MAPokemon.com — Spotting a gap, acting fast and avoiding a gamble.
Pokemon Go — I don’t even have to tell you what it is, whether you’re 5 or 95 it’s everywhere.
Like most other millennials I was eager to get my hands on Pokemon Go. I wasn’t even aware of its existence before it really hit the first App Stores in Australia — although I know many people had been awaiting the app for several months. As soon as I heard that it was released in NZ/Aus, along with the instant virality it gained within the first 24 hours; I was intrigued. I checked the UK app store daily and even considered doing the “hack” of changing my App Store to the Australian version, although later opting to just wait for the proper UK release.
14th July was when I finally got the app and created my account, and immediately my brain started working on ideas, this app was already taking the world by storm — and for the most part I had missed the first week or two of possibilities due to it being released so late in the UK.
On the evening of the 14th I started brainstorming and my main gripe with the app was that my home is in a fairly suburban location with only 1 Pokestop in the visible distance. Quickly I became aware that most other people are in a similar situation with regards to a lack of Pokestops on their doorstep, and would love to know where their local gyms and Pokestops are. A few hours later I had a pretty solid idea, and a list of possible domain names.
The idea: A map that shows all Pokestops and Gyms, with the ability for community moderation and additions.
I knew time was of the essence, my idea wasn’t solid or planned, but in these situations you go with your gut instinct.
Most people who’ve seen me doing any type of web dev work know I’m a shortcut addict, speed and efficiency lover. I quickly fired up Sublime Text, and first of all got to work on the basics of HTML, CSS, JS and then the core of the Google Maps API. At first, all the Pokestops and locations were a static list I had acquired myself as well as exporting KML from other Google Maps that existed. I did a fair amount of data manipulation on the KML exports and then built a MySQL database around this structure and started pulling the results from it. On initial release I saw new locations slowly trickle in from the community, and as the site gained momentum I really wanted to be ahead of the curve and avoid site outages. I put my focus on stability, and started with caching the DB results every 2 seconds to avoid DB load, I even upgraded the Google Maps API to a business account as the places search library very quickly started to hit the daily free limit. Since then CNET, PC World and Reddit all put quite a few links to the site and I was thankful I had done my performance enhancements when I had.
Since it’s initial release I’ve created the logo, added features for voting on locations, a new drag and drop system for adding locations as well as many security and UX improvements, including a fairly basic system that checks people aren’t trying to add Pokestops at sea.
MAPokemon’s full stack:
- Google Business Maps API
- jQuery and ajax for all interactions
- Google Ads/Analytics/Webmasters
- npm: js-uglify
- Hosted on: Media Temple
In the back of my mind I always knew that an actual Pokemon spawn/location tracking map would be the most popular Pokemon Go webapp, but deep down I knew that Niantic would be very unlikely to allow these tracking apps and would quickly ban them. As of writing (3rd August), I’m very glad I had this gut instinct, as most, if not all of the Pokemon spawn tracking apps have been taken down.
Personally I hate adverts, and never built MAPokemon with the aim of making money, however I decided to place one small Google Ad container in the sidebar to help with hosting and domain costs that were quickly racking up. So far in my opinion this ad balance is fairly honourable, I’m not making much more money than the site is costing and the single Google Ad is fairly unobtrusive — and if you don’t like.. just use Ad Block!
In the space of two weeks MAPokemon went from nothing, a vague idea, a basic HTML Google Map to where it is a today; a solid website, reliant on fully paid Google Maps API’s, DB caching, community additions and voting with at least 99.99% uptime.
The site now consistently gets a real time visitor count of over 100 and has had over 200,000 unique visitors in under 2 weeks.
I’m going to keep maintaining the site and adding features, however my gut instinct is that Pokemon Go will be a distant memory by most people by the end of September unless Niantic quickly start putting out some big updates with regards to the social side (friends list/trading etc).
I hope others can gain an insight from this article into how easily, quickly and efficiently websites, apps and more can be created in a niche to make the most of any viral markets and crazes.