Grab Buys Control of Atome for $1.49 Billion to Scale Its Lending Arm

Grab agreed on September 15, 2026 to acquire a controlling 60% stake in Atome Financial, the buy-now-pay-later arm of Singapore's Advance Intelligence Group, for $1.49 billion in cash. The company called it a definitive agreement with the sellers, and it is Grab's largest acquisition to date. If regulators approve, the deal would give the Southeast Asian super-app a scaled consumer lending business rather than the one it has been slowly building inside its own app.

Related reading: Bending Spoons to Buy Miro at a 90% Markdown From Its 2022 Peak · LEGO Digital Play Buys Mobile Studio Offroad Games.

The deal, announced by Grab, is as much about financial targets as it is about the product. Grab used the announcement to raise its 2028 group goals to $1.7 billion in adjusted EBITDA and more than 30% compound revenue growth from 2025 to 2028. It also said its financial services segment, including Atome, should reach $500 million in adjusted EBITDA and a gross loan portfolio above $6 billion by 2028.

The structure protects the buyer

Grab is not paying $1.49 billion for the whole company. It is paying that for 60%, with the remaining 40% to be acquired roughly two years after the first phase closes, under a formula negotiated in advance rather than a fixed price.

The terms are worth reading closely because they show how carefully Grab has hedged a business it cannot fully observe yet.

• Phase 1: 60% for $1.49 billion in cash, of which $260 million is primary growth capital that goes into the business. Completion is expected by the third quarter of 2027, subject to regulatory approvals.
• Phase 2: the remaining 40%, about two years after Phase 1 closes, at a valuation built from 13.0 times annualized adjusted EBITDA plus 2.5 times annualized revenue, each measured over the six months before the second closing and weighted 75% and 25%.
• Collar: the resulting equity value is floored at $2 billion and capped at $4.5 billion. At least half of the Phase 2 consideration is paid in cash.

The formula ties what Grab eventually pays to Atome's demonstrated performance. If the lending book grows and earns, Grab pays more. If it stalls, the floor limits how cheap the second tranche looks relative to what was paid for the first, while the cap protects Grab from overpaying if Atome grows faster than expected. Grab is funding the deal entirely from existing cash, says it will be accretive to group adjusted EBITDA upon completion, and says it does not affect its share buyback.

What Grab is actually buying

Atome Financial is not a small startup. It says it has 25 million cumulative transacted users and works with more than 30,000 brands, and its app is ranked by Sensor Tower as the most popular buy-now-pay-later app in Asia. Its product line has moved well past point-of-sale installment payments to include cash loans, a BNPL card, and digital lending. In Indonesia it operates Kredit Pintar, a leading digital lender. The company is funded by a roster of institutional partners including Standard Chartered, HSBC, Bank Jago, DBS, SMBC, and BlackRock, and its parent Advance Intelligence Group is backed by SoftBank Vision Fund 2, Warburg Pincus, Ares, and EDBI.

That mix is the point. Grab already runs payments, digital banks through GXS in Singapore, GXBank in Malaysia, and Superbank in Indonesia, plus partner lending, insurance, and its own consumer credit. What it has struggled to build quickly is a large, standalone consumer lending book with its own underwriting engine. Atome has one, and Grab's management was direct that buying it was faster and cheaper than building it.

"Atome Financial gives us a proven consumer lending operator and an established merchant base, letting us scale our Financial Services segment significantly faster and more cost-efficiently than building it ourselves," Grab chief financial officer Peter Oey said in the announcement.

The credit quality claim is the one to watch. Grab says Atome's gross loan portfolio stood at about $1 billion as of the first half of 2026, with delinquency rates stable or improving across borrower cohorts and prudent loss provisioning. Those are unaudited figures supplied by Atome, and Grab will be accountable for them once it consolidates the business.

Why this is a distribution play as much as a credit play

Southeast Asia is a favorable market for lending and a hard one for lenders. Grab cites Temasek research that more than 70% of adults in the region are unbanked or underbanked, which describes a large population that traditional banks have not served and that digital lenders see as their core opportunity. Grab says that in 2025, 68% of its driver-partner borrowers accessed formal credit for the first time through the company, and half of them said they did so to avoid predatory lenders.

Those numbers describe the social case, but the business case is simpler. Grab has nearly 54 million monthly transacting users who already pay for rides, food, and groceries inside a single app. Adding a credit product to a payments relationship is one of the oldest cross-sell plays in consumer finance, and it works best when the platform already sees a user's transaction history and income pattern. Atome brings the underwriting models and the merchant network; Grab brings the users and the data. Both companies say the overlap in their markets is minimal, which is a claim regulators will test, since they operate in the same five countries.

The transaction also carries a regulatory dimension that is easy to understate. Financial services in these markets are licensed, and consumer lending is among the most closely watched categories because of over-indebtedness risk. Both companies said their obligations on licensing, consumer protection, data privacy, and responsible lending in each market are unchanged by the deal, and Grab framed the combination partly as a way to share risk-management and collection practices to prevent over-indebtedness. Whether regulators accept that framing, and how many jurisdictions attach conditions to their approval, will shape how quickly the deal closes.

A super-app strategy with a shrinking runway

Grab's bet sits inside a larger pattern. Super-apps across Asia have spent the last several years converting mobility users into financial services customers, because payments and lending have higher margins and more recurring revenue than ride-hailing, which is labor-intensive and price-sensitive. Grab's financial services segment is the part of the business investors now watch for growth, and the company's decision to raise its 2028 targets alongside the deal is a signal that it wants the market to value it on those numbers.

There is also a competitive reason to move now. Buy-now-pay-later has consolidated globally, with large platforms absorbing smaller specialists to gain underwriting scale. In Southeast Asia, the field includes regional banks, telcos, and e-commerce players all chasing the same underbanked customers. Grab buying Atome removes one potential competitor and denies it to a rival, and it does so with terms that defer much of the price.

Caveats and what to watch

• It is not done. Completion is expected by the third quarter of 2027, subject to regulatory approvals and closing conditions. A more than a year gap between announcement and close is long enough for the business to change.
• The financials are unaudited. Atome's loan book, user counts, and delinquency figures come from Atome's management accounts, which Grab notes have not been reviewed or audited.
• Credit risk travels. A lending book bought at a formula-based valuation can still sour. The floor and cap protect Grab on price, not on credit losses.
• Overlap is contested. Both companies operate in the same five countries. The claim of minimal product overlap is the kind of thing competition authorities examine.
• The 2028 targets are forward-looking. Adjusted EBITDA and revenue growth targets are non-IFRS projections with no reconciliation, and Grab's own filing language lists the usual macro, regulatory, and execution risks.

The deal will be judged on whether Grab can turn a 54-million-user app into a lending machine without letting credit losses undo the economics. The structure it negotiated suggests management knows the risk is real. The size of the bet suggests it thinks the reward is worth it.

Cover: Grab delivery rider on a Bangkok road, by Chainwit. (CC BY-SA 4.0) via Wikimedia Commons.