• Successful MVP with minimal rework from prototype.
  • 50% activation rate.
  • 20% retention rate.
  • Arguably better user experience than competitors.

A problem worth solving

Long queues at cafes in Sydney’s CBD along with heavy pockets full of coins motivated us to create an app that would help customers enjoy their favourite daily fixes sooner and ensure they have less coin clutter to lug around.

Another small but significant personal annoyance both cofounders experienced over the years was around one-size-fits-all deals regardless of how loyal we were to the merchant. After even further exploration, we discovered merchants had issues of their own. They had exorbitant payment processing fees and Point of Sales (POS) terminal costs to contend with.

We were passionate about addressing these issues for patrons and merchants alike.

Our users

We had two sets of users to accommodate.


Patrons, being the end-customers who would shop at merchants, varied considerably with how they intended on using our service. Some preferred only to use the service to avoid having to carry cash, others were more interested in deals whilst yet another group wanted to browse menus from their phone but place orders in-person.


Merchants, being the stores that sold coffees and food to patrons, were our customers. Of the merchants we had interviewed during our preliminary customer research, the majority had similar needs.

In four words: increase sales, lower costs.

We uncovered something quite thought-provoking here. While we had been interested in the increase sales side of the equation, we hadn’t really considered the lowering costs side. We had already been planning a solution for pre-ordering and pre-paying which had to work with the basic POS system we were building, so it made perfect sense to address the POS side of our solution to help merchants lower their costs (after our pilot.)

Crafting a smooth experience

Armed with solid knowledge of what our users thought of our hypothetical service, we went about translating our ideas into wireframes. By this stage, I had assembled an informal user advisory group whom I met with regularly. Wireframes and multiple journey options were reviewed on numerous occasions.

A rather interesting observation was made during these sessions. The original homescreen we had intended offered a snapshot of all the local merchants registered with our service within a certain radius. You’d tap on a merchant tile and be taken directly to that merchant.

Our new homescreen in the making.

Our new homescreen in the making.

We had discovered that more people seemed interested in deals to start with. It “lured” them. They weren’t as interested in looking at deals filtered by merchant either (or within a specific merchant’s profile.) The homescreen was changed accordingly to display deals and promotions being showcased by local merchants.

We wanted to be careful to accommodate those users who enjoyed browsing deals and menus vs those who craved efficiency.

We resolved this tradeoff by allowing patrons to favourite not just individual products but entire orders with their specific preferences allowing them to place those orders with a single tap. Furthermore, we made a decision not to allow patrons to pre-order and subsequently line-up to pay – it would defeat the purpose of our service.

Favourites screen before and after.

Favourites screen before and after. Favourites was originally planned as a tabbed section within the Home screen. We ended up assigning it to the top-level menu structure in the interests of efficiency.


To satisfy those users who didn’t prefer using credit cards to purchase pre-paid credit, we designed a simple feature which enabled users to select an amount of pre-paid credit they’d like to pay and request a unique code which they’d take to their local Cornershop merchant with the cash and the credit would be instantly applied to their account.

Another decision we were quite proud of (and received a lot of positive feedback from) was that we didn’t force users to sign up. We allowed them to do as much as possible within the app before having to create an account (e.g. we allowed them to browse deals, menus and other areas of the app.) This makes for a much more enjoyable experience without the pressure to surrender personal details too soon.

Just as important as our functional tradeoffs, we wanted to ensure merchants were able to present themselves in the best possible light. In delivering this, we permitted Showcase promotional items and menus to include custom graphics.

Product detail screen before and after.

Product detail screen before and after. Instead of the original plan to represent products with a thumbnail illustration, we scrapped the idea in favour of allowing custom full-width high resolution product images allowing merchants to proudly display their masterpieces.


Those merchants who are very particular about their branding can maintain a very high level of control with Cornershop. Others who don’t have that same desire could choose from a multitude of attractive stock images.

The minimum viable product

Once we had honed the interactive prototype, we had our blueprint ripe and ready for development. We weren’t going to build everything we had planned (at least not yet) so we had to carve out just enough features to make the product useful while keeping it very light so as not to build features people may not use. This approach would also assist us in delivering a beneficial product to market much sooner.

These features (and nothing more) formed our scope:

  • Pre-paid credit accounts
  • Pre-ordering
  • Favouriting an order
  • General promotions
  • Order history

We decided to target a specific type of merchant for our 3 month pilot and eventual launch: a small and somewhat busy, non-franchise operation who may not have a state of the art POS setup and pre-ordering app but would benefit greatly from one.

The pilot, metrics mania and a pivot

According to Dave McClure’s advice (founder of 500 Startups), there are two key metrics that should be considered before product/market fit. They are: activation and acquisition.

Activation is measuring whether users have a great first experience (account registration.) Retention measures whether those users keep coming back (order placement.)

Of the number of users who downloaded the app, we had a 50% activation rate. Of those active users, 20% would place orders daily. That’s a significant retention rate.

But we didn’t pop the champaign bottles just yet. We needed to uncover some areas that needed addressing.

We wanted to understand why 50% of users weren’t signing up. Were they only browsing the deals and/or menu? Were they just curious about the service but not serious about using it? Were they under the impression they needed to purchase pre-paid credit to signup? Was it difficult to know where to go to sign up?

We also wanted to know why 30% of users weren’t placing orders. Was it not obvious there were two choices for payment? Was the app hard to figure out? If so, in what way? Were they planning on using the service “some day”? Were they expecting more from the service? Oddly enough 5% of our users were creating Favourites but never ordering from them. This piqued our curiosity.

The top-level navigation structure with Showcase front and centre.

The top-level navigation structure with Showcase front and centre.


The pilot reaffirmed the decision to have Showcase front and centre. Thanks to our Google Analytics integration we determined on average, users spent the most time on this screen.

Order status icons (left) and our way of letting you know your order is ready (right.)

Order status icons (left) and our way of letting you know your order is ready (right.)


The next popular screen was Menus followed by Order History (within Orders) so we certainly had the correct navigation structure and an efficient workflow for discovering products, ordering them and keeping up to date with order progress (via push notifications and order statuses.) No changes to be made here.

The interview

We rounded up 20 randomly selected users to interview. We had the questions, they had the answers. We learnt the following key points pertaining to user experience:

  • Of those 50% of users who downloaded the app but didn’t create accounts, their biggest hurdle was the missing loyalty functionality. They didn’t want to use one service to pay and order their coffee and yet another to track loyalty. It was “too hard”. This led us to prioritise loyalty functionality in our public release.
  • Of those 30% of users who had created accounts but weren’t placing orders, their biggest gripe was that it wasn’t obvious you could pay as you go (i.e. you didn’t have to purchase pre-paid credit.) We subsequently redesigned the payments workflow to make PAYG more obvious.

Overall, the app was easy to use and it was a breeze to find what they were looking for. We didn’t need to make any changes to the navigation structure or the layout of our screens. Our interactive prototype and its subsequent iterations seemed to have hit the nail.

Our ability to identify the right metrics ensured we asked the right questions of our users and subsequently made the right changes.

One, two, three, pivot

Perhaps our most insightful feedback came from our merchant. He told us that he’d be reluctant to push this service if it were to feature his competitors.

What the merchant sees.

What the merchant sees. The Cornershop Register screens showing order tickets and their status (left) along with the options available should a merchant choose to decline an order (right.)


This encouraged us to think of the service in the form of a white-label solution instead of its current directory implementation (i.e. the same underlying functional platform with the ability for merchants to brand the app with their own identity.)

After further interviews with other merchants and the overwhelmingly positive feedback, it was obvious this was a better strategy for the company to consider.

The future

Cornershop is now in the process of re-architecting the product into a white-label solution. It’s set for a very bright future. I believe we have now accomplished the first step towards the potential reinvention (and much needed shake-up) of the POS industry.

I have since left the company to pursue other endeavours but I still advise on user experience matters.

other work

Willoughby Squash

Willoughby Squash Centre needed their existing online court booking process squashed and subsequently overhauled. Customers and management wanted a more elegant design that simplified the number of steps in the booking process, was easy and intuitive to use and worked just as well on mobile or tablet mediums.