- What are three conclusions we can make about Kickstarter campaigns given the provided data?
CONCLUSION #1: Categories are key, but don’t ignore subcategories Overall, projects in our data were successfully funded at a rate of 53%. This rate does not hold true across different categories though. If someone wanted to launch a successful Kickstarter campaign, they should investigate the following categories: Music, Theater, and Film. These three categories were the largest categories by count and were also funded at the highest rates of 77%, 60% and 58% respectively. If you removed these three categories from the data set the remaining projects were funded at a rate of 38%.
Subcategories are extremely important too! The rate of the music projects funded was very high, but not universal across different subcategories. In fact, exactly 0% (n=140) of music projects under the subcategory of jazz, faith and world music were funded. The remaining subcategories were funded at a collective rate of 96%. If you are a rocker, a classical musician or EDM producer Kickstarter is a great place to look for funding. Otherwise, it may be best to look elsewhere to get your album off the ground.
The ‘hardware’ subcategory of ‘technology’ was also interesting. ‘Technology’ as a category was funded at a rate of 35%, however the subcategory ‘hardware’ was universally funded at a rate of 100% (n=140). Excluding ‘hardware’, technology projects were successfully funded at a rate of 13%. I conclude that anyone interested in bringing their technology project to Kickstarter ought to make sure that their product can fall into the ‘hardware’ subcategory
CONCLUSION #2: Kickstarter is neither popular nor very successful outside of the United States and Great Britain
88.5% of the projects in our data set originated from the US or GB. 55% of these projects were successfully funded. The projects launched from all other countries were funded at a rate of 33.6%. While there has been an increase in the number of projects launched from these countries over the past several years (6 projects in 2013 to 176 in 2016), there has not been a notable increase in success. In 2016 only 35% of projects were successfully funded from non-US, non-GB countries.
CONCLUSION #3: Projects launched around Christmas succeed at a lower rate
January-November 49%-59% of launched projects are successfully funded, varying a bit month to month. In December, only 43% of projects are funded. Presumably this may have something to do with the holiday season directing consumer’s spending to family and friends rather than Kickstarter projects. February-June appears to be an ideal time to launch a project with May, April and February being the most successful months respectively. It seems that Kickstarters may have wised up to some of these trends as December launches the fewest number of projects and May, June, and July the most (though ideally the later should be shifted a few 2-3 months earlier for optimal success rates).
- What are some of the limitations of this dataset?
It would be nice if there was some sort of regional breakdown within the United States since such a large percentage of the projects were launched here. This could give better insight as to what type of projects are successful where. Along the same lines, it would be beneficial to have some sort of data that shows where the donations are coming from. This information could help potential Kickstarters market their project regionally. I would be interested in seeing data on the timeframe in which successfully funded projects deliver their product. If a play is funded how much time passed before opening night? If a documentary was funded how long until the premier? This data would be useful to both the Kickstarters and the donors. The Kickstarters may get an idea of how long it will realistically take for them to finish their project while donors will know how long before they see the fruits of their donation.
- What are some other possible tables/graphs that we could create?
I created a pivot table by country to get a breakdown of the count of state for every country. This was illuminating as it gave a more succinct visualization of the number of projects launched from various countries as well as their success rates.
A pie chart could be appropriate to show a breakdown of the count of the various categories though I am unsure of how much more effective this would be over the column chart we already created. An area chart could also be helpful, but again, I am unsure if this would be a more effective visualization of the data rather than the charts we already created.