How do they know? Well, first of all, it’s pretty clear if you just look at them. It’s almost impossible to figure out what they are, to say nothing of how on earth to use them. Also, their current customers keep coming back to them asking ‘why can’t I do X with your product?’ Hence, they called us (and when I say us, I’m talking about Fell Swoop, dontcha know).

So there we were, staring up at the projection screen, at a seriously sad and confusing UI. And the Acme exec asked “how would you help us avoid this?” I thought for a minute. I thought about all the fancy stuff I know about user experience and customer-centricity and testing and metrics and wireframes. And then I realized that the answer was much more basic:

“Well, first of all, we’d want to figure out what question this thing is supposed to answer.”

Guess what? That simple sentence became the foundation for the rest of the meeting. Back to basics indeed…but basic waters run deep. In order to figure out what question something is supposed to answer, you have to figure out:

1. Who’s asking (personas)
2. How they are phrasing the question (and whether that’s even the right question to ask)
3. What answers they expect (how the personas think, vs how you think)
4. How your company and products can answer that question in a truly fabulous and unique way (business goals, value propositions, and differentiators),
5. What you want your customers to do in addition to getting the question answered (like buy more question-answering-devices)
6. Then, and only then, how to translate all this into a UI that actually works.

I think the problem is that most people start talking at #6. And that’s not where the answer is. It’s not about moving text and boxes and buttons around.

Long story short, apparently, after our meeting, the Acme folks went around muttering ‘what question is it supposed to answer? What question is it supposed to answer?’ and there have already been some pretty great results. For example, they told us that they asked a customer what he thought a particular product was supposed to answer. Guess what. The customer thought the product was supposed to answer question X, and it was actually built to answer question Y. Horrifying, yet also magical, because now that Acme can see the problem, they can fix it.

So how does this have anything to do with me? Well, here's what happened, according to my peeps at the shelter:

"I asked them how they picked us to be one of the 15 shelters – out of the thousands of shelters in the United States – and the producer told me he was searching the internet for the humane societies and SPCAs in major cities and we came up at the top of one of his searches!"

Hi, how's THAT for ROI. "We redesigned the site, and Oprah called."

Here's the before shot:

And, of course, the after. Please note, I helped them. I didn't do the redesign. There are still plenty of opportunities here.

You get to see me in my Haute Trash outfit, made entirely of ethernet and phone cords, and learn all about my little...proclivities. For example, I've managed to out myself as someone interested in Rabbit Agility.

The lops kill me. I mean really.

Carol Bartz is the new CEO of Yahoo and she held her first earnings call recently. And, bless her heart, she dropped the F-bomb...while talking about things near and dear to my heart. Here's the quote from the Techcrunch article by Erick Shonfeld:

.... Later on she did manage to drop the F-word, though (but quickly apologized for the slip). The exact quote, said at the end of the Q&A in frustration over all the layers of extra management at Yahoo, was:

'There were engineers in almost every country, and way too many product people. We had one product management person for every three engineers. We had a lot of people telling engineers what to do but nobody fucking doing anything. Excuse me. I knew that would slip out one of these times.'

And that was just her second conference call as Yahoo CEO.

Carol Bartz, CEO of Yahoo. Bless her *#*$%! heart.

I love this quote. And I love how much this situation irks her, given that it is a totally irksome situation. Being an engineer must suck so often: they get all these people telling them to do different things, changing their minds, playing politics, being bossy...it must be like being an electrician hired to update the wiring at the Brady house and being supervised by all six kids. And the dog. And Alice.

And then they get to build stuff they know isn't as good as it could be. And then they get blamed.

Now here's where I take issue with Mz. Bartz. This snippet is from a CNET article entitled Bartz lights fire under Yahoo engineers:

We have good engineers but have to hire more and get them focused on the right stuff. It's probably the most important thing Yahoo's going to do to really become a big strong growing international company," Bartz said during a conference call to discuss the company's lackluster first-quarter results.

Ummm...hold on a sec. That's like solving the problem at the Brady residence by sending in more electricians. If I had 10 seconds with Carol in an elevator, I'd say "Love ya mean it. BUT you best figure out the 'get them focused' part before you hire a bunch more engineers. Don't do the Brady electrician thing."

At which point she would be likely to call security, but I betcha she'd wake up in the middle of the night thinking I was totally right.

You know, it seems to me that engineers should be the closest allies to user-centered design people. Oh, and executives should keep one of us around at all times. Basically, we're trying to help everyone get out from under the whims of Peter and Cindy and help them find the time, clarity, and focus to build kick-ass software.

They key is to find the common language between UX people and engineers (and execs). And help them see that the 'soft' stuff we do is actually really truly helpful when it comes to untangling confusion -- and that we can be the translators between engineer-ese and executive-ese. But guess what...it's our job to help them understand that we can help. Not theirs to just 'get it.'

I think this is one of the reasons I was so psyched to see the MSDN article! By engineers, for engineers...and about the value of basic UX methods. Gotta love 'em.

And gotta love Bartz. I'd rather be led any day of the week by someone who gets frustrated enough at stupidness in a corporation to drop the F-bomb than a haircut in a suit. We'll see what she does with Yahoo. I don't envy her the job of untangling a company that gigantic.

Well, actually I kinda do. How fun would it be to try something like that? If being an exec didn't require so many spreadsheets I might actually be interested enough to try it.

You may begin.

Check it out:

We will manually place you and leaders under you from around the world. We will also place your personals with the volume under them. We will provide a system API for you. I am sponsored by the master distributor.

I will be blunt and real. It will take 6 dollars and 95 cent that locks you in in before the masses. Your time is limited, you must to decide NOW to stay on TOP!

We've just finished the system and it is now ready to start locking you and your people in before everyone else gets it. Just lock your special spot now. Get all in fast ... 25% personal match plus 7 other streams.

You will instantly have your personal replicated site and can start using it and locking in your people.

I wish I was sponsored by a master distributor.

And, come to think of it, the idea of 25% personal match plus 7 other streams does sound desirable and significantly better than other offers I've gotten.

And there are days when I want to lock in my people and place leaders under me.

Hmmm....

Well, for ME it's cool anyways.

And the authors, Dr. Charles B. Kreitzberg and Ambrose Little, even writing for tech guys who tend to be very data-driven, even approve of quick-and-dirty personas. Kreitzberg, who is the CEO of Cognetics (a company that does usability and ux consulting) wrote the first part of the article, which included this delicious nugget:

Personas do not need to be complex to be useful. I typically begin by creating brief outlines of personas based on conversations with people who know the audience well, such as salespeople or customer service staff. I call these personas "assumptive personas" because they are not based on actual data.

We used the term 'assumption personas' in the Persona Lifecycle book too...now I like to use "Ad Hoc Personas" (which Don Norman used in his Personas & Empathetic Focus article) because really, they aren't based so much on 'assumptions' as they are on embedded knowledge that exists in an organization.

Ambrose Little, who is a software developer, wrote the second half of the article...and I'm even more fascinated by what he has to say. I've of course thought a lot about personas from the User Experience perspective, which Kreitzberg does a wonderful job of articulating. But hearing it from a dev dude is fascinating to me, because for years one of the challenges has been to help devs understand that even a tool this 'soft' can be incredibly helpful to them.

I love this quote:

...if you find yourself in trouble, which happened to me recently, where you have tons of prickly issues poking out of your design, it may just be that you need to zoom out and attack the problem anew from your persona's perspective. Sometimes it's hard to break old ways of thinking about design, and using personas can help you reformulate your approach. In my case, doing this helped me to see that we needed to somewhat dramatically alter our approach in terms of presenting essentially the same functionality to our target audience.

There's still a sort of slightly antagonistic thing that can happen between UX people and developers, which is interesting to me as someone who is a student of intra- and inter-team communication and the problem of focus in companies. The deal is, I think, that everybody is just trying to get their work done. Especially devs, who have typically been yanked around by ever-changing, random-seeming specs for years. They, after all, are the ones who somehow have to translate the chaos into a product that works and sells. Not so easy.

But when it comes down to it, everyone wants the same thing. Cool software that people like, and use. Devs don't want to create crap. They want people to use their stuff. And I like that Little is saying 'hey, this can help.' That's really what it all comes down to.

He also talks about 'dictatorial specifications,' which is a phrase I'm now officially stealing because I love it. Reminds me of Avinash Kaushik's HiPPOs (Highest Paid Person in the Organization:

...even in cases where the team members are committed to human-centered design, personas can help explain what might otherwise appear to be dictatorial specifications. I've seen this happen more than once, where a design specification was questioned because it didn't make sense to some people on the team. They were looking at it from their own, seemingly intuitive perspective, but the solution wasn't being built for them. The personas provide a known, concrete reference point to explain why the spec is the way it is, which takes the focus of the discussion away from what makes sense to individual team members (or what they prefer) to what makes sense for the people you are building for.

The most astute comment in the article is, of course, this one:

An excellent resource to help you understand personas and their construction is the book The Persona Lifecycle: Keeping People in Mind Throughout Product Design by John Pruitt and Tamara Adlin (Morgan Kaufmann, 2006).

All quotes are from here.

Here’s the deal. In most companies, people are confused. Especially the top brass. Why? Because they go to a zillion meetings, and during each meeting either one thing is decided or someone is assigned to figure out how to decide something. Which means every day, every honcho is both making decisions and getting decisions thrust upon them. Lots and lots and lots of decisions.

And no one is writing them down.

So, at some meeting a long long time ago, all the execs might have all agreed on a set of business objectives, brand objectives, and target markets. They may have even created some kind of document. And sure, they all still have meetings together, and each of them assumes that everyone else is crystal-clear on all those goals--and on how all the goals have been tweaked or changed as a result of the zillions of little decisions.

Wrong.

I bet no one at your company actually knows exactly what your goals are.

What I find is that no one is super-clear on exactly what the brand, business, and customer-experience goals of their company are. And here’s how I know that: every single time I start a new contract, I ask. I ask for a plain and simple list of the top 3-5 business, brand, and customer experience goals. I bat my eyes and look all innocent and ask. And no one can produce them--at least not an ‘official, up-to-date’ set of them.

Why not? Because it feels like career suicide to raise your hand in a meeting and say ‘err, ummm, I’m not exactly clear what our goals are.’ Especially if you are a honcho.

If the leaders aren’t clear, how on earth are the rest of the people in the company supposed to have a focused understanding of what they are supposed to be doing, and why they are supposed to be doing it?

So here’s my assignment, should you choose to accept it.

Send out an email asking key people to give you the following information (for the company as a whole or for a particular project, if that’s more appropriate). If they want to be anonymous, let them do that (they can leave a printout on your desk, for example).

Ask them to answer these questions:

• What are our top 3-5 business objectives? These goals are usually expressed in measurable terms, like "increase signup by 20%."
• What are our top 3-5 brand objectives? How do you describe your brand today, and how do you want to advance, change, or evolve this to be different six months from now?
• What are our top 3-5 customer experience goals? These are statements that you’d love to hear from your customers after this project, like "wow! This product/site/service is really great because…" or "I’m going to tell my friends all about…."
• And while you’re at it, what are our key value propositions and differentiators? (And make them actually valuable! and different!)

Then put all this info together and help someone important understand that the variety in the answers is an issue that needs to be solved. (And for heavens’ sakes, don’t send the answers to me. But do send me a note and tell me how it all went!)

I think you’ll be amazed. (And maybe fired too, but probably not.)

It's almost the day we've been planning since June
It's scary! It's here! It's almost too soon!
It's been weeks since new site tweaks, we're all feeling sober,
Site code has been frozen since early October.

Budgets have flown off like southern-bound birds;
Marketing blew big bucks on Google adwords.
On PayPal! on EChecks! on Rotating Credit!
On digital cash! online wallet! on cards that are Debit!

We pray "O today, let economic crisis be averted!
Help us to get lots of sessions converted!'
Let's go, online traffic! Mirrored servers, get mounting!
Our metrics are waiting for clicks to be counting!

We're watching! We're counting projected sales versus actual!
and assessing right now if SEO was all natural!
New shoppers arrive--on promos to pounce!
But, do they have pause when creating accounts?

Are savings enough? Did their nerves you assuage?
Or do they just bounce right off your home page?
Are you safe and secure? Will new customers have a conniption...
if they find out they're shopping sans SSL encryption?

Do you have enough AJAX? And commerce that's social?
What about in-store pickup at a location that's local?
Are you lucky? Are workers each other a'tripping,
as they scramble past robots whilst pack-pick-and-shipping?

Are fulfillment centers buzzing in overtime glee
as you get out them gifties--ontime under trees?
And what of next year? What new things are coming?
To keep your business in recession still humming?

What happens next time in things click-and-mortar,
when next we all face our first thru fourth quarters?
We'll be dashing and prancing to start-of-year conferences
to talk about new ideas on personalizing preferences.

We'll watch and pretend it's much more than jive
when a panel convenes on 'web-two-point-five.'
Don't get too anxious! Too scared all will fail!
Don't cheapen your brands with huge clearance sales!

Seriously, are we behaving like a whole bunch of rookies,
Making poor Santa ask if HE can leave cookies?
Treat all shoppers nice, for it is just decency
that is the thing that will jump visit frequency.

Great customer experience is all in the details...
else they'll be dying to opt out of emails.
It's not about platforms, or templates, or what's new
it's about doing for customers what others won't d.

The tiniest customer will loudly agree:
Be nice, make me happy, and back I will be!

My wish it is this: may your site bring you gold
May all of your holiday wishes end up 'bove the fold.

Think about this: you've actually gotten someone to convert at your site, fabulous-store.com. They bought something. They gave you their credit card and they are happily waiting for their purchase. Many of them get it and love it. Hooray!

But what about the customers whose purchase comes with a free white elephant gift: an interaction with your customer support system? For example, let's imagine Sarah is worried that her order from your fabulous-store.com isn't going to arrive by Christmas. She emails your reps. What happens?

Possibility 1: Cheerful robots

Sarah gets an auto-generated email saying an email server was happy to receive her message. The next day she gets an email from a 'human' who's bonus is based on how fast they can respond using pre-written blurbs carefully designed to answer a completely different question. Sarah responds, now in even more of a panic because it's one day closer to Christmas. The server is delighted to hear from her again. The next day she gets a different blurb. And guess how helpful it is.

Possibility 2: What kind of fruit falls off a phone tree?

Sarah calls your customer support center. She does the whole 'press one for orders originally made in sanskrit' thing. She listens to music designed to produce seizures in small children, interspersed with messages about promotions that repeat at least four times. She ends up with an operator who asks her for every bit of the same info she 'entered' by pushing the buttons. After five minutes, the rep tells her that he needs to transfer to another rep, who, you guessed it, asks for all that information yet again...and so on.

Possibility 3: Happiness is a warm rep

Sarah gets a response that was clearly written, or at least edited, by a human being. The human being makes sure that the package is expedited, even though the company now has to eat $15 for quick shipping. The cost of the support interaction is around $5. Sarah is so amazed and impressed that, when all of her friends start talking about their own shopping hell, she excitedly tells her story about the fabulous experience at fabulous-store.com. Hmm...wonder if that was worth the $20?

Before you talk about how expensive this would be, let me get this straight... You're spending how much money, exactly, trying to get customers to come to your site? And how many initiatives do you have to improve conversion? What about projects to try to get them to make a second purchase or increase average order volume?

And let me guess...you're trying to save money by increasing throughput in your customer service center? So basically to spend less time with each customer who has a problem, most of whom are among those you have converted?

So, Tamara, what am I supposed to do?
Go order two things from your site. Tomorrow, before they arrive, email customer support to ask where one of them is, and call customer support to ask about the other one.

How was it? Annoying enough to grumble about it to anyone who will listen? Surprisingly good enough to tell your friends?

The most recent interaction many of your converted customers have had with your site, your brand, and your products is a customer support experience. You got them, you converted them... don't piss them off.