Bloomingdale’s Registry Redesign
Role
UX/UI, Research
Duration
1 year
Overview
Registry was a legacy experience within the Bloomingdale’s site. It had not been updated since it’s original creation, it was not responsive and lacked key functionality that would make it useful to someone registering for gifts.
This project was funded based on a prototype I built and pitched. The investment took the shape of a year long lean lab effort. Lean labs are an agile structure where a large effort is broken down into smaller manageable sprints and iterated on.
Follow the north star to find a budget
Context
Bloomingdale’s in store registry experience did not match their online experience. Users had frustration and difficulty navigating the registry site due to confusing user experiences and desktop only pages. Traffic was relatively low within Registry compared to the rest of the site. Sadly, there was also no budget to overhaul it.
Make it matter and win a budget
At the end of the year Bloomingdale’s (much like others) enters code freeze. My team was using this time to think about future forward enhancements and I chose registry as my focus.
My approach was to do a competitive analysis, compile the best features and audit our own site. I signed up for every registry that ever existed and poured through every page and detail of their experience. I read reviews from users on the various platforms and began cobbling together a rough design based on sketches and examples I found. Compiling all the best features I started to build out a north star experience that I thought would get people excited about what our registry site could be.
I took my prototype and peddled it like a door to door like a salesman. First to my boss to rally support, then to her boss, then to our Group Creative Director, the head of marketing, the director of the site itself and then the entire engineering team. The engineering team had one question, when can we get started?
When projects were reviewed for the upcoming year my registry redesign was awarded $1.2 million in funding. The registry redesign would run as a lean lab for a full year. (lean and agile product structure containing 2-3 backend devs, 3-4 front end devs, one product manager, one scrum master, one QE, a tech lead, a copywriter, 2 product designers.)
The North Star
Shown above (from left to right): Bride’s View of Registry North Star version with personalization, BVR added products and most-loved items, Empty registry with curated categories.
Shown above: Other fun features baked into the initial pitch for funding. Features include, personalization, additional interrupter ad space, simplified product detail pages, appointment scheduling, consultant chats and interactive registry checklists.
Kicking off a lean lab
Problem
After auditing our registry experience we found that we had a lot of repetitive links within pages, many frustrating and convoluted rabbit holes to buy a product and some majorly unnecessary pages that were confusing for our users. The product manager pulled what she could from customer feedback to analyze the problem. Most customers were frustrated and some were outright angry responding in profanity. In addition we needed to:
Make the registry experience responsive (and mobile first)
Registry should be easy to us for both registrant and guest
Create more premium feel that fits the brand and coincides with the in store experience
Make it easier to maintain (leverage systems design)
Optimize for SEO and continue to keep latency top of mind.
KPIs
The redesign was broken down into feature releases, each release has specific KPIs but overall generally we were looking at:
Primary: conversion & revenue
Secondary: AOV
Also tracking: customer sentiment through survey results and lower call volume to customer service
Constraints
Registry must remain a separate experience. Dragging/ranking items within a registry, navigation updates (unless small), and gift card registry are all out of scope.
Setting the roadmap for the lab
Based on the UX research and prototype the product manager and I created a year long roadmap of features to tackle and ranked the features with senior leadership. Meanwhile, our developers toiled away setting up the backend framework for our endeavor. Our backend developers worked tirelessly with our front end guys to get a new framework in place to match the rest of the site. The number one priority at the time was to make the digital experience responsive. The existing registry site legacy format could not be manipulated so we needed to build from scratch.
The release schedule shows the breakdown of the redesign into smaller or more manageable feature releases.
Process & iteration
The goal with lean labs is to have UX get out ahead of the upcoming work and have it ready and approved for pickup once engineering frees up. As more projects and features take flight some engineers may stay behind to account for fixes or iteration while one might jump ahead to start the next feature. The whole team works together, engineers are part of the process the entire way, at no point should an engineer be shocked about the thing they are about to go build.
Before code release there would be an onscreen review with the entire team in which feedback and release blockers were addressed.
Defining the product
When working on a project I would:
Comb through existing research from The Knot, Wedding Wire and Bloomingdale’s customer care
Look at other competitors in the field
Use a MoSCoW to determine a list of features and enhancements we wanted to target and what we know is out of scope
Create user flows and run sketch sessions with cross functional team members.
Create prototypes and in lieu of a user testing platform that would be tested in office with users (unaffiliated with our team and project) that fit our profile inside of Bloomingdale’s office network.
Work closely with engineers to validate feasibility of ideas and define edge cases
Wireframing the Registry Manager
A major point of confusion for couples who were registering with Bloomingdale’s was the figuring out the difference between registry manager and the Bride’s view of registry. In order to eliminate this problem I came up with a solution to combine these pages into one useful feature by turning the manager into a secondary navigation within the Bride’s View of Registry.
I ran in-person usability testing with my product manager in the office as Bloomingdale’s at the time did not have a user testing platform. We gathered volunteers from other departments unrelated to our own and asked them to try out the new functionality of the guest and bride view with the newly incorporated registry manager.
This is a high level wireframe example of the Registry Manager feature for the host. Agile updates were made in the UI phase based on feasibility and feedback. Note: Not all annotations listed here, some are dependent on other features and may have been addressed in previous wires.
A year in the lab and many features later, we went from this….
to this…
Above are just a few screens from what went live. Left: Guest View of Registry grid, Middle: Add to Registry overlay clean up, Right: Thank you manager
Above & Below: A few of the most dramatic redesigned pages that went live.
Results
$12.6m increase in revenue by December
We exceeded our original goal which was $12M increase in revenue before the year end. We also had a small lift in registry creations which is a great start considering this part of the site had long been neglected in previous years.
Working in tandem with the development team and analytics, we were even able to do a few smaller tests on the side and make some UI updates that were in our back pocket as a nice to have. We also received positive comments via our web survey from happy customers ecstatic about the ease of use of the new Registry site.
Future Enhancements
As with all MVPs, if you started with a north star prototype you have to scale back for the minimum product. Keeping that in mind, I had some features in my original prototype that offered more personalization for the user. Many comments we received across the entire e-commerce site were regarding personalization. While this was not a target for the MVP I did take steps to ensure a framework was in place in case we wanted to explore a future enhancement.
Personalization
Color Background Content placeholder behind the Registrant’s information where in the future we could allow the user to choose their own image (currently there isn’t a backend system in place to store the images). I also created a great feature called “Top 5 favorites” to help Couples rank their most-loved items.
Collections
Something like this could easily turn into a more editorial experience for this brand, but for the MVP we hardcoded a few trendy categories based on the Little Registry Guide initiatives for that year. Collections would take you to a browse experience with items chosen by the sales team but in the future would ideally tie into the Little Registry guide and have more of a voice and personality.
Recommendations
Currently there isn’t a system in place to create smart recommendations for a Bride. The product recommendation panels algorithms misses the mark a little bit. For the time being, I made space for them, but the recommendations are hard coded by the sales team with items that are top sellers in their categories.
Checklist
More interactive and dynamic checklist would be nice to have. Currently The Little Registry has a printable pdf. I had also pitched a Style Quiz to help get a more curated collection for our couples to shop from.
Wedding Shop
Explore new ways to showcase UGC Bloomingdale’s has formal wear in the likes of suits, dresses, and shoes etc.The idea was to create space within the registry manager navigation to house this cross promotional opportunity. trust, transparency, diversity & inclusion. Additionally, allow the user to register for a gift card which is currently not enabled on the backend.
Navigation
Registry had some of the oldest navigation on the site and many UX problems that needed to be addressed, but this was largely out of scope and an undertaking all its own.