Providing Retailers with an Integrated Point of Sale System
I led the product design process for a 6-month project, launching a tablet point-of-sale system. The project aimed to provide retailers with brick-and-mortar stores a point-of-sale system that directly integrated with our existing inventory platform.
The Opportunity: Turn Headaches into Sales
CommentSold is the leading live video commerce platform that enables businesses and brands to create a live shopping experience for their customers. As we grew our user base, we frequently saw shops with brick & mortar locations struggling to maintain one source of inventory management across their sales channels. Many shops frustratingly adopted a rival point-of-sale system or manually reconciled their inventory.
Goals
Our goal was to provide a more seamless sales experience for our users through the following:
1
Reduce amount of time a store manually reconciles inventory & inventory discrepancy rate
2
Manage influx of orders from multiple sales channels
3
Allow shops to manage omnichannel strategy without extra fees or integrations
4
Provide mobility depending on sales environment
Defining the Problem
Discovery began with an alpha proof of concept; this prototype aimed to assess interest and market viability for a streamlined point-of-sale solution for retail outlets.
User Interviews
We conducted targeted interviews with specific user segments:
Alpha testers
Retailers with physical stores or frequent pop-up shops
Stores generating over 100K in monthly revenue through physical outlets
Retailers actively utilizing competitor POS systems
Key Insights
Insights gleaned from these interviews played a pivotal role in shaping our foundational feature set (P0).
1
Hardware Compatibility
Won’t adopt a new system unless it can be configured with existing hardware (barcode scanners, receipt printers, cash tills).
2
Barcode Support
Needs to support existing barcode system – all products across all channels should use the same structure.
3
check-out friction
Forcing in-store shoppers to have to create a store account in order to make a purchase caused additional friction in the checkout flow and resulted in abandoned purchases.
4
disjointed shopping experience
Existing shoppers could not use or earn account credit in-store.
Design Principles
To ensure we stayed in alignment with user interests, I crafted design criteria to benchmark ourselves against during the course of the product development.
mobility
📱
Touch surfaces and actions need to be large & visible enough that users in motion can easily identify tap areas.
environment
🌳
Provide shop employees with the right data at the right time; don't oversaturate the UI with extraneous details.
simplicity
👍
User actions need to be easy enough that even the least tech-savvy users can complete vital tasks.
Trimming the Fat
We realized that many customer interactions were dependent on whether or not they had an account with the store — we needed to remove this friction for both the retailer and shoppers. The dependency was more detrimental than helpful.
Old checkout flow – The TL;DR: you may have fallen asleep by the time you pay.
drag me, too!
New checkout flow — Congrats, you've entered the 20th century.
Collaborating with my PM and Engineering Lead early on allowed us to uncover any potential feasibility issues and brainstorm new solutions.
Building MVP
At the time of this project, we lacked a design system and were amidst a company-wide visual rebrand. Additionally, we faced hurdles with the outdated and inconsistent UI of the core product. To streamline the transition to high-fidelity mockups, we collaborated with the Brand Creative team, opting for a shade of the brand blue as to build our primary palette around. The hope was once a design system was established, we would continue to iterate until fully adopted.
Early explorations
User Feedback
We had a rough P0.5 available – we took this back to some of our original interviewees to gauge reaction.
One of our users asked about support for split payments. It was here that we uncovered that while the overall CS platform does not support split payments, we technically could support it on the POS, only if someone used account credit as one of the forms of payment. We needed to go back and figure out a flow for this.
However, the compromise was that was all or nothing, the shopper couldn’t apply a specific amount to their purchase.
Challenges
⏰ Ramp-Up Time
🎨 Visual QA
Impact
$1.38M
gMV of single early adopter
$404K
gmv within one month of beta launch
$95K
GMV generated in single Saturday
271
terminals ordered & setup
⭐️
increased employee satisfaction for retailers
User Testimonials
"We recognized a tremendous improvement right away...fewer lines in the store, more accuracy in all of our orders, fewer inventory discrepancies, we were able to process quick cash payments and credit cards, and we had no problem receiving the funds deposited into our bank account."
Future Steps
While the application was a great success with our users, leadership postponed any new features on the 2023 roadmap. To keep the product front of mind with customers, Product Marketing conducted a short survey to ask what features we could develop to improve their experience.
The top three responses:
Support same login on multiple tablets
Support fractional sizes (for shops that sell fabrics or craft items)
Support physical gift cards
Internally, our goal was to update the system with CRISP. Unfortunately, I was a part of a reduction in force prior to being able to revisit this.
Learnings
This was the first tablet application that I had to design from 0>1. My prior experience was in providing direction on a resort in-room application to an external agency.
Working with your engineers is vital. Don’t be afraid to involve them in the design process so they can better understand your decisions.
Kristin Ferrari © 2025