RewardPay web portal
Company: RewardPay
Success metric:~4× monthly signups
I design for the RewardPay portal — when I started they had no reward partners at all, only Amex. Since then: four reward partners added, a subscription model and a referral program, through a period when monthly signups grew around fourfold.
- Project Duration
- 2023–Present
- Services Provided
- Product Design, UX & UI, Marketing Collaterals
What RewardPay does
Small and mid-size companies use RewardPay to pay their everyday bills — tax, payroll, rent, suppliers, super.
- Pay with American Express even where it isn't accepted — RewardPay takes the card and pays the supplier for you.
- Earn your normal Amex points on that spending, on top of everything else.
- Buy reward points as well: pick your partners at the payment step and pay a fee for them.
Qantas, Velocity, ALL Accor and Luxury Escapes — up to 6 points per $1
Amex, Visa, Mastercard and pay-by-bank, across every business expense type
Free, Silver and Gold — the monthly fee buys down the processing rate
Impact
Growth in monthly new-customer signups, Sep 2025 → Jul 2026 — and the curve is still steepening.
Growth in the marketing database over nine months
Transactions per active customer each month — repeat use, not one-off signups
The portal and campaign work I designed ran through this period, alongside new partnerships, paid acquisition and referral programs.
The problem I was there to solve
There were no reward partners when I started. Qantas came first, and taking it from one partner to four was its own long piece of work. Every partner needs its own screens — a way to sign up, a page explaining it, and a pay-and-complete flow where they see the points detail and earn rate.
One product, not four
Add each partner without the portal splitting into separate mini-products.
Keep paying simple
More choice at the payment step can't turn into a slower or harder payment.
Ready for the next one
Build it so the fifth partner costs a fraction of what the first one did.
Adding the partners
The main thread of my work. Qantas set the pattern; Velocity, ALL Accor and Luxury Escapes tested whether it held.
First: Qantas
- Portal banner — a homepage banner explaining the partnership, with one action: connect your Qantas account.
- Linking page — a dedicated page that collects and checks what Qantas needs, including Frequent Flyer number.
- Points tracking — a Qantas section showing points earned, and a second place to link an account.
The Qantas linking flow, from the portal banner through to connected account.
Then three more, and a page for each
Each of the three arrived with its own rules: what you earn, what's needed to connect, how the brand appears.
- A sign-up option for each partner — connect the account, confirm what that partner requires, show whether it's linked.
- A landing page for each partner — what it is, what you earn, how to join. Used in campaigns too.
- Choosing at the payment step — points are a purchase, so the choice sits inside the payment flow without slowing it down.
- Four brands, one product — a co-branded lockup per partner, cleared with each brand team before launch.
Velocity, ALL Accor, Luxury Escapes and Qantas — built from one pattern.
At the payment step
Where the partner work turns into money. A business adds one or more partners to the same payment and picks an earn rate for each — the fee beside it, the points total updating live, campaign bonuses stacking on top.
Subscriptions
Customers pay a subscription fee upfront for a better rate on their Amex payments — Free, Silver, Gold. The saving is a fraction of a percent against a fixed monthly fee, so whether it pays off depends entirely on how much a business spends. The design does that maths for them.
- Show the saving in dollars — a rate difference means nothing until it's dollars against what this business actually spends.
- Ask at the right moment — show a plan when someone is looking at what they just paid, not in a settings page nobody opens.
- Keep every rate honest — once plans exist, every rate in the portal depends on which plan you're on.
The dashboard for a customer on the Gold plan — the plan they're on stays visible alongside everything else.
Member-get-member
A referral program: existing customers invite other businesses, and both sides get something when the invited business signs up. I designed the in-portal flow and the emails that carry the invitation — growth through customers rather than paid acquisition.
- Inviting from inside the product — the invite lives where customers already are, not in a separate campaign.
- Seeing what your invites did — sent, accepted, earned, visible enough that the program feels worth repeating.
Referral invites that converted into an accepted invite.
July 2026 referral invites against the monthly baseline — the same month new signups hit their record.
Referral programme engagement, shown as ratios. The flow is small and highly targeted rather than high-volume.
The referral promo on the dashboard, the invite flow, and the referrals view.
Marketing emails
40+ email templates, many of them reusable — one-off campaigns and automated flows that run on their own.
- Campaigns that explain a change — when a partner went live, the email had to teach the offer, not just announce it.
- Flows that do a job — welcome, onboarding nudges and referral invitations run automatically. These are the ones that compound.
- Proof from real customers — short video stories from businesses already using RewardPay.
Sustained across campaigns — roughly double the 25–35% typical of fintech and B2B.
Click-to-open on the member-get-member sequence — nearly two in three openers clicked through.
Open rate held across the full onboarding sequence.
Six of forty-plus. Partner launches, subscription reminders, onboarding sequences and customer stories.
Portal improvements
Alongside the partner work, the portal needed steady improvement — the everyday admin that decides whether people keep using it.
- User roles and permissions — the portal allowed one user per business. Two patterns: many users to one organisation, and one user across many.
- Wallet management — where customers manage the cards and accounts they pay with. Renaming happens inline.
- Promotional tiles — dashboard tiles surfacing partner offers without getting in the way of paying a bill.
Roles and permissions, the wallet, and the dashboard tiles that carry partner offers.
What I assumed — and what I got wrong
The polished result hides the part that matters: what I believed going in, and what I learned along the way.
I assumed we'd instrument and learn.
I expected to watch how people move through the portal and adjust as I went. Without behavioural analytics in place, the work runs on judgement and the data I can see rather than on observed behaviour.
I assumed a good rationale would settle a decision.
It rarely does on its own. What worked was reframing conversations around the customer's problem and the cost of getting it wrong, rather than defending a design.
Visual feedback is welcome — settling it needs evidence.
Where feedback improved the interface I took it; where it came down to preference I proposed an A/B test. Next time I'd get that testing path agreed upfront.
What I'd change now
Two things I'd set as conditions at the start of the next engagement, rather than asking for them halfway through.
A design review step, agreed at kickoff
Building a short design check into the release process from the start, so the built screens and the design stay in step as a matter of routine rather than something picked up case by case.
Instrumentation as a starting condition
Behavioural analytics as a prerequisite rather than a recommendation — otherwise every argument about a flow is an argument between opinions.
