This weekend the Integrum team decided to participate in Chandler Startup Weekend. Startup Weekends are weekend-long, hands-on experiences where entrepreneurs and aspiring entrepreneurs can find out if startup ideas are viable. Teams focus on customer development, validating their ideas, practicing LEAN Startup Methodologies and building a minimal viable product.
As a team we decided it would be a fun way to spend some time together and challenge our abilities. We told ourselves that we would either join a pitched idea and work as the development team or we would play with an idea of our own and deliver the product. Either way we wanted something that could be done within the short time we had together while leaving no pressure to continue past the weekend. We also knew we were slightly handicapped because we would NOT be working on Sunday.
The Wednesday before Startup Weekend we spent a fifteen minute time box on brainstorming some ideas. Here is the list we came up with:
- * Meme Siri – Voice generated meme creation.
- * People Near Me – Find people near you and engage them in conversation (specifically replace campaign door knocking)
- * Hosted Curation – Move AgileWeekly platform to be multi topic
- * Media Timeline – Show how media portrays events over time
- * Two Face Opinions – Show how public figures contradict themselves
- * Craigslist Car Scraper – Low cost car buying tool
- * Lunch Decider App – Mobile app to decide where to go to lunch using core protocols
Friday night we showed up ready to jump right in to something. We got some dinner from Pittsburgh Willy‘s and then started listening to pitches. As the pitches were winding down we decided to keep it just our team because we would not be around on Sunday and to go with the Craigslist Car Scraper. At this point it was about 8pm and we found a room and got started.
We started out by doing a quick Agile Lift Off. This process took us about 30 minutes.
Vision: Allow vehicle buyers with less than $5,000 an efficient way to locate a new vehicle.
Values: Fun experience, Less BS, Less time on Craigslist, Best deals right away
- * rotate pairs every hour
- * ok to checkout
- * hourly checkins
- * deploy every hour
- * no stories greater than 1 hour
We then did a quick story workshop in about 30 minutes to generate a backlog of stories. After creating the backlog we sorted it and started planning. It quickly became apparent that at no stories larger than 1 hour, that 1 hour sprints probably made sense. When breaking down tasks we made our smallest time box 6 minutes or 0.1 hours. Most tasks were either .1 or .2. We setup two clocks, one to measure iteration time (1 hour) and one to notify of time box (6 minute).
At the end of every sprint we would do a Core Protocol Checkin and then do a quick retrospective. After the retrospective we would take a break and play some NFL Blitz. Then right back into another planning (generally about 10 minutes).
We worked from 8pm – Midnight (4 hours) on Friday and 8am – 8pm on Saturday (12 hours) for a total of 16 hours including time for breaks, meals, etc. We had 3 sprints on Friday night and I believe 6 sprints on Saturday for a total of 9 hours of keyboard time.
This is what we produced http://hooptie.co
What we learned:
- * we can still practice what we preach
- * working agreements are important
- * breaks and having fun increase sustainable pace
- * things that look really big get small really quick when you break them down
- * if you have an eye for simplest solution you can complete more than you think
- * getting user feedback early is important and better than arguing to decide what to do
- * sometimes good user feedback shouldn’t be implemented right away in favor of shipping
- * that deploying after every commit is insanely powerful
- * pairing is still an effective way to develop code even under pressure
- * having a plan is still worth it even if you are only doing an hours worth of work
- * iterations are powerful, the smaller the more powerful