Spend management and credit are likely to become one business. How carefully a company approves and tracks its spending is strong evidence of how it will treat borrowed money.
Every few months, a question crops up in the
public domain: When will business credit in India have its UPI moment? That
might be the wrong question.
Consider a mid-sized distributor. He collects
from retailers on UPI, pays suppliers from his phone, and moves money through
the business in seconds. His working capital tells a different story. The
cash-credit limit may still be based on an old balance sheet, reviewed
annually, secured against property and backed by a personal guarantee. The
bank’s day-to-day view may still depend on a monthly stock statement.
In short, while his cash flow reflects the
realities of his business in 2026, his credit-worthiness could be based on
20-year-old data.
Ironically, as
India processes more than twenty billion UPI transactions a month, empowered by
an infrastructure that has become almost invisible, business credit still lags
far behind.
UPI worked because it sits where decisions are
made: At the counter and on the phone.
When a business plans for a loan, it collects
papers, applies and waits – its financial reliability having little to do with
the health of its business.
It’s not that there has been no progress in
speeding up loan disbursement. Now we have account aggregators, the RBI’s
Unified Lending Interface, e-invoicing, TReDS. Yet most of these systems fetch
the same documents that banks rely on – the only difference being a reduced
waiting period.
The lender then summarizes
these past records into one number and leaves
it alone for a year. Three clocks run at different speeds: the business changes
daily, its data refreshes monthly, its limit is reviewed yearly. Until they
converge, credit will stay slower than money.
The cost is real: the MSME credit gap is usually put at Rs 25-Rs 30 lakh crore. Many small firms also wait 60 or 90 days to be paid, despite the 45-day legal limit for micro and small suppliers, while paying their own vendors on time. The businesses that most need working capital end up lending it out.
Lend against the flow, not the company
Card issuers learned decades ago what business-lending
is still learning. With every swipe, the issuer makes a credit and a fraud
decision before the terminal responds, without an audited balance sheet. It
works because the question involves one transaction, as it happens, with the issuer’s
full knowledge.
Wherever Indian credit has become fast,
lenders have made the same move: underwriting a
flow they can see, not the whole entity.
Dealer finance lends on the basis of a manufacturer; marketplaces do so based
on platform sales; TReDS relies on a large buyer’s accepted invoice, and RuPay
credit cards on UPI cater to the kirana store. The unit of underwriting must
move from the company to the transaction, to the invoice, and the spend.
Spend data can give lenders a clearer picture
of how a business operates. When a company uses corporate cards and a spend
management platform, lenders can see where it spends, how regularly it pays
vendors, and whether spending stays within approved limits. This real-time behavior
can often reveal more about the health of a business than an annual balance
sheet.
Lenders have a fair objection. Collateral
answers two questions: Will this business repay, and what can be recovered if
it doesn’t? Cash-flow data is good at answering only the first question.
For lenders, control is essential. Credit that
can be spent only for a defined purpose, with approved suppliers and within set
limits, is hard to divert, especially when it is visible as it is used and its
flow can be restricted within hours. The lender doesn’t need a building to
seize in two years on account of default. It simply cuts off credit supply.
This is why spend management and credit are likely to become one business. How carefully a company approves and tracks its spending is strong evidence of how it will treat borrowed money.
Paper was never the safety net
More documentation does not automatically mean
less risk. Some of the largest banking frauds came through loans supported by
extensive files, audited statements and pledged assets. But faster, behaviour-based
credit is not safe by default either. Borrowers can manufacture good behavior
before drawing down aggressively. Automated models can scale mistakes quickly.
If many lenders rely on similar data and models, they may also reduce credit
together when the cycle turns.
Speed amplifies good judgement and bad
judgement equally.
The next stage needs discipline alongside digitisation:
transaction-level fraud checks; limits that can come down as quickly as they go
up; explainable models; clear human accountability; and consent that borrowers
genuinely understand.
The guardrails are largely in place; the
harder shift is in how credit is designed.
Continuous, not annual, underwriting: A business changes within weeks. Credit should adapt by reading
meaningful changes in cash flows, collections, invoices, and spending.
Credit inside the transaction flow: When payments, spend controls, and credit sit together, a lender sees
how borrowed money is used.
Reputation built from business behaviour: How consistently a company pays suppliers, how fast customers pay it,
and how often invoices are disputed can form a financial identity, usable with
consent.
Speed that is reversible: A system that raises exposure quickly must also reduce it just as fast.
Payments became instant because controls evolved alongside speed. Credit too
needs the same process.
The goal is not an instant business loan. It
is a financial system where a good business does not have to repeatedly prove
what its transactions already show. The better measure is how fast a good
business gets the capital it has earned, and whether the system can still tell
a good risk from a bad one at that speed. Small businesses already produce evidence
in every payment. What remains is for lenders, regulators, and the companies building
for them to start reading it.