A big barrier we have as a marketplace platform is to get visitors to trust us enough to make their first purchase with us. We hypothesised that users are very likely to make multiple transactions after their first transaction. The goal then was to convert them into a paying user as we have quite a number of users who signed up without making any purchases or sales.
To add to that, we have to solve the chicken and egg problem: for users to sell more on our platform, we need more buyers in the first place. One of the ways to solve this is to increase the amount of offers made, which we solved with a marketplace revamp.
The first thing we did was to research on flash sale drops that are done by other platforms. We were particularly interested in the features they have as well as the user’s process to making a purchase on a drop. We were lucky because at comencement of this project, Nike Sneakers was doing some flash sale drops.
We mapped the users journey, where they come from, what they look for and how will they reach their goal state. This document also serves to inform the developers as to the actions users go through as well as the information they should be presented with.
While mapping out the flow, we realised that users might leave if a product is sold out. We devised ways for users to funnel back into the site. If a product is sold out, they can wait for the next product or if they really like that particular item, they can purchase them at non-discounted price.
This way, we convert potential buyers to make a purchase from the sellers on our platform.
We made quick sketches to test internally for usability, this is to ensure we have our priorities in place and that the solution actually makes sense for our users. With the workshops done, we moved to low-fidelity wireframes and testing. These are clickable prototypes that our users can play around with.
We conducted usability tests on this to make sure we get it right for our users. In that process, we made several quick iterations to improve on the design before we begin work on the final UI.
The final UI was a collaboration between the UI designer, graphics from the marketing designer and me. We made sure we covered all edge cases as well as the interactivity between pages and element. As the flash sales will be live we needed a system that has a before and after interactivity between these states:
When a visitor lands on the dedicated flash sale page, they will be able to preview the next drops. We can choose the timing of the drop for maximum effect so the drop won’t be predictable.
A day before the drop we will tease the product that will drop and then using social media, we promote the timing of the drop at the day itself. The timer will start an hour in advance and users will be prompted beforehand to sign in or register for an account as well as to key in their payment and shipping details. This is a win-win scenario, as it will be easier for users to checkout in the future for purchases outside of the flash sale drops.
Because the flash sales are time sensitive, it should generate quite a lot of excitement leading up to the drop. Because sizes are limited, users will likely camp earlier and then quickly checkout. Users who missed out on the chance to score a flash sale drop, will be given the option to utilise our first time purchase voucher on the same product outside of the sale. This should still convert users who FOMO or new users who have not used the first time purchase promo code.
Our flash sale products were snapped up in mere seconds, but there are other less popular products that stayed for a few minutes. During the flash sale drops we see a tremendous amount of activities from visitors on our platform.
Most importantly, we see that 40% of first time purchasers during this period ended up making subsequent purchases within the next three months.
Envisioning this system was a fun creative exercise into what would get our users buzzing with excitement.
While the traffic and word of mouth we received from our users have been quite positive, we do acknowledge that this is probably an expensive way to grow users.
At this stage at growth, it might be necessary for us to have a system like this but moving forward into the future, we will probably adapt this into a more cost effective mechanism.
Overall, we achieve what we wanted — positive brand enforcement and the acquisition of new buyers.