There is a version of this deal that looks like genius. Grab already knows which of its users order food at midnight, tip generously, and pay their drivers on time. By analyzing ride activity, delivery volumes, and payment histories, Grab can assess creditworthiness in ways that traditional banks cannot. Now, with its $1.49 billion acquisition of a controlling 60% stake in Atome Financial, it gets a proven consumer lending operator to go with that data advantage.
The strategic logic is clear enough. The deal will combine Atome’s businesses, including BNPL loans and cards, consumer cash loans, and digital lending, with Grab’s financial services business. Atome operates in Singapore, Malaysia, the Philippines, Indonesia and Thailand and serves 25 million cumulative transacted users. The 80% revenue growth Atome posted in 2025 suggests strong product-market fit, and two years of pre-tax profitability indicates the unit economics are working. Grab’s CFO Peter Oey put a number on the ambition: the combined financial services segment is expected to generate $500 million in adjusted EBITDA by 2028, with a gross loan portfolio of over $6 billion.
That portfolio figure deserves a pause. Grab’s gross loan portfolio reached $1.438 billion as of Q1 2026. Atome Financial has said it holds a roughly $1 billion gross loan portfolio. Getting from roughly $2.4 billion combined to $6 billion in under two years means more than doubling the book. In consumer lending, speed and credit quality are rarely friends.
Grab’s financial services segment posted an adjusted EBITDA loss of $17 million in Q1 2026. Spending $1.49 billion on a lending acquisition while your lending division is still losing money requires a certain confidence in the trajectory. The deal is funded entirely from existing cash. Grab has also said it does not expect the transaction to affect its existing capital return plans. But the acquisition cost sits on top of a spending spree: Grab previously agreed to acquire Stash Financial at an enterprise value of $425 million (with 50.1% paid at closing), and agreed to acquire Foodpanda’s Taiwan operations for $600 million, meaning Grab has committed over $2.5 billion in acquisitions in a relatively compressed window.
The competitive pressure driving that pace is real. GoTo, Grab’s primary rival in Indonesia, has been building its own financial services capabilities through GoPay and its partnership with Bank Jago. Standing still in this market means ceding ground to a well-capitalized opponent. Buying Atome accelerates Grab’s position, but it also raises the underwriting stakes considerably across five markets with different regulatory regimes, different credit cultures, and different default behaviors.
This is the tension a disciplined investor has to sit with. The distribution advantage is genuine, but Grab itself frames the edge more as ecosystem data plus Atome’s AI-powered lending infrastructure than any single metric. Owning the consumer relationship at the point of daily habit, then extending credit through that same interface, is a compounding loop that few lenders anywhere in the world can match. Grab and Atome Financial have said they plan to share risk-management insights, regulatory best practices, and collection strategies to combat platform fraud and enhance credit underwriting.
The question is whether execution keeps up with ambition. Grab raised its overall 2028 adjusted EBITDA target to $1.7 billion and upgraded its outlook to 30%+ group revenue CAGR between 2025 and 2028. Those targets price in a great deal of things going right simultaneously. The structure of the full acquisition offers some protection: the second phase valuation will use a combination of adjusted EBITDA and revenue multiples, with an equity valuation floor of $2 billion and a cap of $4.5 billion, tying the final price to Atome’s actual performance rather than a fixed promise.
For long-term investors, Grab’s integrated model remains one of the more compelling emerging-market financial platforms being built today. The Atome deal could be the move that tips it from interesting to exceptional. But the lending book has to perform. In consumer credit, distribution is the starting line, not the finish.