COMPANY: Odeko
PROJECT: Building user trust and capturing lost revenue with out of stock recommendations
RESPONSIBILITIES:
UX Research
UX/UI Design
CONTEXT
When hired at my current company, there was no research process in place and the company did not strategically use feedback and research to inform design decisions. My first couple months at Odeko consisted of developing a research process, integrating with support feedback, internal feedback and doing a round of customer research (surveys, in depth interviews) to better understand users and present insights to relevant stakeholders.
Many impactful projects were born out of this initiative, but one of the most impactful was a larger initiative on how our customers interact with out of stock items.
RESEARCH AND ANALYSIS
Frustrations around out of stock made up 38% of user pain points
My first research initiative was a general survey to better understand what the top pain points were for users of the Odeko supply portal. We found that frustrations with our current out of stock experience made up 38% of survey results.
DIGGING DEEPER INTO THE ISSUE
Lack of transparency, frustrations around finding alternatives and late notice were some of the top pain points within the out of stock flow
We found that out of stock frustrations could be broken up into two main categories - frontend out of stock frustrations (when/how users are presented with an out of stock item on the portal) and back end out of stock frustrations (what happens when an item goes out of stock during pick/pack). The two types of out of stocks had different root causes and subsequent pain points, but were correlated in interesting ways, and most users had experienced either one or the either in their customer lifetime. Below are the main themes and pain points within frontend and backend out of stocks.
How widespread were these issues?
In order to get the full picture, I wanted to understand how widespread these frustrations were. Did they really impact everyone on a large scale? I dug a bit deeper into some relevant metrics.
We saw that backend out of stock issues were actually made up a very small percentage of total orders, yet the frustration around these issues seamed very high because there was more at risk (users didn’t get the things they needed for day-to-day operation).
A huge opportunity to make up for lost revenue
Whereas, frontend out of stock issues were a bit more widespread... In Feb we had an average of 278 errors per day. In Jan there were 800 attempted orders for Martinellis, 1200 attempted orders for Saratoga, even after advanced notice comms went out about these issues.
This shows hundreds of users getting the front-end out of stock modal when trying to order items. Given the context from the survey and follow up interviews around why these front end out of stocks were so painful to users (see below), this felt like a very widespread, impactful frustration…
Lack of transparency around out of stock items. We don’t present users with info about out of stocks until after they’ve created an order.
Users are frustrated with the multi-step process to add an alternative when presented with the OOS modal:
“Because baristas are doing this while they're also working and serving customers, and so if they go through this process, they count up all the inventory, hit the order for review, then they get this message that says ‘Oh, go back and pick something else.’ That could push them into a situation where it's 5 minutes to cutoff and they still have to place this order. And now that they've lost that time, it becomes annoying for them.” - Cafe Owner
“Its so frustrating when an item is out of stock because you have to clear your entire order to add something new and because we place orders at 3:30, we often don’t have time and we just have to get it from Costco early the next day.” Cafe Manager
Because users often don’t have time to order something new through Odeko or don’t have visibility of possible substitutions, there was huge opportunity to capture revenue that would have otherwise been lost (users only option on the modal was to ‘continue without the item’).
IDEATION AND STAKEHOLDER INPUT
To the drawing board…
I had many solution ideas around solving the out of stock problems (both frontend and backend). Since many of the solutions involved other stakeholders and teams (procurement, eng, operations, warehouse) I wanted to keep my solutions very very low fidelity in order to get the right stakeholder input before moving to a higher fidelity ideation.
I held an impact / feasibility brainstorm with a representative from each key team to help provide context.
The perfect intersection of impact and fesaibility
Due to procurement and eng complexities identified, many of the solutions had a heavier lift, but we were able identify one solution that was an impactful + quick fix…
THE SOLUTION
Subsitutions on the out of stock modal
We worked with procurement to create a very manual list of out of stock alternatives that integrated with the portal so that (depending on the market) all users could see out of stock alternatives suggestions (rather than just a singular option to continue without the out of stock item…
SUCESS METRICS
This project was a big success. We captured an averaged of 250 new items added to customer orders from out of stock substitutions and 20% of these items became consistent repeat orders. We captured $46k in revenue on these substitution suggestions in a 6 month period.
