Gigantic Peace Loving Decoy

Whew. What started out as an idea for a few updates for turned into a yearlong quest to figure out the right way to get current, accurate, and useful information about the Portland beer scene into your hands. The way people use websites is also changing due to the proliferation of devices like smartphones, tablets, and everything in between. While desktop and laptop users may take more time digging through the site, mobile users may need to find something quickly, easily, and bandwidth-friendly. Since I’m a data-wrangler at my day job, this seemed like a fun problem to solve: a few weekends and it would be finished, right? Oh optimism, you are a quite a liar.

For those of you that want to jump off start exploring, check out some of my favorite features below. If you want to know a bit more the hows and whys of what we did, continue reading.

  • Beer Finder—Want to explore Portland beers by numbers? Search beers by setting ranges on IBUs, ABV, and color! Looking for a dark beer, on the sweeter side, with the kick of a mule? Start here.
  • Recently Mentioned Beers—Want to know what beers people have been talking about in the last few minutes? In almost every case, these beers represent what people are drinking right now! This means you don’t just have to read it, you can see what all the buzz is about for yourself.
  • Near Beers—Need a visual clue about what’s going on near you? The last 100 beers mentioned are available on this interactive map. If a beer is within 1000 feet of you, it will be highlighted!
  • Beer and Brewery Details—On beer landing pages, you’ll not only be presented with details about that beer, but also how it measures up against other beers from that brewery. Like that beer but want something a little stronger, or more bitter? With these charts, it’s easy to see where this beer fits in the context of the other beers. It works the same way for brewery landing pages. The brewery charts compare all of the breweries in Portland by average IBU and ABV. This is a great tool to see which brewery is bucking the trends and making beers outside of the norm.
  • Interconnected Data—Every beer listed on the site includes the style and the brewery. Clicking on the style will show you all of the other beers brewed in Portland in that style. Clicking on the Brewery will bring you to the brewery page where you can focus on only beers brewed by that brewery. In fact, as of this morning, the site has over 3900 unique internal pages—a number that grows every day.

It's Alive! has always stood out from the other blogs by being able to connect people with beer. Providing recent beer and brewery data, maps, statistics, and event listings has distinguished us from the rest of the beer sites in Portland. But all of this was all managed by my hands alone. Manually pulling out relevant facts from emails, Facebook events, phone conversations, and sloppy notes was extremely time consuming. New beers were released faster than I could type, and new brewery openings were just around the corner. When a hobby becomes an overwhelming chore, it ceases to be a hobby worth pursuing.

Enter the robots. In a nutshell, actively seeks out publicly shared beer and beer-related information across the Internet. To keep the data fresh, there’s a focus social media sites like Twitter, Facebook, Instagram, Flicker. When the site “hears” about some beer-related data, it goes through the following process to ensure that it's relevant to our readers:

  • Is the data valuable? Using tools like Natural Language Processing, Keyword Extraction, and Concept Tagging, this site can decide if this data is relevant to us.
  • Was it mentioned in Portland, OR? Geofencing allows the site find out the origins of most data.
  • Was it mention about Portland, OR? Does the relevant data pertain to our city? Again, geofencing and data analysis allows us to tell if the data is about Portland, or pertaining to something within Portland.
  • Does this data exist yet on the site? While we never want to duplicate data, it can be important to know that the data was referenced more than once. For example, if the data is a beer, and 900 people reference it within an hour, the site knows not to add the beer again, but it will classify it as extremely important. In fact, using Sentiment Analysis tools allow us to tell if those 900 people are loving that beer, hating that beer, or simply mentioning it.

Once each piece of data goes through this process, the site queues it up to be added to the site immediately. It took a lot of tweaking, but the data that appears on the site has been testing consistently over 99% relevant.

On the Go!

Mobile usage is a huge focus for the site as well, but fitting so much data in a screen is always a chore. My primary focus was to shorten the distance between the user and the data. To this end, I concentrated on two fundamental concepts: search and related data. Simple and accurate search functionality ensures that the user can find what they’re looking for in a short window. Providing accurate related data is a way the site can keep their attention. If they are looking at a specific beer, the site offers up other related beers, related styles, and articles related to that beer. Secondarily, but very important, I’ve made sure that the content is optimized for delivery to mobile devices. Optimizing server response times, sending device-centric content, and making use of data caching for content that doesn’t change often, have all helped reduce the latency of the site, while increasing the delivery speeds of the content.

Damn, it's 2015!

As the great Marty DeBergi once said: "Enough of my yakin'. What do you say? Let's boogie." Get clickin', check out what we've done, and let us know what you think! The team is excited to get back into the Portland beer trenches with you in 2015. Happy New Year Portland!