What I learned from my second startup experiment
My name is Stan Bradbury, Founder of Graffiti (www.graffitiover.com). The individual who helped me run this experiment are Steve Sherman (https://trello.com/szsherman). The link to my specific hypothesis on Trello is https://trello.com/c/cwOcieoy.
I am part of an experimental incubator that gives startups and entrepreneurs funding in exchange for testing and sharing lean startup hypotheses.
I THOUGHT that I could get 100 meme creators will install my Graffiti Chrome extension at http://bit.ly/2u7Feqa in the five days from Oct. 17–21 as we expand our marketing to Reddit and Facebook because it makes making memes easier and provides more tools for creativity.
ACTUALLY only 55 of 733 people did this after clicking through to the Chrome Store. In total, 286,000 meme creators were exposed to my Reddit, Facebook and AdWords campaigns at an expense of $361.62 for an average cost to acquire a user installation of $6.58.
The MOST SURPRISING metric was budget increased a total of 31% compared to the same five days in the week prior when only AdWords was running yet our conversions per day decreased 17% during that time period. I BELIEVE this happened because the quality of click-throughs we were getting from Reddit (only 0.10%) and Facebook (0.22%) were not actively seeking a meme solution like the users that searched for a meme word on Google to see our AdWords campaign. As a result our conversion rate fell 25% from 4% to 3% during the same five day periods.
The NEXT STEP is to increase our conversion rate from 3% to 9% by sending traffic to a second Chrome Store location. The main Chrome Store URL reported an unaccounted for 1196 and 1119 impressions for each five day period that were not from click through traffic from our ads. Our second location has no random, unaccounted traffic on it yet. I BELIEVE nearly all the random traffic does not convert and install our extension. You can find that experiment on Trello at: https://trello.com/c/nkVgcMUw. The funding pays for our advertising campaigns and gives us the flexibility to do a lot more A/B Hypothesis testing as we move forward.
