Case study: Big Stock

The optimization team at BigStock, an online stock photo provider, noticed their customers were having problems with their searches. If they misspelled a word in their search, visitors were less likely to see the results they were looking for. The team came up with a solution: adjust the search algorithm to create a “fuzzy autosuggest” when running searches. As a result of this algorithm, they estimated that they would see an uptick in people selecting results from the auto suggest.

BigStock-Seach1

With this hypothesis established, they ran their experiment. When the results came in a week later, Bigstock found that visitors selected results from the fuzzy auto suggest 9.6% more often. They used their analytics, coupled with their results, to find a 6.52% increase in the number of images added to visitors’ carts and a 3.2% increase in image downloads.

BigStock-Seach2

By centering their hypothesis around a problem they found their customers were having based on analysis of their data, BigStock was able to solve a real issue with their search algorithm. Their results came quickly and had a direct impact on their conversions.