To get to the new blog, simply visit www.maderalabs.com/blog.

To subscribe directly to the feed for the blog, use this link.

Thanks for everyone who drops by to read.  Sorry about the inconvenience of moving things.

I will be taking this blog down around the end of February, and redirecting the usabilitymatters.net URL to the Madera Labs blog at that time.  Thanks!

Here’s why I think this is.  Saying “I” allows those who you’re communicating with to immediately position you as a potential outlier.  It’s much easier to discount what you think…harder to discount what a group of people think.  See, when trying to convince someone of a point, the other party is often automatically defensive to your point.  Any potential gap in your logic means an opportunity for that person to refute your point.

The problem is, using “I” often creates those gaps.  At that point, it’s easy enough for the other party to simply refute you by saying “Well, you do it that way, but I do it this way”, resulting in a stalemate.  The key to this is another behavior that I think is common: people doubt themselves first.  By positioning them against a large, faceless crowd, it’s much harder for them to hold their side out of stubbornness.

Give it a try next time you’re working on persuading someone to your point.  Use “they” or “people” or “many groups”…whatever collective pronoun you wish.  Also, notice how much people use “I” in their daily conversations - you’ll see how it makes for a weak argument.

[I'm not advocating that you outright lie about what a group of people think or how a group acts.  This argument is made on the assumption that you've come to your side of the argument through research and careful examination.  Saying "Well, most people think the sky is brown" isn't going to work.]

Lately, as more websites offer users the ability to rate content (books, recipes, hotels, etc.), sorting and narrowing a set of search results by rating is becoming an increasingly popular method employed by users.  It makes sense, as it harnesses the crowdsourcing nature of the web and allows users to cull out the products that the community has decided are good.

However, how these ratings are sorted is handled differently from website to website.  Recently, I had been searching for cookbooks on Amazon. I sorted by rating to cull out what I thought were books with the best community reputation.  Below is a screenshot of a couple of these results:

Notice this:  these products are shown in order of decreasing rating, and the book with one 5-star rating is positioned above the book with 159 ratings, which have resulted in a 4.5 star average.

Now, technically, this is a correct way of sorting.  Yes, 5 stars are better than 4.5.  However, does one 5-star rating give as accurate a quality indication as an average of 4.5 stars over 159 reviews?  Probably not.  The 5-star book could be given a 1-star rating by the next user, resulting in it dropping to a much lower average.  This kind of volatility means that the ability for the user to make a culling decision based strictly off rating (and many users do use one factor like this to get a list down to a manageable size) is unreliable.

So, what’s the solution? I think it’s a calculated score of some sort.  It may be as simple as total number of reviews/average score (159 reviews/4.5 stars).  I’m not sure if this holds up on all ends of the scale - I’ll let the math contingent chime in with their suggestions for a better system.

As a final note, the recipe site Recipezaar does appear to use some sort of calculated method for sorting by rating.  Note the following results:

Aha.  This time, the result with a slightly lower star rating, but more total reviews, is ranked higher than the result with a higher star rating over fewer reviews.

What do you think?  Is Recipezaar onto something?  Is Amazon wrong?

So, I was really excited when my wife (@molliebrad) brought me the following photos she snapped at a gas pump the other day.

As she was explaining the story, she said that after she had put her card in the pump, she kept pressing the button for credit, only to have nothing happen.  Here’s a picture of the keypad in mention:

As she was standing there, she says, she kept pressing the “1″ button, trying to get the machine to accept her input as selecting “Credit here”.  The number 1 is obviously labelled as such, but pressing the button didn’t seem to do anything.

The problem, as she figured out after a few moments, is that “Credit here” isn’t a label - it’s a button.  What?  Sure doesn’t look like one.  Evidently, she’s not the only one who’s made this mistake.  In fact, so many have, the gas station has affixed the following note to the top of the keypad:

Aha!  Evidence of a terrible interface design.  The ever-so-telling instructions on how to correctly use the interface.  Sad.

So, what can we learn here?  There are a least a couple things wrong:

• Poor use of affordances -  On the keypad, the numbers are clearly buttons.  They are raised and beveled, and look like they can be pressed.  However, what about the buttons around the edges?  The manufacturer, probably due to aesthetic considerations, has removed all the button-ness from the buttons!  Lesson:  let objects read like they want to read.  Buttons should look, feel and act like something that can be pressed, which involves visual cues like raised edges, bevels, shadows and sometimes texture.
  
• If you need to provide instructions on how to use an interface, something is probably wrong.  Here, there’s a note added after the fact that is attempting to prevent people from making a mistake.  The interface, however, should provide all the clues necessary through it’s physical design, not through instruction.  As my wife has proved, user’s don’t read instructions!  Despite this note hanging there, it took her several frustrating moments of failing to finally realize that there were instructions above the keypad.  All users of interfaces, on the web or in the real world, do the same thing - we try first, and read later.

If you ever find yourself feeling like an idiot because you can’t figure out an interface, rest assured, it’s not you…it’s them.

Recently, I spoke at BarCamp Nashville.  My talk was about some usability design principles that I gleaned from Don Norman’s book The Design of Everyday Things, and how I think these concepts from the physical usability world can be transferred over to the web.  In the talk, I covered three main areas:  affordances, visibility and feedback.  There are a ton more great ideas from this book, but alas, 25 minutes isn’t a bunch of time.

If you’re interested in checking out the presentation, it’s posted on SlideShare here.

Here’s a great blog post from Will Sloan, where he takes some points I covered and expands on them (thanks Will!)

And, here’s the original presentation page on the BarCamp Nashville site.

In addition, I gave a condensed talk about affordances at Nashville Ignite, a really fun presentation event where presenters are given 5 minutes to talk, with your slide deck automatically advancing every 15 seconds.  You can watch that here.

Thanks to everyone who came out, I really appreciate it!

If I tell you that your website gets 3,200 hits per month, is that good or bad?  How do you know?  You’ve got the data - what’s the problem?

Often, data is mistaken as information - a leap in logic that results in improperly drawn conclusions and frustration with the data collection itself.  However, the data’s not to blame - it’s the failure to translate that data into information that’s the culprit.

We love numbers.  We want analytics on our website so we can see how many hits we get.  The big letdown comes when we realize there’s no context to consider the data inside of - nothing at all to translate data into useful information.  This is why strategy is so important (and so hard).  Strategy and goals are what provide context around data and make it useful.

Two areas I’ve seen this happen often in are web analytics and surveys.  Web analytics, as mentioned earlier, often end up useless because the owner of the site never thought about what data would be important and what benchmarks and goals the numbers should hit.  Is 3,200 visitors to your site good? Sure it is, if your strategy is to get a 10% market penetration into pygmy water buffalo farmers.  If you know there are only 10,000 of these people in the world, 3,200 visitors is damn good.  If, on the other hand, you’re Chevrolet, it’s probably not.  (Also, don’t get caught up on visitor count.  Visitors don’t mean much, conversions and user actions are where the paydirt is.  Think critically about what data you need to evaluate to assure that you’re meeting some sort of user engagement and conversion benchmarks.  10,000 people walking into a store everyday and buying nothing makes for a short-lived business).

Surveys are just as bad.  I’ve seen countless surveys conducted that result in useless data, because the survey creator never considered what information they were looking to extrapolate out of the data.  Every question you ask in a survey should result in you taking some action based on the information pulled out of the data.  Asking questions like gender or race are probably useless (unless your product’s success depends on you targeting a group of customers based heavily on this criteria - for most businesses, however, this isn’t the case).  Think about what problems you have, what information you need to make a decision, and design a way to collect the right kind of data that will allow you to glean that information.  It’s shocking how many people don’t do this.

When it comes down to it, information is just data with context.  Context is almost always determined by strategy, which means deliberate and thoughtful planning.  Be careful not to abuse the glut of numbers you have access to.

Guess what?  People are saying the same thing about your business.

When times are going well, your business smells great.  Honest relationships with customers, selling to them because you truly want to offer them something to improve their life.

When things start to go badly, you start to stink.  You start to sell things just because you need the cash.  You start to think of your customers as a homogeneous group that you can drill for money.  You start to have the stench of desperation.

The ironic, and damning, thing about this is that customers can smell it.  You think you’re offering “deals of a lifetime”…they smell the stench.

What does this look like?  Here are a few examples:

• Spamming your email list with offers and discounts
• Scraping websites for email addresses or buying lists
• Telemarketing
• Putting “Buy”, ‘Subscribe” or “Purchase” links in every conceivable corner of your website.
• Saying things like “If we can just get more subscribers, then an average of 5% will convert, and we’ll make $X.XX”

You get the idea.  Look, when things go south, it’s easy to react and start resorting to these kind of measures.  The problem is, your customers can smell the desparation on you.  No one wants to go home with that guy at the end of the bar.

All companies do things wrong.  However, we often look at successful companies and draw the conclusion that all of their tactical moves are inherently correct.  They’re doing well, so it must be because all of the things they’re doing are correct. So, we do what they’re doing.  Problem: that doesn’t always work.

If you’re going to correlate cause and effect, do so in a narrow and restrained way. If you’re flailing about, trying to prove a point that you’re insecure about, it’s easy to jump to wrong conclusions.

Case in point:  A marketer was hell bent on placing large, intrusive subscription ads on their website.  In an effort to justify this move, they cited successful companies that are doing the same thing.  The correlation they made:  company X is successful, and it must be because of the ads on their website.  The problem: these things might have absoutely nothing to do with each other.  Without the right kind of data to back it up (and this particular person didn’t have that data - just a far-fetched correlation), wrong conclusions are drawn. When you make actions based on those wrong conclusion, you’re in serious danger.

Be careful with correlation - make sure that the result you’re seeing is really due to the action you’re proposing it’s caused by.  Don’t use incorrectly drawn conclusions to support your poorly researched ideas.

Companies do this all the time.  ”Our customers just don’t understand how great this is for them”, “We’re ahead of the times, they’ll catch up”, “Why aren’t people buying this stuff?!  They need to know how awesome it is!”.

The problem is with control.  As a company, you’re not in control.  You don’t have any bargaining power when it comes to the market.  If the market decides not to give you money, you’re dead.  Too often, we try to change the market, instead of really finding out what the market wants/needs.  Often, this is due to being overly precious about our products.

If people aren’t buying your stuff, it’s because you’re wrong, not the market.

Far too many companies and marketers try to force consumers into the conversation that they, the company, wants to have.  The problem is, consumers are self-centered.  They don’t want to talk about what you’re talking about.  They want to have the conversation they’re already having.

This is lazy marketing.  Instead of putting time and energy into honestly listening to others and truly caring about their conversation, it’s much easier to shout at them and hope they talk back.

The key, as made clear in the seminal book on relationships, How to Win Friends and Influence People byDale Carnegie, is to take a geniune interest in other people, not to assume you can get them to care about you.  It takes work, and it takes restraint, but it’s the only way to be successful.

So, stop blasting out your messages and hoping people care.  They don’t.  Find out what they’re talking about and join in the conversation.  As soon as you start to try to turn the conversation to your agenda, however, you fail.  Be real, be genuine, and you’ll be trusted.  When you’re trusted, you win.