Came across an interesting experiment from Etsy on infinite scrolling. Day McKinley give a interesting talk on why infinite scrolling failed and how they can better approve the implementation process.
Etsy implemented the infinite scrolling on their search result page, assuming people like to see as much results sooner on the page. After the assumption, the team comes up with this idea using infinite scroll, where you can keep scrolling and see as many results as possible. They did a AB testing and got the result that infinite scrolling made people stop using search. Their learned lesson is to conduct the testing sooner in the process. But, would that really help solving the problem? What is really wrong with the process?
I think the assumption is correct that people do want to see more results and see the result quickly. However, the solutions didn’t really take into account the context. Instead of asking how we can have a solutions that meet the assumption, we should really ask what do the users need and why. Why they want to see more results? What happen when they see the result? Do they go back to a certain item? Do they need to remember where the item is so they can find it again? In the case of something like Etsy, the user is searching for something in particular within the list of results. As ux designer, we have to always keep in mind the scenario, user goal and put users in context.