Among the most interesting findings of User Research activities to improve Seller Onboarding flow in OLX was that most sellers have no idea about the proper price for their stuff to sell. This is due because OLX is a Customer to Customer (C2C) second-hand classified platform, where most of the sellers are common people, and for many, it’s also their first time selling their preloved, personal item online.
The current pattern we identified was for these casual sellers to first do research, either online or by asking their friends, to understand better the common price range. Some managed to get a good price for their stuff, while some find it hard to sell their stuff as it turned out they were asking for a price above the market rate.
The latter segment tends to have a bad experience selling in OLX which ends up leaving the platform for good, contributing to the churn rate. Fortunately, OLX’s Big Data already records thousands if not millions of successful transactions in the past, and this is something we can tap into.
I was thinking of shortening the research activity for our common sellers or even removing it completely, by creating a price recommendation feature where OLX would be able to suggest the ideal selling price, based on past recorded transactions.
The feature ends up as a price range suggestion, highlighting the tradeoff between faster speed of sale but lower profit, and higher profit but slower sale speed.
As this was a new feature, and a non-interactive one, we don’t have the baseline of measurement nor a proper way to measure its performance, however, from the follow-up research we discovered that it’s becoming one of the features that users would like to keep.
As an additional note, the common seller’s churn rate was also decreased, however, there were a couple of other improvement initiatives running in parallel hence we can’t really isolate the effect of this price recommendation feature.