Recently, the amount of “impressions” we get in Google search results has dropped. It didn’t happen on the exact date of the infamous “Panda” update, but the symptoms were the same.

So why did Google all of a sudden decide our pages were less relevant? I’m not sure, but one thing people keep bringing up in relation to Panda is duplicate content. We have a lot of duplicate content.  And Google hates duplicate content.

So why do we have so much of it?  Well, I promise it’s not to scam Google. I Maybe I’m kind of old-fashioned, but I feel like the URL of a web page ought to mirror the file system in which the web pages live. Or at least pretend to.

Let me explain. Take our page for everyone’s favorite movie, The Big Lebowski. There are two paths to this movie through our site: Movies -> Crime and Movies -> Comedy.:

And depending on how you arrived there, the URL in your browser will be either…

http://www.commontastes.com/4_Entertainment/6_Movies/10_Crime/4594_The_Big_Lebowski

…or…

http://www.commontastes.com/4_Entertainment/6_Movies/9_Comedy/4594_The_Big_Lebowski

Why two URLs? Well back in the day, one could navigate to the “next level up” by simply removing the last segment of the URL. Someone who saw the first URL and wanted to see more crime movies could simple change it to:

http://www.commontastes.com/4_Entertainment/6_Movies/10_Crime

…and see our crime movies. And do the same with the second URL to see our comedies.

OK, I admit that few people out there are nerdy and tech-savvy enough to try that, but it’s something I really wanted to have on our site, if for no other reason than to stroke my OCD.

In the eyes of Google though, this is the dreaded duplicate content. Google sees it as two separate web pages with identical content, and red flags go up.

And it’s similar for things that appear on our other websites. Take our beer ratings and recommendations site Barley Buddy. Our beers are categorized by region and by type. So the page for Anchor Porter is avilable here:

http://www.barleybuddy.com/89838_By_Type/89953_Porter__Stout/453220_Anchor_Porter

…and here:

http://www.barleybuddy.com/89837_By_Region/463095_United_States/453220_Anchor_Porter

Confounding that even further, is that Common Tastes, as our “umbrella site,” contains everything the spin-offs have, too:

http://www.commontastes.com/75_Food__Drink/76_Drinks/79_Beer/89838_By_Type/89953_Porter__Stout/453220_Anchor_Porter

http://www.commontastes.com/75_Food__Drink/76_Drinks/79_Beer/89837_By_Region/463095_United_States/453220_Anchor_Porter

That, my friends, is four URLs. Not to try and scam Google, but just our attempt to provide a logical and consistent site structure.

Doesn’t help us with Google though, does it?

Canonical URLs to the Rescue?

Google, in their Webmaster Help, recognizes that there are valid situations where you may have duplicate content on your site. (In fact, one example they use is a site-hierarchy with multiple paths, just like we have.) Their advice in this case, however, isn’t the usual “don’t worry about Google, just try and build a great site!” Here, they recommend using canonical URLs.

The canonical URL is the, let’s say, master URL for a page. In the case of The Big Lebowski, one of those two URLs has to take precedence over the other, and Google will always send users there.

We’ve implemented it now, throughout all of our sites. If you go to:

http://www.commontastes.com/4_Entertainment/6_Movies/9_Comedy/4594_The_Big_Lebowski

…and view the page source, you’ll see a line in the header like this:

<link rel="canonical" href="http://www.commontastes.com/4_Entertainment/6_Movies/10_Crime/4594_The_Big_Lebowski" />

That lets Google know that this is not a duplicate but in fact the same page.

Similarly, viewing the source on:

http://www.barleybuddy.com/89837_By_Region/463095_United_States/453220_Anchor_Porter

…will show you this:

<link rel="canonical" href="http://www.barleybuddy.com/89838_By_Type/89953_Porter__Stout/453220_Anchor_Porter" />

It also works between our sites, too. Viewing the source on either of the Anchor Porter URL’s on Common Tastes will show that same canonical URL above.

Will this fix our problem with Google? Will the GoogleBot figure out that we’re not trying to scam anyone here, man? Only time will tell.

I really liked Jason Cohen’s blog post a while back On the (un?)importance of design. It’s a fascinating read and turned the seemingly-obvious question “is a better site design really better?” into a very interesting debate in the comments.

We recently did some design re-vamping of our own, on Common Tastes and our beer recommendation engine Barley Buddy. What’d we do?

Common Tastes

One piece of feedback we heard over and over about Common Tastes was that its color scheme was “dreary,” “depressing,” “like a rainy day inside my computer.”^[1] We certainly didn’t feel that way, but on the other hand, it’s hard to view something with fresh eyes when you stare at it for hours each day.

So we played around with increasing the amount of visible whitespace, and in the process realized just how obsessed we were before about colored backgrounds – boxes with pastel blue backgrounds; headings with grey backgrounds; comments with alternating backgrounds… We took the axe to a lot of it, and just kept going until we eventually chipped it away to a much cleaner, more usable interface.

Barley Buddy

If people were, shall we say, direct about Common Taste’s color scheme, they were downright brutal about Barley Buddy’s. Not only the colors, but a lot of people told us that it wasn’t clear, at first glance, where you’re supposed to look or what you’re supposed to do.

We took the same approach, removing backgrounds, borders and unnecessary headers, and just look at the difference:

When Everything Competes For Your Attention

I spent 10 years as a web developer for private companies, and every time a product manager came through my door to get something added to the home page, they wanted it bold. Or brighter. Or larger. Or in a Johnson box. They wanted their new and exciting feature to stand out and get noticed, until eventually just about everything on the home page was competing for your attention. It didn’t look good. It made for a site where everything competed for the user’s attention, and resulted paradoxically in nothing actually standing out.

Despite seeing this first-hand for 10 years, somehow we let it happen on our own site.

So Did It Make a Difference?

So where do I weigh in on the above-mentioned (un)importance of design? Like Jason Coen, we haven’t seen it make a difference in traffic and conversions. But we have received a lot of compliments, and hope to hear a lot fewer testers say they wouldn’t trust a site that looks this unprofessional.

I think, in our case, it does matter, because we have in an application. When it comes to us and design, it’s more than just “Does blue look good with green?” or “Does this font say ‘Beer!’ to you?” It’s “Will the user’s attention be drawn to the right place to get started?” It’s “Can he tell the difference between the navigation and the main content? Will he find the ‘Signup’ link?” And I think what we’ve done has gone a long way toward moving us in the right direction.

Do you have any feedback on our design, either of Common Tastes or Barley Buddy?

Well we didn’t win the RISE fast pitch competition, and through proud and pleased we were to have been finalists it is of course now time to ask ourselves: How could it have been better?

Today I came across 5 Reasons Your Idea Pitch Sucks by Mike Figliuolo. While I don’t think our pitch sucked exactly, when I read his first point…

You Don’t Have a Compelling Problem. People pay to have problems solved.  I don’t care if the problem is boredom, inefficiency, poor decision making, or not being able to find your keys – problems sell.

…my heart sank. Because, for a while now, I’ve had the nagging feeling that our problem, as we’ve been presenting it, is not compelling.

So what is the problem we’re solving? Thus far we’ve stated it as “bridging the gap between all the people out there with opinions to share and the 70% of consumers who trust online opinions” along with some cited studies about how many of those people exist.

But that’s not really it, is it? Suggesting that there are millions of people all sitting at their computer, money clenched tightly in their fists, ready to buy something if only they could find some more opinions on it? That isn’t really the problem we’re solving. (Or a problem at all?)

Mike’s Figliuolo suggestion that inefficiency is a problem worth solving brings it more into alignment. While it’s true that there are lots of people out there with opinions to share and lots of people out there who trust online opinions, the real problem is that there really is no streamlined, non-tedious way to get from point A (the opinions) to point B (your purchasing decision).

That’s still a pretty inelegant statement of our problem, though.  Lucky for us, Mark Suster summed it up perfectly in his blog post Make the Internet Smarter at Helping Us:^[1]

The Internet has a problem.  It’s one you know well … there is so much information out there now it’s hard to separate the good stuff from the rest … I guess I sort of want “Pandora for everything else.” Why can’t I have it?

Well, Mark, good news: You can have it, because we’re building it. Already you can try our beer recommendation site Barley Buddy, or Gluten Freer, our recommendation engine for gluten-free products, or one of our other subject-specific sites.

Where we differ ever so slightly with the good Mr. Suster is on how the recommendations should be “sliced” (to use the term he coined and we’ll be stealing from now on). He suggested three different methods:

1. By social graph
2. By influencer graph
3. By people like him

For reasons I’d be happy to elaborate on in the comments, we’re not focusing on the first two.  But the third one: that’s the ticket.

Luckily, Mark Suster mostly agrees, saying about the last one that “mostly I want my recommendations sliced by random people I don’t know,” and his reasons are spot on with ours.

So thanks to Messrs. Figliuolo and Suster for helping us spot the problem with our problem, and bring it into focus. We’re not just “bridging the gap” between the masses with opinions and the masses who are buying.  That’s already been done.  We’re building a better bridge, one that only lets the opinions from people with common tastes to your own pass, packaged as recommendations. More signal, less noise.

So what was the original spark that gave life to Common Tastes?  The moment that started it all?  The original problem what got that itch a-itchin’ and so badly in need of scratchin’?

So glad you asked!

A few years back, I was living in Europe and realized there’d probably never be a better (or cheaper) time to learn about wine. Being in Germany, I’d already, shall way say, “studied” the local beer offerings, intensely and first-hand,^[1] but not wanting to take the same brute force approach to wine, I got on the Google to find a site where I could say “I liked this wine, didn’t like this one, and thought this was so-so.”

Being uninterested in expert opinions, or which wines were the most highly-rated by “the masses” (who are notorious for having terrible taste), I wanted a site to compare my tastes against a massive database of other users’ wine ratings, identify the people with tastes similar to my own (let’s call them my “common tasters”), and recommend me a few wines that I haven’t tried but my common tasters enjoyed.

I couldn’t find this website.^[2]  It was obvious what needed to be done.

I mulled the idea over with my pal (and now co-founder) Steve and we got to thinking: Why not include cheese? Wine lovers like cheese, and some interesting taste-crossovers could emerge there! What about cigars? But some people don’t like cigars. What about cigars and spirits?

What about, instead of a website, a platform that lets us spin off websites, slicing and dicing it up by any subject we please, all pulling from their own cross-section of one big database?

Now that sounded like a great product.

I’ll be pitching at the RISE Fast Pitch competition in Austin tomorrow (June 28th).

If you’re in the area, come by and see me and give me feedback!

Wish me luck!

So last week I posted to Reddit asking for feedback on Barley Buddy and the response was way more than I’d hoped for.  We got so much great feedback on the site!  (And our system got a lot smarter due to all the beer ratings that Redditors added!)

The top remarks we heard were:

1. We need a mobile app. Agreed, we do! Just imagine with a site like Barley Buddy, being down at the bar or the supermarket to see what you oughta try, or scanning a UPC code to see what the community says, or how many stars we predict you’d give this beer.
2. The design is ugly. Sorry about that, guys. We’re primarily programmers. We’ll pretty it up just as soon as possible.
3. Implement a Pandora-like interface to get new users started quickly. Not only did people complain that the home page left them a little stranded, unsure of where to go, they also suggested this as a remedy.  And this is a great idea to quickly and easily guide new users through their first handful of beer ratings so the system can get a feeling for their tastes ASAP.

There were some other great suggestions we’ll be implementing, but these are the ones that we heard the most often.

We can’t thank you enough for the help, Beerit.

So we just launched a bunch of new changes, and we’re really pleased.  IMO this raises our sites from their previous “functional” state to being just a hair within “1.0 status.”

So what’d we do?

• Profiles! That’s right, if you’re reviewing with any one of our websites, you can now set up a profile and link it to your Twitter, Facebook, or personal blog page! You can also use your pic from Twitter, Avatar, or Gravatar with us to show your happy, a-smilin’ face next to all your reviews!  Here’s my profile page.
• Re-designed home page. The old, bland design has been snazzed up with a little series of graphics showing the flow from “I’ve got an opinion” to “I’ve discovered something new!”
• Suggestions tacked onto the “Who’s Doing What.” When you hit the home page, it also shows the 10 most recent ratings, just like before, but now it also says “…and based on this rating, we think this user would like X.“ We hope that this, and the above change, will make it instantly clear on the home page what the site’s all about.  But just in case it doesn’t…
• Calls-to-action on the pages for individual items! If you land deep within the site somewhere (having arrived from Google, perhaps), there are now more prominent Johnson boxes telling you why you should enter a rating, and what you’ll get if you do! The boxes evolve from asking you to sign up, to asking you to rate something, to asking you to review it, depending where you are in the process.

How’s it working out?

So far feedback’s been positive on the design.  But more importantly, the number of signups and ratings from search engine traffic has gone up!  Thanks, call-to-action boxes!

Please check out Common Tastes and let us know what you think of the home page, profile page, and the page of something you might want to rate.  Any feedback is very much appreciated.

Wow.  I still can’t believe how much opportunity there is to network here.  I’ve met so many interesting people who are doing their own thing, and all in the first week!

I went to a bootstrap interactive presentation by Conversion Scientist Brian Massey (which was great), met some cool folks at Open Coffee, and made some new contacts at the Tech Ranch Camp Fire.

I had good expectations to begin with (I mean Austin’s a happenin’ startup town, ain’t it?), but just shooting the breeze and discussing our business with other entrepreneurs (and their businesses!) has proven way more fascinating (and useful!) than I even anticipated.

Can’t wait to hit some more!

Man oh man was that a long drive! And despite going through 4 states, it felt like half of the whole trip was Texas; only Texas. Things really are bigger here.

We are all so pumped to start working on Common Tastes — and all of its spin-offs — together, face-to-face. After having done it all remotely so far, I can only imagine how much faster things are going to move from here on!

First impressions on Austin: Great. On day 1 I attended a free presentation from Conversion Scientist Brian Massey, rubbed elbows with other entrepreneurs, and have four more events on my calendar before the weekend. That so much is going on here is very exciting.