Planaroo Blog

Go big or go home

I had breakfast with a friend and former eBay co-worker last week, and he asked me this simple question:

"What's so great about your site?"

Leave it to Leonard to ask an easy question that's exceptionally difficult to answer. 

I went through the list: lots of recently updated content, lots of photos, the ability to add things to your plan (coming in two weeks), etc.

He said, "Yeah, but what's so great that people won't have a choice but to use it?"

I was stumped.  What he was really saying is that a travel guide is just a travel guide... unless it isn't somehow.

In other words, Planaroo can't just be a travel guide.  It has to be the very best travel guide on the web, or there's no point in doing it.  It doesn't need to be the most complete (focusing on key areas is fine), but it has to be the best at the things it does cover.

So... I'm heading back to Las Vegas tomorrow morning at 6 a.m. for four days.  Steve and I got a lot of photos a few weeks ago, but we missed a ton of stuff.  Sometimes the stuff you missed is only obvious after the fact. 

This time I'm going to get more photos, better information, and lots of video.  I also commissioned some custom maps that will blow away anything else I've seen online. And photos will be available online in their original (super huge) sizes.

After all, isn't Vegas really about bigger, better, and brighter?

Keeping up with Las Vegas

I've been plugging away at the Las Vegas content for almost two weeks now.  Steve and I took over 1,000 photos on our trip last month, and now I'm writing the content that will go on the casino, restaurant, and show pages.  (And I still have to go back and add Universal and Sea World.) The research is taking longer than expected

Here's the situation:

• A surprising amount of information in printed 2008 guidebooks is out of date.
• Wikipedia cannot be trusted.  I've found about 5 examples where information was either outdated or just plain wrong. I've updated a couple of Wikipedia pages in the process, but this experience has given me serious concerns about Wikipedia content in general.
• Many travel web sites use static HTML without database content management.  It's very common to see the new Planet Hollywood referred to as the Aladdin (old name) one some pages, but not all.
• Google News Alerts are proving to be very helpful in keeping up on major topics.
• I believe Google is using a page's update date as a ranking factor, but all too often they're displaying seriously outdated web pages on pages 1 or 2 of search results if the site is big enough to have a high PageRank score.  There is a major Disney fan site that hasn't updated parking prices since 2004, but it performs very well on Google.
• More than ever I think there's a need for a web site like Planaroo that is constantly updated with solid information.

So, in the spirit of doing one or two things well instead of doing 10 badly, I've decided to focus Planaroo only on Orlando and Las Vegas for the foreseeable future.  My goal is to have the best travel web site available for those two cities before expanding further.  I have a lot of tweaking and organizing to do, but I feel good about the strategy.

I'd like to hear what you think about it.

Disney World raises ticket prices

Walt Disney World increased ticket prices, effective August 3, 2008. 

It's almost an annual event.  The Orlando Sentinel reports that price changes used to occur in January, but Disney, Universal, and Sea World are now changing prices in late summer to meet the publication deadlines for independent guidebooks. 

Thanks to the immediacy of the Internet we have the updated Walt Disney World ticket prices on Planaroo right now.

Disneyworld.com is showing the old prices on their (exceedingly difficult to find) ticket prices page, but it may be updated tomorrow when the changes take effect.  This gives you less than 24 hours to buy tickets at the old prices!

Planaroo at 30 (days, that is)

Planaroo.com launched in beta form 30 days ago today, and we're about to drop the "beta" designator.  I thought it might be interesting to share some of the things we've learned so far.

1. Traffic is growing organically through search
   The numbers are small because the site started from zero, but traffic is growing quickly -- about 10% a day on average.  Sundays and Mondays are heavy traffic days.  Fridays are dead. 
2. Google sends the most traffic, but Microsoft performs far better than anticipated
   80% of traffic to Planaroo from search engines comes from Google, and the other 20% comes from Microsoft's Live or MSN search.  Google has indexed 287 out of 831 pages in the sitemap, and almost all of the non-indexed pages are photo pages without content.  Microsoft has indexed 177 pages as of 8/1/08. seems to have indexed most of the content too, but they don't provide specific information on indexed page counts.
3. The original content strategy is working
   Creating so much content was difficult and time-consuming, but it seems to be paying off.  The site is doing unexpectedly well on search queries for terms like parking, directions, and ride height requirements. We're even on the first page of search results for some search terms.  I didn't expect that so quickly, especially with no links to the site.
4. Yahoo is missing...
   There is no traffic from Yahoo, despite the fact that their spider (named Slurp) has visited the site countless times.  Planaroo content pages are not showing up in Yahoo search results at all, so I assume that either their indexing is exceptionally slow compared to Google and Microsoft, or I've done something to piss them off.  (I have friends there... maybe one is secretly mad at me.) Hopefully this will resolve itself soon.
5. Over 64% of Planaroo.com traffic in July was on Firefox
   This striking number is almost certainly because of a blog post I wrote about Amazon's S3 service that created huge traffic numbers on July 21 (which I factor out in the overall growth rate).  The programmers who found that post on Hacker News tend to use Firefox on Macs, but that's still a huge number.  25% of visitors are using IE, and 9% are on Safari.
6. The UK is sending close to 20% of Planaroo's traffic
   The weak dollar makes an Orlando vacation much more affordable for our friends in the UK, and it shows in my traffic stats.
7. People are not visiting photo pages as often as expected
   This surprises me.  My friends and I spent a lot of time (not to mention money) taking photos for the site, but users are simply not clicking through to see the larger versions of the thumbnails or the gallery of images.  I'm not sure if that's because the thumbnails are big enough to get the point across, or if it's not clear that the images can be clicked, or if people are just more interested in the text.  It won't stop me from adding more photos, but it's something to keep a close eye on.

It will be very interesting to see how the Las Vegas content (coming very soon) changes things because it will be the first time that hotels and restaurants are part of the content mix.

Thanks to everyone who has helped Planaroo get to this point.  Special thanks to Jen, Steve, Pat, Charles, David, Caryn, Stella, my mom, and Art.

Hard at work on Universal, Sea World, and Las Vegas

Things may seem eerily quiet at the old Planaroo Ranch, but there's a lot of work going on behind the scenes to prepare for the launch of Universal Orlando, Sea World, and Las Vegas content.  Expect to see more a bit later this week.

Now available: Email to a Friend

Email to a Friend is now available on attraction information pages. This is the first feature to send e-mail from the site to users, and it lays the groundwork for future applications.

 

It costs $319 to remove grass from a dog's nose

This is my bulldog, Dexter:

Dexter started shaking his head violently on Monday night, and it was pretty clear that something had gotten in his nose somehow. We raced around Willow Glen and Campbell looking for an emergency vet, but the one close to eBay has closed.  By the time we got there he stopped, anyway.

It started again on Monday morning on the daily dog walk, so I made an appointment with our regular vet.  It turns out that Dexter had inhaled two stands of decorative grass that I picked up from Home Depot a few months ago and planted near the back door.  I ripped it out of the garden as soon as we got home.

Dexter's fine now, but he's still a little out of it. Here are the lessons I learned this week:

• Decorative grass seems to attract dogs.
• Vet office can close permanently, just like a Starbucks.
• It costs $319 to have grass removed from a dog's nose.  (It required anesthesia.)

Who knew?

Working on a few new features

Other than the Amazon S3 problems, things have been pretty quiet here at the old Planaroo ranch while I was working in Las Vegas.  But that's about to change quickly.  Here are a few of the changes coming up soon:

Features and functionality

• Caryn from Lightwise Design is working on a great new homepage and logo design right now.  I really like what she's come up with so far.  The new logo is a variation of the current one, but it looks much better.  (The kangaroo will still be in there, no worries.)
• Checklist pages (like this one for Epcot) will soon have sortable columns.
• Users will be able to add attractions to the "My Plan" feature, which can be shared with family and friends.

Content changes

• Sea World Orlando and Universal Florida pages will soon join Walt Disney World to complete the Orlando content set.
• Next week will see the launch of the Las Vegas pages, including every casino hotel/resort on the Strip and Downtown.

Traffic Update

• We're starting to see more organic traffic from search engines, especially Google and Microsoft.  This is great news.
• The blog post about Amazon's recent downtime created a spike on Planaroo.com.  Of course that traffic won't linger, but it was fun to see.

Responding to Amazon's S3 outage

Today's S3 outage has disrupted a lot of major web sites today including Twitter and SmugMug, both of which rely on S3 to host photos and/or icons.  The service is coming back online now, but it's been down for almost 7 hours total.

We host all of our photos at Amazon S3, but as a new site with low (but growing) traffic levels Planaroo obviously won't feel the impact that these major sites will.  However, we're optimistic that Planaroo.com will be a big site one day so it's interesting to think of ways to prevent bad things from happening.

Here are the lessons we're taking away from today's problems:

• Backups are a good thing. The one thing we did right was to keep current backups of the photos we host on S3.  That allowed us to manually replace the URL in our photos database table with a new URL quickly and easily. 
• Fix it. Don't hope for the best.  S3 went down at 9:06 am Pacific, and we had photos up and running by around noon.  We didn't do anything for two hours because we though -- incorrectly -- that the system would probably come back up quickly.  That was a mistake.
• Keep backups up-to-date and ready-to-go. Planaroo is based in a private home at the moment, and we depend on our Comcast connection to upload files to our servers at Slicehost.  Normally that's not a problem since the entire code base is small, but Planaroo has well over 1000 photos, and we store small (225x315) and large (500x700) versions of each one on the server.  Comcast's upstream speeds are slow, and they throttle down after the first few MB are uploaded.  It took over an hour to upload all the photos, and that's just not acceptable.  In the future we'll keep a backup ready to go in the event that Amazon S3 goes down, and we'll automate the switchover process.  We'll also think about seeking out a faster upstream solution.
• We were right not to host critical CSS, JavaScript, or icons on S3.  It's bad enough to serve broken photos, but broken CSS or JavaScript files can render a site unusable.
• Google's AJAX Libraries could go down, too.  We're using Google's AJAX Libraries API to serve the prototype.js file.  Google could have a similar downtime, and we need a failover plan for that too.

We would like to hear what other small startups are going in response to today's problems.

Photos restored

All photos are now being served locally.  Amazon is making progress with S3, but the system is still mostly down.  We'll stick with local photo hosting until we know more about the problem.