Credit Card Bill Payments: Fair game in post-BBPS world (#75)
Banks’ conflict with repayments, complications behind CRED’s big bet, and CCBP's evolution into a plug-and-play feature
Welcome to the 75th issue of Unit Economics. In this article, I talk about how CRED transformed credit card bill payments (CCBP), and how the BBPS-mandate has made it a level-playing field for all fintech companies.
I have been mulling over this topic for the last few weeks, and will also spend time talking about why the CCBP experience became the problem that it did. Dive in!
Credit cards and the revolving credit model
On the face of it, an interest-free credit line, which also pays you back on use, should not make for a good business.
Yet, with credit cards, banks have found a way to turn it into one of their most profitable products. It’s a story in two parts.
1950s: All about loyalty and the MDR
In the earliest days of credit cards, banks made almost all of their credit card money by taking a cut from the merchants who accepted these cards. The commission used to be as much as 3-5% of the purchase value.
Merchants were okay with sharing as much since credit cards enabled them to make even more money off the customers. How? Because every merchant - be it a restaurant, a grocery chain, or an apparel store - wants their customers to do more repeat orders and with higher average order values. But the competition makes it challenging to maintain either.
With credit cards, banks realised that customers (a) place a higher number of orders, and (b) have higher average order values, than if they were spending without cards. That is, credit cards were enabling customer loyalty for the partner merchants.
For a merchant, the loyalty justified the 3-5% costs. And for banks, this margin was sufficient to recoup the costs of rewards, credit lines, and other operations. In fact, they could make a killing off whatever was left.
But this model, where credit cards - as charge cards - were designed around merchant discounting, could not maintain the same economics for long.
By the end of the 1950s, the number of card issuers and carded customers had grown. And every reasonably sized merchant had started accepting credit cards. This meant that the impact of a credit card on a customer’s behaviour was not as obvious now to the merchants.
The 3-5% fee started to pinch. Merchants began pushing back, arguing that the costs were too high. Many, who could get unionised or had some monopoly power, started negotiating much lower rates. Gradually, the average merchant discount fee fell to ~2-3%. Which was still significant. But clearly not enough to sustain the card programs.
Banks were forced to innovate.
1960s and later: Introduction of revolving credit
In the 1960s, banks came up with what is today known as the revolving credit model. In this, cardholders were allowed to carry over their outstanding balances from one cycle to another in exchange for an interest cost.
Over time, this interest income grew to contribute a larger share of credit card revenues in most portfolios. Particularly in the last couple of decades, where the interchange fee, and in effect the merchant discount rates (MDR), have been more strongly regulated. In the UK, for instance, the interchange on domestic credit card transactions is capped at 0.3%. In Australia, it’s capped at 0.5%.
In lower-income countries, in addition to the interchange regulations, customers show little willingness to pay any annual or joining fee for the cards. Making interest income more important for the card businesses.
The nature of interest-based income was not new for banks. But the shift away from the merchant fee to revolving credit changed how banks ran their credit card business. This had multiple second-order effects:
Underwriting adjusted to intake the less creditworthy
The early charge card model pushed out anyone who couldn’t pay on time. This meant that the underwriting models were built to approve only the most creditworthy customers to the programs.
That changed with the revolving credit model.
With the card P&L dependent on the interest income from customers who would revolve their balances, the underwriting policies had to become more accommodative towards riskier customers.
In fact, they had to be intently designed to intake a healthy mix of those who would revolve their balances consistently, customers who we call revolvers today. This made credit card underwriting a more sophisticated exercise.
Revolving credit model forced the high interest rate pricing
Banks needed a healthy revolver mix to subsidise the rewards for the entire portfolio. But with more revolvers, the risk of defaults also increased.
Which meant that the banks could not keep the revolver mix more than a certain percentage of their overall portfolio. Over time, through trial and error, Indian banks figured out that number is somewhere between ~30-40%. That is, the majority of the customers should still be non-revolvers.
But this led to some very odd economics. Since the revolvers had to shoulder the interest-income burden for the portfolio, the interest rates on revolving balances started to look unreasonably high. That is why, for most credit cards, the annualised interest rate on revolving is at least 30% and sometimes more than 50%.
This is even though these revolvers can get loans at probably half the rates elsewhere, even with the same banks. This is why, even though consumer courts may debate the ethics of charging such steep interest rates, it's hard to justify a business case against it given the economics.
Instead, and rightly, a lot of effort from the regulators is directed towards driving more friendly disclosures from card issuers (Reg-Z ‘68, RBI, CARD Act ‘09, etc.), rather than in controlling the rates themselves.
Banks’ conflict of interest in improving the repayment experience
For the longest time, paying your credit card bills wasn’t a very pleasant experience. You would have to make sure you remember the due dates and don’t miss any card statements.
Then, to pay the bill, you would either log in to your bank’s website. Or find an online payment gateway like BillDesk, where you would have to enter the 16-digits, CVV, expiry, and your name. You’d usually verify it thrice after entering, because of how anxiety-inducing the experience was. And then once you had made the payment, you would come back to your bank to verify if the amount had been credited to the card or not. Which could take anywhere from 1 to 3 days. Exhausting!
Banks knew this. But with the revolving credit model, they had little incentive to change the status quo. Today, while the quality of banking mobile apps has improved — not much else has changed. For instance, payment reminders are hardly standardised, despite the many efforts of the RBI.
Customers had no alternatives to solve this repayment problem.
That is, only until 2018. When CRED completely changed the credit card bill payment experience. It made a lot of noise with its cashbacks and offers. But what is usually missed in the noise is how transformational its product was for the credit card industry. And how well it solved some very tough problems to make the repayment experience what it did.
Let me break down the experience into three parts: card aggregation, bill discovery & management, and bill payment.
Card aggregation and its impact on the credit card industry growth
The Indian credit card industry is lifted entirely on the shoulders of 50-60 Mn white-collar individuals. Over time, it has meant that banks end up selling more credit cards to already-carded customers than to the never-carded customers.
This is why, on average, a carded customer in India today holds ~2 or more credit cards. And often each is from a different bank.
Earlier, the gaps in repayment experience anxiety increased your chances of involuntary revolving. This likely influenced many to restrict their card holdings to not more than 1-3 credit cards.
But with CRED neatly aggregating all of your cards on a single app – which in itself was a first for the industry – an average user could find more comfort in holding a higher number of cards.
As a second-order effect, I’d imagine it to have influenced the decision of hundreds of thousands to apply for another credit card and contributed to the overall industry growth.
Complexity of a bill discovery & management machine
Aggregating cards is only one part of the experience, and comparatively, the simpler one to solve. Once the cards are aggregated, the more challenging problem, however, is identifying and ensuring that no upcoming bill payments are missed.
The ask is straightforward. But since banks share bill information with no one but the customer and there are no common servers to get it from, it meant that CRED had to solve three things very well:
Identify and validate the card details - number, CVV, etc. - against the customer, an operation that only became more complicated with the poorly planned tokenisation guidelines from the RBI,
Scrape a customer’s SMS or emails to find the bill against that card, which requires custom OCR’ing logic to be built for each card program. To understand the extent of effort in this, note that there are 30+ card issuing banks, and SBI Cards - an issuer - alone runs 60+ card programs; and
State-manage the bill against each card, which requires constantly tracking the due amounts & dates against any potential part or full payments already made on CRED or outside, and building state-specific reminder journeys to make sure customers don’t miss any due dates.
Optimising success rates, credit time, and building payment trust
Once a card is identified, and verified, and a bill recognised against it – the repayment leg starts. With the higher ticket-sizes on credit card bill payments, the platform is expected to be that much superior in their payment experience to overcome any trust deficit. Especially when any failures or delays can have consequences on your credit profile.
To a CRED, the payment leg was broken into two async parts: (a) pay-in: accepting a customer’s payment through their preferred mode, (b) pay-out: once accepted, processing its transfer directly to the customer’s card account via a fund-transfer system.
Till 2022, when the card details could be stored by a merchant like CRED, this was, say, a medium-difficulty problem to solve. You could partner with a Razorpay, Cashfree, or a PayU and handle both legs. The bigger challenge until then was in (a) managing customer expectations around the credit time of the payment towards their account (usually took T+1 to T+3, especially when processed over weekends), and (b) ensuring a high payment success rate by optimising routing across PA/PGs. To CRED’s credit, it went two steps further in optimising both by building direct fund-transfer rails with multiple banks.
In 2022, when tokenisation of card details was mandated, and merchants had become blind to a card’s details, managing pay-outs became increasingly complicated. Since either the issuers or card networks had to get involved in processing card tokens. Over time, the tokenisation ecosystem also matured, but not without a lot of effort and time having spent on it already.
All-in-all, it took a mountain to climb for CRED to solve the repayment problem. But it paid off. By 2024, CRED was processing credit card bill payments for ~13 Mn+ very active customers. Which is roughly 30% of all of the country's carded customers. And achieving market-leading M12 retention rates (60%+), with very little competition.
It did not do that out of charity, of course. Credit card bill payments by itself provide no revenue for CRED. But when used as an acquisition pitch, it acts as a filter to acquire carded customers, which – given how banks run their card underwriting – are expected to be amongst the most affluent Indians. Ones you would categorise under “India 1”.
CRED’s thesis was that the more affluent the customer base, the more meaningfully you could monetise them eventually. Especially in Indian consumer payments, where hardly anyone focused on the long-tail or the mass-market has proven good unit economics.
Today, this affluent-first strategy is largely accepted as the more preferred strategy for a consumer business in India. And CRED’s success in building a $300 Mn+ topline within 5-6 years is a big reason behind the narrative shift. For fintech companies, it has also given them confidence on credit card bill payments as a potential acquisition play.
But since matching CRED’s repayment experience seemed a complicated and costly venture, not many tried for a long time.
This changed last year, when the RBI characteristically mandated that all credit card bill payments (CCBP) be processed over Bharat Bill Payment System (BBPS). What did it mean?
BBPS-shift: Turning a complicated product into a feature
In doing so, the RBI centralised a lot of the core capabilities that a CRED had spent years building.
You did not need the full card number or a token to identify and link a card anymore. On BBPS, it could be done simply with the last-4 digits of the card and the mobile number.
You did not need to scrape emails or SMSes to find bills. You could now get it directly from the banks that are registered as billers on BBPS.
You also did not need to manage multiple PA/PGs for pay-in or pay-outs. You could simply integrate with a customer operating unit (COU) on BBPS, who would handle everything from authorisation, settlement, and disputes for you.
Effectively, the BBPS shift lowered the barrier to entry for anyone who wants to offer CCBP, turning what was a complicated product offering to a plug-and-play feature.
The gap between a BBPS and a CRED’s capabilities reduced fast as the critical mass of banks onboarded onto the BBPS very early (97%+ coverage today), with the exception of American Express, Standard Chartered, and certain co-brands.
Nine months in, and there are already early signals into how others in fintech are thinking. Below, I highlight that and also take some bets on how this may unfold:
Large-scale fintech companies want a piece of the CCBP pie
Today, when monetisation and customer quality are continuously questioned, all consumer-focused fintech companies are heavily incentivised to acquire through CCBP. And we are already seeing early signs of how those capital-rich ones are approaching it.
PhonePe, Navi, Jupiter – for example – have doubled down on CCBP and given it high-attention in the first viewport of their apps. Some, particularly Navi, are offering monetary incentives as well to drive the switch.
While still early days, for large fintech companies, I expect CCBP to (a) be placed more distinctively than other bill categories on the app, (b) be used increasingly on top-of-funnel branding for acquisition, rather than just as a cross-sell to the existing base, and (c) have budgets cut out separately for monetary incentives to the customers. We can expect this to get competitive soon.
Unlikely for a CRED 2.0 to exist
In the post-BBPS world, it appears unlikely that any platform could emulate what CRED did in 2018, and rely solely on CCBP for platform acquisition.
I say this primarily because (1) more apps are pitching the same offering to customers today, which will make it a more expensive affair, (2) there is limited product differentiation, if any, that you could bring to the functionality due to the maturity of CCBP biller ecosystem already, and (3) there are multiple headwinds across each cross-sell opportunity that CRED utilised - from personal loans to rent payments, making it a tougher sell even to the VCs.
CRED, to their end, was smart in diversifying away early from CCBP and towards UPI for acquisition, with a lot of their CCBP efforts today focused on retention than on acquisition.
Tail-end of CCBP volumes to be spread across fintech companies
CRED’s differentiated positioning and experience will help it retain the majority of the 13 Mn+ active base.
But I believe that there would be a pre-existing 5-10 Mn carded base engaged on UPI-first apps such as PhonePe, Navi, or super.money, who could switch towards these apps for CCBP.
Realistically, I would argue that each such platform could get a 1 Mn+ engaged base on CCBP within the next 18 months, given the scale of their UPI today. If they could dedicate themselves to building a good experience from card discovery to bill payments. The only risks, I’d assume, would be of their own distractions!
If you want to share any views or talk casually on the topic, please reach out to me by email or LinkedIn. Until next time!
Thank you for deeply researched and insightful Article!
One question though - Now that most of the Banking Apps offer users the option to pay credit card bills directly, why would someone use a different app(Cred) for this purpose?
..as multiple players vying for a finite market, so would we be seeing times of differentiation and consolidation? Does this also mean we may see partnerships between a platform and credit card companies?