Merit · Product · Internal Explainer
Merit's other business sells commerce infrastructure to loyalty programs. This one sells directly to the person holding the points. One wallet, every program, and a balance you can actually spend. This page starts at the one-paragraph version and ends at the four-market launch plan.
If you only read one section, read this one.
A Gulf consumer holds seven to ten loyalty cards and actively uses fewer than three. Airline miles sit in one app, bank points in another, telco points in a third, and none of them talk. In KSA alone an estimated SAR 2.4B of points expire unused every year. The value is real, it belongs to the consumer, and it is unreachable.
The B2C Super App is the consumer layer that unlocks it. It aggregates points from many issuers into one wallet, then makes that wallet spendable: gift cards, physical goods, and a checkout that takes points, cash, or both in the same transaction. Underneath it is Merit ID, one identity across the ecosystem, and the mTrust Score, a 1 to 100 transaction-integrity signal.
This is the mirror image of the E-Commerce Solution. There, a client owns the member relationship and Merit is the engine behind their brand. Here Merit owns the consumer directly. It exists because B2B sales cycles are long and the client always owns the customer, so Merit needs a moat it controls.
The whole product in one picture
Many issuers, many balances, none of them spendable where the consumer actually wants to spend. Merit aggregates them into one balance and gives it somewhere to go.
Android since March 2026, iOS released 30 July 2026 after the guest-browsing fix cleared App Review. Gift cards, physical-goods marketplace and Moyasar payments are operational.
Revenue is booked as full transaction value across first-party and third-party. Take rate is a margin metric, not the revenue metric.
Across four markets, with a December 2026 checkpoint of 125K monthly transacting users and $11.04M GMV per month.
The B2C Super App is a New World product end to end. It is built on the modular Superplatform services and it does not run on, read from, or fall back to the legacy platform. If you have come from the E-Commerce Solution side, the Old World and New World split that dominates that product does not apply here. There is only New World.
Four roles. The important shift from the B2B side is that the consumer is now Merit's own user, not somebody else's member.
B2C against the E-Commerce Solution
Same company, same supplier network, opposite ownership of the end customer. This is the single most useful contrast to hold in your head.
An issuer on this side of the house is often a client on the other side. STC appears as a point issuer for the Super App, as a client for the E-Commerce Solution, and as an Apple distributor supplying Al Fursan. When someone names a company, check which relationship they mean.
| Need | Person | Role |
|---|---|---|
| Product direction, roadmap, priorities | Brian Arfi Faridhi | Product Director, owns B2C plus Ecom, Marketplace and Platform |
| Company strategy and final calls | Fred Barry | Chief Product and Growth Officer |
| All point mechanics, earn and burn | Tamer | Product Director, KSA. Owns LMS |
| LiveOps, data, mTrust, central wallet | Marthino Yuda | Product Director, LiveOps and Data |
| Platform services the app consumes | Akshay Chennupati | Sr. Engineering Manager, Platform |
| Growth and acquisition | Rufus Giwa | Growth Marketing Lead |
| Delivery coordination | Khawar Rasheed | Project Manager, Tintash |
| The mobile build and App Store | Mudassar Raza | Mobile Engineer, React Native |
The daily 14:30 WIB ceremony is the B2C, Seller Portal and PIM standup, and only the first of those three is this product. Seller Portal, storefront, PIM and the Ecom order and fulfilment services belong to the E-Commerce Solution, and gift-card merchandising belongs to Marketplace. Hearing an update in that meeting does not make it a B2C item. Tickets for this product are on the Jira B2CR board, and the team moved from sprints to Kanban on 31 July 2026.
Six streams. Two behave like software, one behaves like retail, and mixing them in a P&L will produce a number nobody can defend.
Revenue definition first, because it trips everyone up. Revenue is full transaction value, GMV, for both first-party and third-party transactions. The take rate is a margin metric. If someone quotes "8% revenue" they are quoting margin, not revenue.
Where a transaction's value goes
The same GMV yields very different margin depending on whether Merit is passing a third-party sale through or selling its own goods. Never blend the two.
| Stream | Basis | Indicative | Margin character |
|---|---|---|---|
| 3P transaction take rate | Percentage of points spent or traded | Blended ~8% | Software-like. COGS is payment processing plus fraud |
| 1P merchant margin | Margin on first-party goods, dark kitchens and exclusives | 25% to 30% | Real retail. COGS is goods, logistics and returns. Model separately |
| Issuer revenue share | Share when a partner's points are spent | Per partner | Pass-through split. Net take is after the issuer's share |
| Data-as-a-Service, mTrust | API access sold to financial institutions | Per call or subscription | Very high GM software. COGS is compute and data ops |
| Point top-up spread | Spread on points bought or converted | Per transaction | Spread capture, low COGS |
| Consumer premium, Merit Elite | All-access lifestyle membership | ~SAR 299 per month | Subscription GM. COGS is funding the partner perks |
Monthly GMV, checkpoint model
Five committed checkpoints from the Q3 2026 close through to the December 2027 model. Hover any point for the paired user number.
The B2C top-line is under revalidation and the pricing section of the Product Page is still an indicative draft pending Finance sign-off. The $50M year-one figure is organic-execution-led: it assumes product quality and market launches, not the full paid-acquisition budget. Do not quote any of it externally without checking first.
What is actually built, what is being built, and what is a roadmap promise. Status as of early August 2026.
What lands when
Every bar below is owned by the B2C team. Bars are positioned by target ship, not by effort. Hover any bar for its dependency.
| Component | What it does | Status |
|---|---|---|
| Mobile app | The primary surface. Manage, exchange and spend points. React Native, built by Tintash. | Live, KSA Android |
| iOS release | Same app on the other half of the market. Cleared App Review once guest browsing shipped. | Live, 30 Jul 2026 |
| Wallet and Ledger | Source of truth for the user's points and cash balance, and for deduction at checkout. | Live |
| Mixed Payment engine | Points and cash in a single transaction. Payment saga: reserve, charge, commit. | v1 shipping |
| Gift cards, digital goods | The highest-volume thing people buy in the app. | Live |
| Physical-goods marketplace | Real products, browsable and buyable in the app. | Live |
| LMS loyalty stack | Cashback, referral, gamification and P2P point mechanics. The single most critical Q3 dependency. | Integration then features Aug to Sep |
| P2P point transfers | Free transfers by phone number or Merit ID, split bill, and the viral loop where the recipient must download to claim. | Build Aug, ship Sep W4 |
| QR Scan-to-Pay | Offline merchant payment on network rails. Needs Merchant Locator, POS integration and a banking ledger. | Highest ship risk in Q3 |
| Gamification and mTrust display | Score 1 to 100, badges, streaks, leaderboards. B2C owns UI and triggers only. | Sep to Oct |
| mTrust Engine | The scoring model itself, owned by Data Platform. Also sold as Data-as-a-Service. | GA target Oct 2026 |
| Point Exchange | Any-to-any point conversion. Development not started, market research ongoing. | A2A beta Mar 2027 |
| API and SDK | The B2B layer, so gateways and e-commerce sites can accept Merit points using Merit ID. | Roadmap |
| Virtual Card and NFC | Merit-branded Visa or Mastercard in Apple and Google Wallet. | Beta Jan 2027 |
| Blockchain, $MERIT | Points-to-token bridge and on-chain settlement. | Testnet PoC Sep 2026, production 2027 |
| Analytics, Mixpanel | Every OKR is unmeasurable without it. Hard dependency, not a nice-to-have. | Instrumentation in progress |
First, there is a Tintash transition freeze from July to mid-August 2026 with no new feature development, and every roadmap date assumes it. Second, and now resolved: the iOS release was rejected under App Store guideline 5.1.1(v) because the app forced login before a user could see anything. Guest browsing was the fix, and once it shipped iOS cleared review and went live on 30 July 2026. The follow-on, guest checkout, is still open and still worth doing on conversion grounds alone.
If you internalise a single thing about how this product is built, make it this one. It decides what the B2C team is allowed to build.
The boundary
Everything above the line is the B2C team. Everything below it is a Superplatform service the app calls. The app never reimplements anything below the line.
We never build a parallel points backend. This comes out of Fred's Superplatform mandate, one multi-tenant stack, 100% New World by December 2026. It is the reason a feature like Daily Spin looks trivial in the app and still carries a real dependency: the spin is UI, the points it awards are LMS.
Why one late dependency moves half the quarter
Almost every loyalty and gamification item is gated on the LMS integration landing first. If the root slips, everything downstream of it slips with it.
| Dependency | Type | Why | Owner |
|---|---|---|---|
| Merit ID | Hard | Universal identity and SSO. Ties the wallet to one person and is the single front door for every funnel. | MFC |
| Wallet / Ledger | Hard | Balance truth and deduction. | MFC |
| Point Exchange | Hard | Any-to-any conversion engine. | MFC |
| LMS | Critical | Backend for every point mechanic. | Tamer |
| mTrust / Fraud | Shared | Transaction integrity and the DaaS product. | Data Platform |
| Notification Service | Synergy | Push, email, SMS and WhatsApp dispatch. | Platform |
| P2P Transfer Module | Synergy | The transfer mechanics. LMS moves the balance. | Platform |
| Mixpanel | Hard | Every OKR and every model recalibration depends on cohorts. | Platform |
Nothing in it is built by the B2C team, and nothing in it is a B2C deliverable. When one of these services is discussed in a shared standup, it is that team's update, not this product's status. The B2C deliverable is always the surface on top.
Three verbs. Everything in the wallet is one of them, and they are at very different stages of maturity.
The loop, and how much of it exists
Burn works today. Earn is the Q3 build. Exchange is the differentiator and the least built. The ring shows the loop; the fill shows how real each arc is.
The competitive field splits cleanly. Read-only aggregators like AwardWallet can see your points but cannot spend them. Single-program embeds like PointsPay can spend one program's points but aggregate nothing. Card optimisers like MaxRewards route a card choice but touch no loyalty balance. Curve consolidates cards, not points.
Capability coverage against the field
Filled means yes, hollow means no, half means partial. The empty column down the middle is the whitespace.
Regional enterprise loyalty vendors such as Dsquares and TADA sell into clients, so their programs stay siloed per client by design. Bumped, sometimes cited as a comparable, has been defunct since December 2022 and was never a points aggregator.
The checkout is where the product either delivers on its promise or does not. It is also the hardest engineering surface in the app.
The whole basket is paid from the aggregated balance. LMS burns, no cash leg. Requires login, because points are bound to an account.
Points and card in one transaction. This is Merit's proprietary piece and it removes the wall that stops someone redeeming on a high-value item when they are slightly short of points. v1 is a split; the slider is v2.
Card, Apple Pay or Google Pay through Moyasar. The only rail that can in principle work without an account, which is why guest checkout is scoped to fiat only.
The mixed-payment saga, step by step
Two ledgers, one transaction. Step through the happy path, then force the card leg to fail and watch what has to unwind.
Every gate you put before the money is a gate a percentage of people will not walk through. Guest browsing moved the account ask from the front door to the checkout. Guest checkout moves it past the money entirely, to a one-tap offer after a successful purchase, when the person has already decided they like you.
From opening the app to points landing back in the wallet, using a gift-card purchase with cashback as the reference case.
Walk the flow
Press play, or click any stage. The stage detail appears below the track.
With guest browsing, search, category and product detail work with no account. Add to cart also works. The account ask moves as late in the funnel as it can go.
A gift card is issued instantly with no logistics. A physical product inherits a whole separate machinery of supplier selection, shipping and returns, owned by the E-Commerce Solution team. Most of the app's volume is digital, which is why the app can be live and stable while the physical side remains the harder operational problem, and why that problem is not this team's to solve.
Every stage above emits a Mixpanel event. That is what makes checkout completion, first-time-buyer rate and month-one retention measurable, and it is why analytics is a hard dependency rather than a reporting nicety.
Year one is KSA first, then UAE, then Jordan, with Australia running separately. Each market is a different payment rail, a different regulator and a different point-issuer landscape.
Year-one GMV by market, base case
KSA carries well over half the plan. Hover any bar for its rails and its launch timing.
| Market | Year-1 GMV, base | Priority point issuers | Payment rails |
|---|---|---|---|
| Saudi Arabia Primary, live | $28.8M | STC Qitaf, around 15.7M members. Al Rajhi Mokafaa, Saudia Al Fursan (already integrated), STC Pay | Moyasar, integrated. mada acquirers: Geidea, HyperPay, Nearpay |
| UAE Mirror launch, Nov 2026 | $14.0M | Emirates Skywards, Etihad Guest, Al-Futtaim Blue Rewards, Carrefour MyCLUB | Network International, Checkout.com, Aani for CBUAE instant payments. Separate PSP and contracts |
| Jordan Gift cards first, Dec 2026 | $2.5M | Bank of Jordan, whose bare cash-to-card redemption is the strongest wedge. Royal Jordanian, Arab Bank Arabi Points, telcos | CliQ via a Network International acquirer, JoMoPay |
| Australia Synchro, as-is | $5.0M | Everyday Rewards (Woolworths), flybuys (Coles) | EFTPOS and NPP (PayTo) via a licensed acquirer, plus Synchro |
What a new market actually costs, using Jordan
Nine workstreams, only one of which is language. This is why market timing is a Fred decision and not a sprint item.
In KSA, blockchain production sits outside virtual asset regulation entirely. The token is internal and closed loop only: transfer-restricted, spendable only inside the Merit merchant network, with no fiat conversion offered anywhere in the product. That design is what keeps Merit a technology service provider rather than a virtual asset service provider, so production is gated on engineering and a legal opinion. In Jordan there is an open question with local counsel on whether the product sits as a loyalty overlay or needs an EPSP licence. Data residency and KYC requirements apply per market. Australia runs as Synchro as-is in year one, with Merit integration a year-two dependency.
Organic execution alone caps the business at $3M to $4M ARR. The gap to the target is closed by three stacking paths, and each one is gated on a decision rather than on engineering.
The gap, and what closes it
The organic core cannot reach the target on its own. Each path is a decision Fred has to make, not a thing the team can build its way to.
| Path | What it is | Contribution | Gate |
|---|---|---|---|
| A · Institutional | Onboard government and state-owned point issuers with large captive bases and idle liability. Sales motion, no product build. | $8M to $15M | Senior BD hire, around $80K |
| B · First-party merchant | Take 25% to 30% real margin on the same GMV instead of an 8% pass-through, starting with a dark-kitchen pilot. | A margin layer, not new GMV | Ops Manager hire plus a SAR 500K float |
| C · Paid acquisition | CPI-optimised performance campaigns, target CPI at or below SAR 25. | Stretch accelerant beyond $50M | MMP tool, around $20K, plus budget beyond the confirmed B2C share |
Five key results, four of them about quality
The quarter's theme is reliability, then retention, then revenue, in that order.
Two of these are named products with owners and dates. The rest are openings.
mTrust, the 1 to 100 signal
It does double duty: it protects the transaction inside the app, and it is sold to financial institutions as an API. That second use is the highest-gross-margin line in the whole revenue model.
Behavioural trust and risk scoring on a 1 to 100 scale, engine GA targeted October 2026, owned by Data Platform. B2C owns the display, the badges and the score-based perks, not the model.
Best-method routing at checkout: which card, or points, maximises value on this specific purchase. It is the one feature where a competitor, MaxRewards, already has a comparable capability, so it is a parity play rather than a differentiator.
Ranking by spend behaviour and loyalty tier. Blocked on behavioural data volume in exactly the way you would expect, so this follows instrumentation rather than leading it.
Real-time monitoring of suspicious activity, feeding and fed by the mTrust signal.
The Mixpanel measurement framework is still being set up, funnel definitions are still settling, and current dashboard numbers are not yet reliable ground truth. Confirm any metric with Marthino before quoting it, internally or externally. This is also the practical limit on what can be trained, evaluated or personalised right now.
The terms that will come at you in week one with no explanation attached.