COMPANY: Odeko
PROJECT: Creating a more intuitive ordering experience that decreases time to conversion and increases discoverability
RESPONSIBILITIES:
UX Research
UX/UI Design
SETTING UP A RESEARCH PROCESS
In my first few months at Odeko, I completely redesigned the research process. Actually, I could say that I created the research process because there was no process before me. I set up a repository (after lots of competitive research, settled on Dovetail), designed a system to integrate with CS tickets to capture external feedback and pain points, created an ‘internal form’ to capture internal feedback and conducted a survey to understand the general landscape of how users felt at Odeko.
FINDINGS
Ordering pain points were within the top 5 themes mentioned within feedback and survey analysis
After analyzing all feedback I grouped pain points and patterns into “tags” in Dovetail and identified major themes. I presented themes to the appropriate project managers. The top 5 themes included out of stock issues, frustration with available roles and permissions, the way items were organized and general ordering frustrations (mainly related to how confusing and time consuming the process was).
CURRENT STATE (DESIGN ANALYSIS)
It takes 6+ steps to place an order…
In order to understand the ordering process that users were experiencing so I could have the context to understand their painpoints, I did some deeper analysis of the flow. I found that to order a new item through the catalog, Odeko users first have to add the item from the catalog (using an ambiguous ‘add’ button) and then navigate to their orders page, scroll (or search) to find the item within its category and then enter the qt of the item for the date they want it delivered, review the order and confirm. This requires 6+ steps (depending on where they’re ordering from), creating lots of friction for the conversion process.
BUSINESS METRICS
Average time to convert from catalog to order within a session was 7.5 minutes… and average time to place a first order was 8.5 weeks…
I understood from the current state why people might be frustrating with ordering. But, I needed to make sure these problems were widespread - and that there really was business impact here as well as user impact.
As a human centered designer, I wanted to embark on solving a problem that had a high rate of user impact and business impact, simultaneously. My hypothesis was that any UX problems with ordering are likely to effect conversion rates and ultimately overall revenue. I also needed to find these metrics in order to get stakeholder buy-in, so I did some deep diving and found that:
Of catalog conversions within 1 day (totals - so the same user could drop off, then come back) (in the past 30d) conversion from catalog -> order (after the user has decided to add the item) was 77% but the total time to convert was 1.2 hours.
And… if we look at conversion criteria of within 1 session, conversion to add an item and complete their order is incredibly low, with a total conversion rate of 25%. This is a very low conversion rate, given that they are adding an item that they typically want to order.
The average time to go from Catalog Item Added > Order Completed in a single session is 7.5 minutes. Given that users are clicking a + icon to add an item and are ultimately completing the order, the time to conversion per session is high.
The First Time Order Gap
From week 1 (signup), it takes an average of 8-10 weeks for a customer to increase their order size and monthly spending with Odeko, then it flattens out or follows the market trends.
While this definitely in part reflects the poor onboarding experience, adding items to the “grid” in order to place an order is extremely unintuitive for new users, which contributes to the first time order gap, and decreased revenue.
FINDINGS
Ordering was frustrating and confusing for both new and existing users
After being sure there was a business problem here, I did some deep diving into the customer feedback.
Ordering is not intuitive for new users. They need to add all items to the grid before they can order them.
Ordering a new item is time consuming (due to above) and takes a lot of scrolling and clocking to complete (add to list -> scroll to find -> update qt -> review)
Users get frustrated that they have to clear an order every time they navigate from the grid -> catalog to add something new (this is especially prevalent when something is out of stock)
“Too many clicks”
"There's just a lot of clicks to get to the catalog, like I know how to do it, but I know I would have to sit down and really show someone if we were to ever hire someone new or add someone on board. It's just very click-y. I don't know another way to put it. Like I have to do supply catalog, go to my catalog, or if I want to add something else or it's just too many clicks.” - Cafe manager,
“Confusing”
“And if there is something like the caramel sauce that I need, and I didn't - to have to go into the catalog and add it to my order list before I even order it can get a little confusing.” - Cafe manager
“I have to back out and start all over again. It’s super frustrating”
I personally don't like the whole orders vs. catalog thing. I would rather it just be catalog because if I ever need to add something that's not on my regular orders thing or just to see - I have to back out and literally like delete everything and start all over again. It's super frustrating.
IDEATION
How might we make ordering more intuitive for users and allow them to add new items from the catalog to an order without friction?
THE SOLUTION
A shopping cart with a twist
The solution I designed was a traditional shopping cart with a twist. Users can save items to their “order calendar” (traditionally called orders) but it is no longer a required step to place an order. They can choose to simply add and complete a cart order for next available delivery (simplistic, intuitive experience) or they can save an item to their order calendar to schedule deliveries in advance and create standing orders. Creating a curated list of saved on their order calendar also allows them to control what their baristas can order (baristas do not have catalog access).
Adding this more traditional eCommerce experience introduces an easier, more flexible way to place orders quickly. It’s also highly beneficial for new users who may not be used to ordering through the order grid. The cart is likely to capture lost sales due to the confusion caused by the grid, and boost conversion rate. It also allows us to introduce typical eCommerce marketing techniques like abandon cart nudges and unique CTAs to entice the user to check out.
SUCESS METRICS (COMING SOON)
In development
This project is currently in development with an expected release date of Feb 28th