know, and it’s not good. Space and storage is cheap these days…so this feels antiquated. But even so, I’ve worked with organizations that have mail quotas that are instituted much more intelligently – the recipient gets a message that their mailbox is over it’s limit, all messages are still received, and they simply can’t SEND any messages until they clean up their room. It’s still a little old-school, but at least it’s on the owner of the mailbox to fix it.

GRRRR.
This is just one of those days that poorly designed technology is infuriating me. Add the fact that I had to type this twice because during my first attempt I hit “Backspace” (I know, hitting Backspace while typing? I should be committed) and my browser decided I wanted to go to the previous page and erased what I was typing. Thanks, Internet Explorer! That was a helpful keyboard shortcut!

And how about when I went to a page earlier to demonstrate some functionality, started typing my username and password, and after a few seconds the page automatically placed the cursor back in the first field and I exposed my password to the whole room?

As a designer, I know there are technical reasons these things happen. Let me reiterate: as a user, I don’t CARE. Enough apologetics. Fix it.

It doesn’t have to be this hard people. It really doesn’t.

The gist is that providing people with more choices is not necessarily a good thing – and in some situations can actually be debilitating. It makes sense – maybe you’ve had a similar experience to me: I’m in no way an expert on investment funds, but when it comes time to figure out where to put my money, I’m faced with a daunting list of options – and I’m such a novice that I can hardly even make heads or tails of the differences between them aside from a few very broad categorizations.

(Illustration Courtesy of Scientific American)

What’s worse, once the decision is made, since it affects me on an ongoing basis I continually ask myself “did I make the right choice? Was/is there a better one? How will I know? What am I missing out on?” As a result, there are plenty of people who just disengage altogether because they can’t or don’t want to deal with the various stresses associated with it.

So how can we tame this beast?

1. Don’t Provide Choices That Are Invalid And Prevent Users From Achieving Goals

The first time this resurfaced for me recently was at An Event Apart, during a usability session where Robert Hoekman, Jr. was doing on-the-spot usability reviews and was looking at a website for an airline (we’ll let it go un-named and give them a chance to act on the collective feedback of the audience at AEA). Even though the airline only flies out of certain airports, in their “select a departure city” drop down list there are hundreds of major and minor ones – many of which (or most of which, based on our results) they don’t even have flights out of.

Oof.

Is this an out-of-the-box list or something? Why in the world would they do this? Better yet, the drop-down for destination city includes the same list, which isn’t truncated in any way after the departure selection is made. If selections are made that aren’t 100% compatible with available flights, the user feedback is simply “Sorry, there are no flights that match your criteria.” This means that even if you are lucky enough to select a valid departure city, if you select an invalid destination, you get the message above and no prescriptive feedback indicating where the problem lies. Aaargh, we could spend paragraphs on this alone – but the faults should be obvious. The user experience is only a few notches shy of this.

2. Don’t Provide More Choices Than Are Necessary in the Interest of  “Showing Off”

Next I had an interaction that I’ve experienced plenty of times over the years in different guises. Without going into too much detail it boiled down to a desire to use valuable screen real estate to display numerous options and features that aren’t available (and therefore aren’t helpful) to users. This was an effort to show off the robustness of the product with all of the bells and whistles turned on. Instead of considering the tyranny of including choices that are smugly turned off, the thought was that a user would want to know why they couldn’t access those options and would try to find out how they could pay to turn them on. To me this is akin to visiting a car dealership and the salesperson trots you past every make and model on the lot regardless of your price range or other personal preferences and parameters. That doesn’t sound like a whole lot of fun, does it?

Walk The Lot With Me...

It’s OK to provide users with information about what other choices are available, but at best it should be when they ask for it – and at worst, it shouldn’t be a hard sell. Most people will want to get in and accomplish their goals quickly, intuitively, and with minimal distraction – not be constantly reminded of (and confused by) things outside of their workflow. Otherwise, your advanced user will see what you’re up to, and your novice will be confused and frustrated. Make your core product a killer, and people will ask what else you can do for them on their own terms.

3. Don’t Make Users Responsible For Making a Choice Unless Absolutely Necessary

This one is more of a stretch than usual and my example isn’t technology based, but I still feel it’s a great illustration of the mental strain that choices can create. Without getting melodramatic, our dog very recently succumbed to a debilitating disease and had to be put to sleep. There’s enough emotional baggage inherent in that as it is – but as my wife and I tried to decide what to do, we came to the conclusion that we simply couldn’t make the decision while it was a CHOICE. There would be too many “what if” questions; so we would do whatever we had to do (short of completely draining our bank account) if things were still at a stage where it was a “choice”.

Even when things were looking pretty dire, initial meetings with vets at a few locations had them saying things like “well, you could try acupuncture…or you could try putting the dog on wheels” – both of which would have done nothing for our dog in his particular condition, but were choices thrown out there (somewhat offhandedly, and I felt, irresponsibly – because of the pressure they placed on us) that confused us and made us wonder if we were doing the right thing. Those additional choices just didn’t seem consistent with what we knew of the disease and what we could see with our own eyes, but still we lost sleep over them.

Finally we met with a compassionate vet at Rau Animal Hospital who removed the tyranny of choice, explained the fallacy of the other proposed choices, and suggested to us that the most humane thing would be to let him go. While that was tough to hear, it felt like someone pulled a 10 ton weight off of us. The burden of choice had been removed, and we could make the decision to let our dog go without further loss of dignity or quality of life.

Rau Animal Hospital GETS IT

This is an extreme example with a heavy emotional component, but it proves a point – often times when you think you’re providing benefit by suggesting choices, it’s an easy enough thing for you to do but can have crippling effects on the novice users of your system. We’ve all interacted with clumsy systems that have way too many nodes of choice even when they could easily learn from individual user behaviors (ATM machines, anyone?), and some that even force you to make choices when there is only ONE OPTION (don’t get me started on the haphazard implementation of EZPass).

You don’t gain repeat business and loyal customers by increasing the amount of time someone spends interacting with your system – you gain it by minimal inputs required to accurately achieve a desired result, minimal cognitive (and emotional!) processing, and keeping advanced features and functionality available for the nerds who want to look for them but out of the way of those who just want to get on with their day.

The Culprit

That resulted in my heading out to buy an over-the-counter (OTC) inhaler for mild cases such as hers – only to find out that at 4 out of 4 stores I visited, after being told the aisle in which I could find them, they were no longer stocked. None of the pharmacists seemed to have a clue as to why, except for the last fellow, who believed they had been discontinued…but didn’t have any details. The only alternative? Being told that perhaps taking some medicines designed to treat a multitude of unrelated things might help, or taking the one product I could find (in only 1 of the 4 stores) that was supposed to treat asthma….which turned out to be a pill, subsequently took hours to kick in, and kept my wife up all night. Score!

We had no idea why OTC inhalers were no longer available, just that they weren’t – and nobody seemed to know why. Even those who really should be in the know said it was a pretty mysterious thing and didn’t have any additional insight. To me, that felt like reaching out for a banister that had always been there – and finding it missing. It would only take escalation of a few conditions to turn that into a trip to the emergency room instead of just up to the drug store. I want to know why I can’t get that good hot salsa at my local grocery store anymore, but I really want to know why I couldn’t get my wife something she had previously relied on when she was short of breath and there was no decent substitute. Not only does the lack of communication feel insulting and make me resent those responsible, it’s downright dangerous. When they discontinued the contraceptive sponge, everyone knew it (heck, I did, and I was just a kid) and was abuzz about it – where was the communication this time?

So I turned to my most reliable source of information – Crazy Aaron, who, before I ever even asked him, sent me an article he had read in “Spray – Technology & Marketing for the Aerosol Industry” (I kid you not), because he was intrigued by the human factors issues involved with the discontinuation of CFC emitting inhalers. You can’t make this stuff up.

Spray didn’t provide an answer as to why there are no new OTC inhalers (although the change from CFC is probably at the root of it), but it seems that the folks in charge of the change-over to the new HFA 134a-propelled prescription inhalers (buy stock!) missed a few opportunities when switching to the new product. New mandates were designed to improve the product, but resulted in a bit of a bumpy rollout (particularly for life-long users of the previous versions) due to a lack of understanding of the full needs and habits of the audience. Here’s what some good solid user-testing could have helped avoid/better manage:

1. Dose Counter

Older products would, as you got close to the last few doses, become noticeably weaker – a strong indicator to replenish. A new regulatory requirement states that every dose has to be the same strength and potency. Sounds good on the surface, but how do you know when it’s kicked? You don’t, until it is. We can all do the math on the dangers of that scenario. Suggestions included telling patients that they’d have to keep track of the doses they’d already used themselves (talk about user-friendly). But chalk one up for GSK: they are (so far) the only ones to introduce an inhaler that has a built-in clicker on the side that tells you how many times you’ve used it. Win!

2. Taste

Apparently, the old stuff that everyone was used to had a mild taste of “rubber or cabbage” (sweeeeet). Not ideal, but it went with the turf and people had learned to deal. There are two kinds of formulations available now – one without alcohol in it (which is more tasteless than before and surprisingly garners mixed reviews for that reason), and one that has 10% ethanol and produces a burning sensation at the back of the throat and can taste weird due to impurities in the alcohol. I wonder if the lack of a standing ovation for that little development came as a surprise?

3. Speed/Energy

Older products would give patients a “punch” of “chilled” air that many of them equated to it “working” and getting into their lungs faster. The newer products deliver a slower moving plume of air with finer particles in it – proven to be more effective. Some patients think it is weaker and not working as well as the old ones. Sounds like solid communication and training might be the answer for that one.

4. Finer Particles = Nozzle Blockage

Those finer particles mentioned above? They get the medicine to the lungs more effectively, but they’re more likely to clog the nozzle – so now patients have to regularly clean it to prevent that – something they never had to do before. Is there a mechanical fix here that was too cost prohibitive to undertake? Is there some alternative to adding additional maintenance chores?

Beyond that are a few other issues like storage (the new inhalers are much more susceptible to moisture and can be rendered ineffective much more quickly depending on the environment) and even electrostatic charges making doses less effective. Did they miss the boat entirely? Absolutely not – the negative feedback came from a minority of patients, and we all know that when it comes to pharmaceutical regulations, you can have your hands tied in so many ways that you’re lucky if you ever get a helpful product to market – let alone one that doesn’t have associated gripes. But, things like the dose clicker in particular are easy wins that some more focus on the psychological needs of the patient, not just the physiological, can achieve.

Oh, and I never did find out for sure why OTC inhalers seemed to have vaporized. Anyone have any idea?

In my opinion they made only one critical error in judgment. This error will haunt them for the rest of their lives – mostly because I refuse to stop bringing it up. Sometime in the late 1970′s, my parents gave me a massive set of LOC BLOCs. What? You have no idea what a LOC BLOC is? Nobody did. All the cool kids had Legos. I was a Pariah.

Warning: these are not Legos

My father cited that the LOC BLOC set was ‘close enough.’ My mother noted that we got a lot more LOC BLOCs than Legos for the money. That’s all fine, but as the Wikipedia article states:

Incompatible

The blocks were of a very similar grid pattern to the Lego system, but not quite. The knobs were too tall and spaced just a little bit off. The result was that children who owned the much cheaper LOC BLOCs were often excluded from group play with other children who owned the overpriced, but more popular, Lego bricks. – (wikipedia, 2/12/2009)

Anyway, where am I going with this? Well, the feeling I had when I first tried to attach my LOC BLOC rocket ship to my neighbor’s Lego space station is the same feeling of disappointment that I experience when I encounter app-like interfaces on the web only to learn that they only look like apps – they don’t act like them. For example, behold the “Pictures of the Week” interface on time.com:

Time.com's deceptive interface

When I encounter what looks like a photo flipper with dedicated “Back” and “Next” buttons – I expect the items within the frame to update when I click a button. In the example above, the entire page refreshes. This was acceptable in 1998, but today, clicking on that “Next” button and watching the whole page refresh just feels like someone went for the cheaper, “close enough” version. Why pretend to be something you’re not? At the very least, simple text links outside of the image frame might have deterred me from thinking I would be interacting with a self-contained app.

I understand that there are business reasons for not wanting an actual image flipper on your site. Page refreshes provide those much-desired view counts advertisers want. Just don’t trick me into thinking I won’t have to sit through an entire refresh when I click your button. Apps are becoming more and more ubiquitous, and even casual users now have expectations regarding the style and speed with which they react. Be what you are, not a wolf in sheep’s clothing. Or, a LOC BLOC in the Lego pile.

UPDATE: there is hope out there. Some traditional print media outlets are exploring actual app behaviors in their interfaces – such as the New York Times’ Article Skimmer Prototype. Now, if I can only resist the temptation to rob a toy store for a set of Legos, maybe I’ll make my Mom proud and run for public office.