Affirm
Each of these is a real decision Affirm is working through, written so you can make the call yourself — incomplete information, someone who pushes back, no answer key.
Affirm is a buy now, pay later (BNPL) financial technology company that provides point-of-sale installment lending to consumers and payment infrastructure to merchants across the U.S., Canada, and the U.K. The company generates revenue through merchant commissions and consumer interest charges, serving 24.1 million consumers and 419,000 merchants as of September 2025. Amazon and Stripe are among the partners that integrate with Affirm through its partner and merchant APIs. The Affirm Card, the company's newest financial product, marries the functionality of a debit card with access to Affirm's buy now pay later financing options and is one of Affirm's fastest-growing products and an increasingly significant contributor to overall platform GMV. Affirm is also building its first fully-featured ads platform to create a shopping destination for its merchants and customers. Affirm Bank is preparing to launch as a de novo Industrial Loan Company (ILC) that will launch a deposit product, drive loan originations, and bring card issuing online.
Your first one is free. No card.
66 live right now.
The rest
Decide which order owns the payment plan once store orders split
The merchant needs 99.5% of pilot orders and returns matched to one plan within 24 hours. You must weigh a holiday launch across two channels against the cost of an order record that cannot be safely corrected later.
Stop duplicate offers when shoppers return to checkout
Duplicate Pay in 4 offers appear when shoppers return from the payment page. You must weigh a fast integration change against preserving the merchant’s existing checkout return flow.
Match vendor charges before servicing capacity gets squeezed
Unexplained vendor charges are slowing invoice approval while shoppers still need timely repayment support. You must weigh faster evidence matching against the human review required for every figure.
Stop stages without a next step while sellers approve every field
Enterprise retail opportunities are advancing with qualification details split across calls, notes, and Salesforce. You must weigh faster seller preparation against the review needed before a commercial promise enters the record.
Flag suspect arrivals of repayment data before reports reach merchants
Merchant reports can publish before missing repayment events become visible, then require corrections later. You must weigh faster incident triage against the evidence and human review needed before data is held.
Trace partner evidence before the monthly recommendation goes out
The same checkout shift is producing conflicting partner briefs, and claims are spread across SQL pulls and Python notebooks. You must decide what AI assembles, what strategists verify, and how evidence remains usable after handover.
Route connection alerts before checkout failures spread
Connection spikes are leaving some cardholders unable to complete a purchase while engineers reconstruct the path. You must decide what AI can assemble and what an engineer must verify before acting.
Prove cardholders can recover accounts after a regional storage outage
UK cardholders could lose access to account recovery during a storage disruption. You must weigh launch readiness against the recovery evidence needed to protect records.
Route checkout alerts while shoppers wait at authentication
Checkout authentication alerts are taking too long to classify while some shoppers are left waiting. You must decide what AI can assemble and what an engineer must judge before action.
Prove drift in thin-credit approvals is real before holiday reporting
Credit-thin Pay in 4 approvals are shifting in smaller cohorts while overall approvals look stable. You must weigh faster case assembly against evidence a risk manager can independently verify.
Sort reward feedback while cost concerns hide inside it
Early rewards feedback mixes questions about earning value with worries about financing cost. You must decide what AI groups, what PMs verify, and what evidence is strong enough to shape a hypothesis.
Design a first reward for repeat use while keeping financing costs clear
Invited cardholders have no clear reason to choose the card again after a successful purchase. You have to weigh a reward people notice against the cost and disclosures that keep it responsible.
Cluster repayment feedback while PMs still need the evidence
Recurring shopper concerns are being interpreted as different problems across separate research reads. You have to decide what AI can group and what Product Managers must verify before it shapes discovery.
Trace account errors while engineers still verify every cause
Cardholders are waiting longer when settled transactions produce account outcomes engineers must reconstruct from scattered evidence. You have to weigh investigation speed against the evidence an engineer needs before changing a customer record.
Surface installment choices after larger card purchases settle
Eligible purchases over $500 are settling with no clear path to a financing choice. You have to weigh a useful post-purchase flow against exact loan attribution and reviewed terms.
Trace checkout failures while shoppers wait at payment
Marketplace shoppers are reaching payment and seeing the purchase flow stall after product pages load. You must weigh faster cause ranking against the evidence engineers need before changing a live checkout path.
Link merchant feedback to sources so pricing reviews agree
Merchant feedback is being interpreted differently across pricing reviews, slowing decisions about conversion and incremental sales. You must weigh faster evidence assembly against the product manager’s need to verify what the feedback actually supports.
Build deadline search for movers without implying approval or delivery
Time-bound home shoppers leave the marketplace when Search cannot help them narrow options for a purchase date. You must weigh useful planning signals against clear repayment terms and realistic expectations.
Commit a ranking policy for home listings before the campaign locks
Home shoppers leave after repayment details change what the purchase feels like. You must weigh merchant traffic gains against repayment clarity and downstream payment risk.
Clarify repayment cues in Search without implying approval
Home shoppers are leaving product pages after the full repayment terms change their view of a purchase. You have to weigh faster product discovery against an honest view of what financing may cost.
Show monthly payments across products so home shoppers stop leaving
Home shoppers are opening multiple product pages without a clear way to compare monthly commitments. You have to weigh a richer comparison moment against a fast, focused path to checkout.
Protect repayment terms while search traffic surges
Search shoppers are reaching checkout with repayment terms that may no longer match what they saw. You must weigh discovery continuity against committing to a customer-facing handoff that cannot be cheaply reversed.
Match plan previews before shoppers abandon the next step
Mobile product-card previews disagree with checkout terms in affected sessions. You have to restore trust without destabilizing the shared API.
Test a change in plan education without overstating what shoppers owe
Merchants see shoppers view interest-free plans but still choose another payment method. You have to weigh higher plan selection against clear, accurate repayment expectations.
Set guardrails for new approvals while first payments remain unknown
Credit-thin shoppers are receiving more interest-free offers, but repayment evidence will arrive after checkout. You have to weigh broader access against the signals that should trigger a closer review.
Settle the evidence for shopper consent before 500 people use checkout
The first UK checkout will place an interest-free plan beside longer-term financing without an agreed evidence standard. You must weigh a simple purchase flow against proof that shoppers understood and qualified for what they accepted.
Give shoppers control of old sessions without disrupting checkout
Shoppers who see an unfamiliar device cannot remove it without recovery or support. You must weigh fast account containment against signing out the shopper who is still trying to pay.
Predict identity checks before shoppers lose their checkout
Late identity checks interrupt shoppers after they select an interest-free plan. You must weigh earlier fraud signals against unnecessary checkout friction.
Set a readiness bar for launch while contact drivers stay unknown
The first merchant launch has no proven path for the questions shoppers will ask after checkout. You must weigh launch speed against a readiness bar that keeps answers accurate and compliant.
Test Pay in 2 on home carts without raising early delinquency
Higher-value home carts are leaving after shoppers see that Pay in 4 will not cover enough of the purchase. You have to weigh incremental checkout completion against early repayment performance and merchant economics.
Settle one route for UK complaints instead of a manual review queue
UK shoppers are receiving conflicting answers about whether their concern became a formal complaint. You must weigh a clearer route for shoppers against accurate classification and sustainable cost-to-serve.
Unify channel metrics so merchants can compare where plans are used
Merchants cannot reliably compare interest-free plan usage across online and in-store checkout. You must weigh a useful shared metric against governed access and definitions that remain stable.
Build a path for platform merchants to launch without weak integrations
Platform merchants are asking for interest-free plans, yet many never reach their first transaction. You have to weigh scalable partner activation against the readiness needed for durable GMV.
Surface purchasing power before larger carts reach checkout
Higher-value shoppers discover interest-free limits after they have built a cart and started checkout. You must weigh earlier clarity against making a credit outcome sound guaranteed.
Show the date of the next payment without duplicate charges
Shoppers can see a next payment date yet still contact support to ask when money will leave. You must weigh clearer action before a due date against the safety of an automated repayment schedule.
Reconcile vendor hours before the payment-date queue grows
Payment-date callers are waiting longer even as billed staffing hours rise. You must weigh fast invoice approval against evidence that the paid coverage actually reached shoppers.
Renew the home merchant while return disputes keep lingering
Return-related disputes take a median of nine business days, and the merchant’s renewal closes in 21 days. You must weigh a credible shopper-facing commitment against the margin and operational capacity it requires.
Recover Pay in 4 before a home promotion launches
Mobile Pay in 4 completions at a home merchant are down 14% after a checkout update. You must recover financed checkout volume without asking the merchant to undo its redesign.
Commit a safer model while holiday approvals keep rising
First-payment misses are climbing among newly approved credit-thin holiday shoppers. You must weigh approval access against repayment risk before the next model release becomes the live decision path.
Restore Pay in 4 decisions before shoppers leave checkout
Pay in 4 disappears for 3.2% of shoppers on one slower decision path, even though a retry usually works. You must weigh faster decisions against preserving the signals that protect repayment quality.
Set the hardship boundary while repayments keep moving
Australian Pay in 4 shoppers seeking hardship support saw different outcomes while scheduled payments continued. You must weigh broad customer protection against a filing position that can withstand scrutiny.
Restore Pay in 30 disclosures after shoppers accept too soon
Some Australian Pay in 30 shoppers receive their terms after accepting the plan. You must protect informed agreement without interrupting a working checkout path.
Negotiate one rate before the retailer assigns its rollout slot
A retailer wants one commission rate for fashion and furniture even though their economics differ. You must weigh a fast group-wide launch against terms that still hold across both brands.
Move merchants from the activation email into integration before review
Only 22% of eligible SMB merchants have activated Pay in 4 after launch. You need to weigh a scalable partner motion against the live guidance merchants are responding to.
Set net-plan rules before holiday reports replace the old extracts
Partial returns are producing different net Pay in 4 totals in merchant reports and merchant records. You must weigh a durable published definition against the cost of changing historical reporting.
Clear the duplicates in merchant reports instead of filtering by hand
A source event is causing 2.8% of Pay in 4 plan rows to appear twice in merchant reports. You have to restore accurate counts without changing valid historical reporting.
Choose a servicing plan for two quarters while repeat contacts hit 18%
Repeat contacts have reached 18%, and compliance QA defects are rising in the same queues. You must weigh lower cost-to-serve against dependable, compliant answers when the allocation cannot be cheaply reversed.
Choose which plans the partner locks in before its window closes
Pay in 4 is lifting completed purchases, while higher-value partner merchants are losing margin. You must weigh purchase completion against the economics of a configuration that is costly to change.
Restore Pay in 4 selection after checkout order changes
Pay in 4 selection fell from 16.8% to 14.2% after the payment-method order changed at affected merchants. You must weigh a fast placement experiment against the need to protect checkout completion and unit economics.
Commit a complaints workflow before the old intake path closes
Payment-plan complaints are being classified differently, leaving some shoppers without a recorded case or clear next step. You must weigh a fixed transition date against the proof needed for compliant, consistent handling.
Align identity-check guidance while shoppers receive conflicting answers
Shoppers affected by new identity-check reason codes receive different explanations for the same outcome. You must weigh speed of agent readiness against the need for one verified, compliant explanation.
Hold fraud under 0.45% while changed-device shoppers get offers
Recently changed-device shoppers are being declined more often, even though confirmed fraud has not risen with them. You have to weigh restored approval access against a fraud control that may be catching risks the current loss rate has not revealed.
Validate a model shift while holiday shoppers keep checking out
A purchase-history signal is gaining weight for credit-thin shoppers while the overall approval rate remains steady. You have to weigh a fast, observed mix change against evidence of inconsistent underwriting decisions.
Move first-time shoppers from 61% to 68% without hiding schedules
First-time approved shoppers complete at 61% when several plan choices appear, even as returning shoppers finish more often. You must weigh a faster path to confirmation against proof that shoppers understand their repayment schedule.
Clarify Pay in 2 dates as shoppers leave the terms review
Pay in 2 completion falls to 71% after approved shoppers open the terms review. You have to weigh a faster confirmation flow against making both payment dates unmistakable.
Move card transactions to TiDB without risking data divergence
Projected card transaction growth will push recovery beyond its tested safe window. You must weigh permanent data consistency against the availability risk of waiting.
Stabilize card authorizations when database failover strands connections
A database failover leaves cardholders unable to complete purchases for more than four minutes. You have to weigh faster reconnection against confidence that each authorization is handled exactly once.
Lock down recovery sessions while shoppers replace their phones
Some recovered accounts keep a valid session on an old device while others fail on a second attempt. You must weigh permanent session invalidation against shoppers reaching checkout and their upcoming payments.
Clear stale connections when checkout authentication slows
Checkout authentication reaches 400 milliseconds too slowly for 1.4% of requests after backend recovery. You must weigh a fast connection fix against the risk of disrupting valid checkout sessions.
Decide how plans are chosen before the release window locks
Fourteen percent of affected monthly-plan cardholders miss a payment within 60 days. You have to weigh accessible financing against a change that may permanently narrow some shoppers’ choices.
Raise plan conversion from 54% to 66% without changing credit terms
Forty-six percent of cardholders who reach the payment summary leave without converting. You have to weigh faster completion against a clear view of what the plan will cost.
Fix gaps in repayment schedules until the old partner path retires
Approved Pay in 4 purchases are reaching the Manage tab without a next payment, and some loan events lack a repayment schedule. You must weigh partner checkout continuity against a permanent contract that keeps servicing and reporting complete.
Fix month-end Pay in 4 dates without changing existing schedules
Some month-end Pay in 4 plans show a next-payment date one day later than their terms. You have to correct the calculation without changing existing schedules.
Commit the loan match before delayed details rewrite repayments
Wrong-loan corrections are running at 42 per 10,000 settled card purchases. You must weigh stable settled accounts against correcting late-arriving transaction details before the cutover becomes permanent.
Stop duplicate refunds before balances fall twice
A retry is causing 0.8% of settled refunds to reduce cardholder balances twice. You have to protect fast refund recovery without letting a delayed response create another adjustment.