One 97 Communications Ltd

NSE:PAYTM

Thank you for joining, and welcome to Paytm's earnings call to discuss our financial results for the quarter ending on 31st of December 2024. We will start our call with a Q&A after introduction to the management. [Operator Instructions] From Paytm management, we have with us Mr. Vijay Shekhar Sharma, Founder and CEO; Mr. Madhur Deora, President and Group CFO; and Mr. Anuj Mittal, SVP, Investor Relations. A few standard announcements before we begin. The information to be presented and discussed here should not be recorded, reproduced or distributed in any manner. Some statements made today may be forward-looking in nature. Actual events may materially -- differ materially from those anticipated in such forward-looking statements. Finally, a replay of this earning call and transcript will be made available on the company's website subsequently.

We will start our Q&A now. [Operator Instructions] The first question is from Pranav Gundlapalle from Bernstein.

Two questions. One is in terms of your GMV, what percent of the transactions would be cases where Paytm is at the merchant side versus those where the app is on the consumer side. I'm just trying to reconcile the consistent growth in GMV versus the loss in overall UPI market shares. Just trying to see where -- if there's a difference in the consumer market share versus merchant market share. Sir, that's the first question. I'll come back for the second.

Thanks, Pranav. We actually have a little bit of disclosure in the back of our earnings release, which talks about number of transactions and it says merchant transactions. So you can see in December '24, we had 1.1 -- sorry, we had 1,000 crore merchant transactions. And we had 1,232 total transactions. Now it is the case that in merchant transactions, some of them might also be our consumers. But if you just look that how many transactions we are doing for merchants, that is in the disclosure in the back. I think it's fair to say that if you take a year-on-year view, then the merchant side of the business has grown where as the consumer side of the business, even adjusted for discontinued businesses has been flat or slightly degrown.

Understood. That's helpful. And second, could you share some color on the RuPay on UPI transactions and what it could mean for the payment margins if it increases significantly from here?

Yes. So as I think there is a gist of the question on RuPay on UPI, we do make MDR unlike bank-linked UPI. But it is still a very small percentage, growing fast, but it's still a very small percentage of the merchant transactions right now. So as that grows, that does add another monetization lever for us.

Pranav, I'll also add one extra line that we get extra bps, which is significant extra bps, and it does include payment margin between other instrument versus RuPay credit card. And we do believe that credit card running on QR gives us, as an acquiring side, much larger revenue than on the consumer side because consumer side, you don't get larger revenue, even though there is some bps that is base to the consumer app also. But acquiring side gets disproportionately more valuable and useful there, number one. Number two, Pranav I did this survey. I asked my team to do a survey that we saw this new -- you would have seen a couple of million new customers added.

I said, why did you sign up so fast without any marketing and advertising. And our QR being present, our brand being present on ground was one of the big -- actually #2 trigger. First was literally that they wanted to try out Paytm once again before they got to know in the press. And then second was that I saw the QR code. So our acquiring will add it to the consumer acquisition is what our belief is. And that is why we will aggressively expand on our merchant acquiring because that gives us an edge in consumer also. So there is a little bit of flywheel that we are seeing here. And that is probably one of the reasons that we didn't see a large number of exodus either and vice versa.

Next question is from Sachin Dixit of JM Financial.

So my first question is basically on regards to DLG and the mix within merchant loans. So we did some maths around the AUM numbers that were shared. And somewhat we are coming up with almost 80% of merchant loans in this quarter came from DLG sort of arrangement. Are we directionally in range? Secondly, is this the right steady state or number or mix from DLG that we are hoping for? Or we are trying to optimize for a certain number there?

That number is in the ballpark, Sachin. And give or take a few percent, I think that's probably right. Like we have said, we think DLG is a good model. Non-DLG is also a good model. So relative to each other, it's -- we are quite indifferent. If a partner thinks that it is easier for them to do business of lending to merchants in a DLG model or if they can do more business as a result, then we are quite open to doing DLG-based business with them. So we don't give a target number in that sense. But yes, ballpark, that -- it should be in that range.

And you are seeing interest from more lending partners on the same model, I'm guessing?

Yes. So that's right. So we're seeing more interest from lending partners in general across MCA as well as personal loan. MCA, in particular, has been a product that has done consistently very, very well for us and for our partners. So we are seeing more interest in that. And yes, within that, there are some partners who are more interested in DLG models and for -- there are some partners for whom it doesn't really make much of a difference.

Sounds good. My second question is with regards to the plan in international. So I see that there are some subsidiaries that we are looking to put in place. Is the plan there to do largely UPI sort of enablement, some of these countries also have UPI extension that has happened? Or we are planning to build an entire merchant ecosystem sort of thing like we have in India in these countries?

Vijay?

Yes. Sachin, I'll take that. And the intent here is to first go on the merchant side because merchant side business model is a very long-term business model and every economy and geography that I'm meeting different, different senior executives in either in Central Bank or government, love it because SME credit is missing everywhere. So if you can solve for payment and solve for future forward receivable, that business model is a template in my opinion, in our opinion, technology-wise, capability-wise, we've demonstrated at scale. So our primary plan will be that.

Next question is from Alok Srivastava of UBS.

So I have this question on DLG model that ECL number that we have given, I mean that number has come down to 4.5% to 5% versus 4.75% to 5.25%. So how are we measuring these numbers? We are measuring these numbers or lenders are sharing these numbers?

These numbers are based on our collection data. So there's a couple of things, right? So one is when the lenders put a program in place, they have in their program an expected credit loss. And in our conversations with lenders, we ensure one of the questions that we obviously touch base on is whether the program is within the bounds of their expected credit loss. But the specific data that we put out, including the bucket-wise resolution, is based on our collection data. As you know, in merchant loans, we help our partners with vast majority of collections from the -- because these are active merchants of ours. So basis that, we have a sense of how this is trending towards expected credit loss.

So Madhur, sequentially, the decline would also be because of DLG that we are doing? And earlier, it was largely the normal one that we were doing? Or sequentially, there has been change in some underwriting standards and so on?

So DLG does not affect the ECL -- so just to clarify. So ECL is the [indiscernible] of the portfolio, where our loss guarantees are versus where the partners' losses and provisions are is independent of how ECL is measured. The DLG book is doing at least as well, in fact, slightly better than the rest of our book. And that's just -- sometimes some books do better, sometimes some books do slightly worse. So there is a -- I wouldn't necessarily draw a pattern there. And the reason for improvement is generally collection efficiencies.

Obviously, with a lag, it is also indicative that your underwriting is getting tighter. I would just also add that at the beginning of 2024, we had some sort of a little bit elevated churn amongst our merchants. So that may have affected some of the ECL as well. But as the business has stabilized and growing and our partners -- our merchant partners are growing very well with us, we are seeing improvements in collection efficiency as well as a lot of the work that we have done to bring even greater amount of sort of efficacy to our collection practices.

Sure, Madhur. That's very helpful. I have a couple of more clarifications on FLDG. So FLDG number that we shared last time that we are doing less than 5%. Does that still hold true? That's one. Secondly, all the DLG costs, which are upfronted, are they fully captured in the other direct cost? And also if there is any unutilized DLG, which may happen a year from now, where will it show up in P&L? Will it be adjusted against cost? Or will it become part of revenue?

So first question correct. On a blended basis, our DLG cost is significantly lower than the 5%, and it has broadly remained the same as last quarter. On your second question, yes, we -- so far, all the DLGs that we have given, we provisioned them immediately in that quarter, and that is in other direct expenses even though the DLG invocation by the partner is obviously with a lag as and when GCLs hit. And the third is that if we do have a situation where we have provisions, which are getting -- which are not getting used, then it would be a reversal in that cost.

Next question is from Piran Engineer of CLSA.

Congrats on the quarter. Just firstly to clarify on one of the previous questions. Is it -- is DLG given only on merchant loans or on both?

There is a small amount of DLG that we give in personal loans as well with select lenders, only in the collection model I would say, not in the distribution-only model.

Correct. Okay. Okay. Fair enough. So it was 80% of just the merchant loans are disbursed through DLG?

That's right.

Okay. Fair enough. And just secondly, if you can give us a sense of how many merchants currently have a loan outstanding from you?

It is -- broadly it's 5 to 6 lakh merchants who have a loan outstanding. So it's roughly been 4% to 5% of our device space.

Okay. Okay. Fair enough. And that can go to what sort of a number in your opinion, this 5 to 6 lakhs. Yes, I'm just trying to think about the medium-term growth rate here.

Yes. We -- in the sort of, I would say, 2 to 3 quarter -- 2- to 3-year period, I think we definitely have an opportunity of increasing the percentage penetration to maybe 10% to 15%. One of the things that I should point out is that, that is only 1 of the 3 drivers of the overall growth in merchant loan volumes and revenues. The other 2 drivers are that over time, the ticket sizes have increased and they continue to increase. So I remember 3 years ago, this number was about INR 1.10 lakhs. Currently, the number is about INR 2 lakhs. And part of that is because the value per merchant has grown in Paytm because the merchants obviously are doing a lot more digital payments today than they were doing 3 years ago, especially our best merchants to whom the loans are sort of qualified for.

And the second is a large percentage of the loans are repeat loans, where our lending partners have greater confidence because that merchant has already shown good behavior of taking a loan and repaying the loan. So that's the second driver. And third driver is obviously the growth in the base of merchants, which has obviously grown by -- basically doubled in the last 2 years, and we still see a huge runway ahead, right? So -- and over the last couple of years, we have actually kept the penetration percentage relatively flat despite the fact that the volume of disbursal has gone up very, very meaningfully. So we're going to continue to be cautious. We're going to continue to have tight white list. But I would say over the next 2, 3 years, we should -- we have an opportunity of taking this to north of 10%. And I think the long-term opportunity is probably even higher than that.

Got it. And Madhur, just when you increase the ticket size, does the tenure also correspondingly increase? Or has that remained at the 12 to 18 months...?

The tenure decision is independent of whether you're increasing the ticket size. So if you have more confidence on the stability of the merchant, then the partner may increase the tenure as well, right? So yes, it is the case that for select merchants for a repeat loan, the tenure might have gone up, but you don't increase the tenure just to increase the loan size. You look at how much you think you can collect per day from the merchant. And that is one of the variables that goes into the loan size. And then you look at how stable the merchant is with you and that sort of decides kind of what the maximum tenure you are willing to do. And those 2 things sort of effectively end up in terms of what is the maximum size of loan that you would be willing to do.

So those are slightly independent. The tenure has not grown very much because, again, that is a conservative assumption that we take, which is that we don't want to increase our loan sizes and loan book just by increasing tenure and then fundamentally changing the nature of what we are trying to do here. If the merchant is a good merchant, they can take a loan 12, 13 months later, they would have repaid that loan. And at some point towards the back end of that journey or after repaying the loan, they're eligible for another loan, right? So the merchant with good behavior can have access to credit instead of saying, hey, they give me a super long loan.

Got it. Okay. Fair enough. And just my last question, just a clarification. On Page 6, you mentioned that your payment processing margins was comfortably above the guided 3 bps margin, but sort of when I just assume, say, INR 90 or INR 100 rental income per device, the margin -- the residual margin comes to about 2.5 bps. So what am I missing here? Is my rental income assumption incorrect or what?

While we can go through your math off-line, if you don't mind. But yes, your rental assumption is slightly higher than what our actual is, but it still doesn't quite get me to 2.5 bps, so I can -- Anuj or I can look at it...

Fine. We'll take it offline. Yes, yes that's fair. Cool, that's it from my end.

Next question is from Shubhranshu Mishra.

Two questions. The first one is around the associates, various associates, we have someone in Ivory Coast, in various other African countries. So just I was curious, why do we have so many associates where probably we're not in business right now? Second is this particular international expansion, what is the total market size that these countries have and what kind of market share are we targeting there?

On the first question, we do have many subsidiaries actually, direct or step-down subsidiaries in Middle East, Southeast Asia, South Asia and Africa. Nearly none of them have to do with Paytm's business. They have to do with the old One97 business which is a part of what we disclose as marketing services, where we provide services generally to telecom operators. And this is a business that One97 has been doing for, call it, 20 years. And in many of these countries, there's a requirement that in order to build those local telecom operators, we have to have a local subsidiary. We are in the process of rationalizing these number of subsidiaries because some of these were set up many, many, many years ago, and we may have little or no business in some of these countries. So over the next 3 to 6 months, we'll look to see if we can reduce these number of subsidiaries.

In terms of international expansion, we have mentioned subsidiaries in 3 specific -- sorry, Pranav, could you just put [indiscernible] on mute? Thank you. So we have mentioned 3 particular subsidiaries, 2 in Middle East and 1 in Southeast Asia. We think that the opportunity in Middle East and Southeast Asia is massive, and many of these countries are -- need more and more solutions and distribution in the merchant ecosystem. We have been able to build great technology at very low cost in India, and we think we can service the merchants here really well. But we are going to, like we have said, explore select opportunities. Generally speaking, opportunities in countries where there is a large local market.

And so we should also say that it is very, very far ahead. So don't misunderstand it to be showed up in a quarter or 2, for sure because these are entity formation, license application, operating. So let's just please understand that it will take some time.

Yes, I should mention that we have said we are exploring these things. Setting up a subsidiary is -- has a long lead time from setting up a subsidiary to, in some cases, getting licenses and then eventually having products launched and merchants signed up and then starting to generate revenue and profit. The good thing is that these are largely B2B businesses. So they don't have large upfront spend, which is why we have said that to begin with, we are saying up to INR 20 crores of investment and -- in each of these markets, which is also a bit of a placeholder number. We don't really think that we need large amounts of money in these countries.

Sure. Can I just squeeze in one last question?

Sure.

Sure. Sir, and what are our targets for personal loan disbursement on a monthly basis and merchant loan disbursement on a monthly basis for FY '26?

We're in the process of concluding this, but I could answer that directionally. See on personal loan, it is very market dependent, if I had to be honest, because we absolutely respect the fact that our partners are being very cautious through the cycle. And this is unsecured digital PL and the partners want to just make sure that this reflects their view on the credit cycle and -- especially at this point in the credit cycle, one should be cautious. So I would say, honestly, it's hard to say where we exit, for example, if you said March 2026, right, what is a good number, what's not a good number. I think the way we are measuring this business right now is how many partners are doing meaningful business with us.

So that's what we are trying to drive at this point in the cycle. And then when the partners have a more growth mindset in terms of their PL book, we think we can be a beneficiary of that. On the merchant loan side, we expect continuously steady growth. Currently, we are doing -- currently, we are already higher than where we were in January, and we expect continuously steady growth, largely because of what I said in response to Piran's question earlier, I think it was Piran's question where we talked about medium-term penetration and there is upside as well as the base is growing and the ticket sizes continue to inch upwards on a blended basis.

Next question is from Anand Dama of Emkay.

Congrats on a great set of numbers. My question again is related to PL disbursement only, right? Do we expect that to bottom out soon that now you have started offering DLG and PL as well. In FY '26, you may not give a number, but should be better versus FY '25, if things go positive?

Yes, we definitely expect -- I mean, unless there are big changes in -- and if FY '26 -- unless the FY '26 macro looks worse than FY '25, we don't expect -- we do expect that FY '26 will be better than FY '25, and especially given that we are going to have additional partners going into FY '26, we do expect an increase in this business and a rebound from the lows that we saw earlier this year.

Okay. And the DLG and the PL business could be on a higher side versus the merchant loans, is that safe to assume?

Not necessarily, no. I think within the PL business, the DLG is on a blended basis lower than 5%.

And is there any update on the non-PL non-merchant loan products that you were actually planning to launch?

Vijay, do you want to talk about that? Non-PL, nonpersonal loans so this might be referring to -- sorry, Anand this...

What kind of other product are we talking here? Non-PL...?

So basically, any home loan or mortgages is what basically we were talking about earlier on. So is there any update over there like...?

No, they are very low margin. They increased the line expenditure on us, and we have not been able to find out a commercially logical reason. The CAC was -- I mean, the opportunity cost of executing that versus other things, I'm not yet a big fan, to be honest about it, of those loans, but we did try them out, but we sort of receded from that.

And I think if on the net payment margin, you said that the subscription revenue is still doing well. Are all the device merchants now paying you full rental or there are still some merchants where you are waiving the rent because you want them to gradually pick it up?

There are certain merchants who get rent refund based on certain volumes and features. Always, now or later [Foreign Language].

Yes. So just to clarify, we don't give waivers basis in activity, we give waivers basis if you're doing a large amount of volume with us, then we give that, which is what Vijay is referring to as a refund. So there are programs that we would run that if you do very large volumes with us, then you don't have to pay the subscription. And the logic is that those merchants are great merchants for lending eventually and also we make [indiscernible] incentive.

Sure. If I can basically squeeze one more question is on your CapEx. You said that basically a large part of the devices you are actually refurbishing and supplying it. How long do you see that CapEx will remain so low? And -- or any guidance for the [indiscernible] for that?

Yes, I think that is one of the few factors. That is an important factor for sure, but it's also cost of device has come down as we are ordering at large scale and so on. So we don't expect the CapEx to go back to FY '24 levels. And we think that the refurbishment, sort of the savings on CapEx due to refurbishment should generally be the case over the next 2, 3 quarters. This number is not something that we actively manage on a month-on-month basis or a quarter-on-quarter basis. So that might go up and down a bit. But refurbishment supply is there for the next sort of 2, 3 quarters. And even in FY '26, we don't expect the CapEx to go back to FY '24 levels. On depreciation, I mean, I don't want to give a specific guidance other than to say that as the high CapEx of FY '24 relative to FY '25, as that starts to sort of roll off because these are 2-year depreciation for Soundbox, then we should see a reduction in CapEx -- sorry, reduction in depreciation. And we're already starting to see some of that, but we should see that just sort of accelerate a bit next year.

Next question is from Manish Shukla of Axis.

Now that you started giving DLG, as a lender, what is an incentive for a lender to go a non-DLG loan with you?

I think these are sort of good faith discussions that one does. Some lenders say, I don't really need DLG because it doesn't really help me. I'm comfortable doing it on a non-DLG model. I quite like the model earlier of you keep the sourcing fee or I pay you a sourcing fee and then eventually, you get a collection incentive in the end versus some lenders say that, hey, it's kind of important for us. It will help us scale the business, right? So these are good faith discussions. There isn't a -- yes, I mean, if one had to not think of win-win and they just had to think about, hey, why not just take the DLG, then I can see where your question is coming from. But that's generally not the tone of the discussion. Generally, the tone is, listen, we are comfortable doing x amount. But yes, it might be helpful for us to have FLDG and we can do more. And on the basis of that, one sort of agrees to a certain model.

Okay. The cash that you've got from PayPay stakes which is sitting in the Singapore entity, what is the proposed end use of that money?

We're still working on that. The cash is sitting overseas. And we haven't -- and obviously, cash is fungible. If we needed cash in India, then we would have also brought back, but we don't need this cash in India. We have over INR 10,000 crores of cash in India. So we will work this out over the next couple of quarters, whether there is an overseas use of funds of this money or whether we should bring that back to India. Most likely, eventually, we'll bring it back to India in some form.

And if you were to get it to India, will there be tax implications?

We're still working through the tax implications of this.

Okay. Next question on contribution margin. Where do you think it stabilizes going ahead? I mean you've been really tightening your belt on expenses, but I'm just trying to understand where does the steady state contribution margin, what does it look like?

Yes. So we've talked about without UPI incentives just being at 50% to 55%, so 50% to 55% and including UPI incentive for this to be 55% to 60%, so we continue to believe it will be in that range. We do see -- and I should just point out that the contribution margin last quarter and this quarter, particularly this quarter, has been impacted by DLG costs, which comes in other direct expenses. And from last quarter to this quarter, it ramped up because last quarter, we did it in the middle of the quarter with our largest merchant lending partner. So this quarter, the number was much higher, not because we were giving more DLG as a percentage, but because of the full quarter impact. So as that number sort of plateaus a bit relative to our overall disbursal, we should see an uptick in contribution margin just basis that factor alone.

But Madhur, you're also making corresponding collection fee upfront as well, right? So there's a revenue impact also of the DLG not just the cost?

Yes. So what we make upfront is still sourcing fee. The -- what I said in the last call is that the trail revenue is significantly higher, right? So there are 2 impacts. One is the trail revenue is significantly higher? And I know there are a couple of questions about sort of our lending take rate which has gone up meaningfully, partly because of this, not only because of this, but partly because of this. And second is the DLG cost has gone up meaningfully this quarter compared to last quarter, but over time, it plateaus. So what happens is if we have given x amount of loans under DLG, then that starts to generate the trail income over the next 3, 4, 5 quarters. Whereas the DLG cost is not going up meaningfully.

Sorry to squeeze in, then what really explains the Q-o-Q jump in take rate? If you can just help us with the delta, 7% to 9% that we...

It's partly the trail revenue from -- given the previous quarter as well as a little bit of loans given this quarter. So whatever you gave in October, November, you have higher trail revenue. And second, these are collection efficiencies have continued to improve, like we mentioned earlier, which means that in our, if you will, old non-FLDG book, we are getting higher collection revenues than we did before.

Next question is from Rahul Jain of Dolat Capital.

Yes. So if you could help me out, what is the ideal revenue run rate of the non-lending part of financial services? And what could be the ideal gross take rate on DLG part of MCA book?

Sorry, what was the second part of the question, Rahul?

So what is the gross take rate on the DLG part of the MCA book merchant loan?

The gross take rate over the lifetime of a loan in the DLG business. So if you remember, earlier, we used to talk about 3% to 4% on sourcing and 1% to 2% on collection, right? So to that, you can add the FLDG amount. So let's say the FLDG amount was 3.5% or 4%, you can effectively add that in order to get gross take rate. But to be clear, most of that is not in the quarter of disbursal. So in the quarter of disbursal, it will be the sourcing fee, which we have talked about as 3% to 4%. And the collection revenue goes -- effectively goes up by the amount of FLDG cost that we have incurred. Does that directionally answer your question?

Yes, yes. So essentially, we should reach 8%, 9% on a trailing basis. That is how the maths would eventually work.

Yes. Broadly, that's right. It obviously depends on the FLDG cost and depends on various other things, but that's broadly right. It can be a bit higher than that because -- so that is the model we used to talk about. Like we have mentioned in the last couple of quarters, our collection efficiencies on MCA has gone up. So that 1% to 2% that we had indicated we make on a trail basis, that number is also inching upwards. So yes, but broadly, the math that you did was right. Second is on nonfinancial services -- sorry, non-lending revenue in financial services, I don't want to give you an exact number because that's, we think, slightly additional breakup. But broadly, it has been 10% to 20% of our financial services revenue almost consistently over the last 8 or 9 quarters. So when lending revenues have gone down like they did in Q1 and Q2, then it was a little bit higher. And currently, as lending revenues are recovering, then it's a little bit lower, but it's been in the 10% to 20% range.

Got it. And just last one more question. Given the kind of efficiencies that we have displayed on the indirect cost side and our commentary on the depreciation side, and of course, the amounting cash position and cash generation expectation in the coming quarters. And probably, we might have achieved adjusted EBITDA breakeven at least on an exit month basis, if not for the quarter. So is there any specific goalpost that we are aiming for going into next fiscal or calendar whatever?

So I think just laying this out, like we said, we are very close to EBITDA before ESOP profitability. This is without UPI incentive, just to point out. So for example, if we got this quarter's share of UPI incentive in this quarter, then we would already be profitable. But obviously, the sort of phasing of that works a certain way. And what is also going on when we look at EBITDA to PAT is that our interest income is going up largely because we have more cash than we did before. Our depreciation is going down, as we discussed earlier. And our ESOP costs are going down as we have disclosed in the back of the earnings release. So as a result, the gap between EBITDA before ESOP and PAT is going down very meaningfully. I think in 2 or 3 quarters, that gap will be basically 0. So just mathematically, we are going to get very close to -- we're going to get to PAT profitability once EBITDA before ESOP is profitable, either in -- maybe in the next 1 or 2 quarters sequentially.

So that's obviously a clear -- I wouldn't even say a target. That's just something that we are just sort of marching towards. But eventually, the point is that we want to get very efficient as an organization. We want to drive higher revenue growth. We have had obviously some headwinds with respect to where we are in the sort of credit cycle and personal loans and so on, which is fine. I think one just has to be cautious and build great long-term businesses. And as we start to get that revenue growth, a lot of the work that we have done this year on becoming more efficient should translate into fantastic operating leverage. And so while EBITDA breakeven and PAT breakeven are milestones, but obviously, that's not the goal. The goal is to have double-digit EBITDA margin relatively soon and then have that translate into a substantial amount of PAT.

Understood. So first goal, as you rightly said, is the difference between adjusted EBITDA and PAT would vanish in a couple of quarters, maybe Q4, Q1 and then the double-digit EBITDA margin kind of probably 4 to 6 quarters from that point. And from a growth point of view, as you rightly said, and we have rightsized the business in various aspects. So what should be a good benchmark for growth on a 3- to 5-year perspective basis for the -- assuming the same set of business stays in the same form and shape, there's no new incremental revenue line that we might see or maybe a recovery on 1 or 2 products that we are not working on. So assuming the same business stays what should be a 3- to 5-year goalpost on the growth path?

Yes, for reasons I'm sure you'll understand, Rahul, I don't want to give specific guidance or specific ranges off the cuff. But I think things that get us excited is obviously payment volumes in India growing very meaningfully. So maybe 20%, 25% year-on-year relatively safely. The second is the take rates are starting to have some levers of tailwinds. So somebody asked earlier about RuPay on UPI. Just like that, there are several other things which are inching take rates, which are putting a bottom with the take rates, but also increasing take rates. Merchants' willingness to pay for devices and other hardware and software is very encouraging as well. So I think that is another monetization source.

We have a whole page in the earnings release about why we are excited, particularly about the merchant payment side of the business. And then the last thing we point to is that only about 6 lakh customers or merchants have taken a financial services product from us last quarter. And we're not even getting into product per customer, right, which is I know in financial services is something that more mature organizations track. We're just talking about a number of people who have taken at least one financial services product for us -- from us. And that is only at 6 lakhs, which is less than 1% of our base, right?

So clearly, there's an opportunity to distribute to more of our customers in merchant financial services. And if we're able to do a good job of that, these things obviously, even at less than 1% scale, it is north of INR 450 crores of revenue for us. So if we are able to increase that penetration by finding the right products for the right customers and having them transact on that -- on our platform, then there is a huge amount of high-margin revenue that is available to us. So maybe I'm sort of stating the obvious, but I think these are the 2 or 3 components which get us really excited about our business in the medium term.

Next question is from Jayant Kharote of Jefferies.

Just one more on the DLG. If I got the construct correct, it seems that we may be starting off on an incremental basis with a contribution margin of less than 50% on these products, incremental DLG disbursals. And then as 3 to 4 quarters sort of pass and you get the collection incentives that can move back to 50% or even higher. And if that is the case, then does it mean for the next 2, 3 quarters, we should expect contribution margin to be nearer to the 50% mark rather than the 55% mark?

So I agree with the first 2 points that you made. So just to be clear, in the month of disbursal, the merchant loan is less than 50% margin, right? Because we get a sourcing fee like we have always mentioned, is 3% to 4% and then we have an FLDG cost which is meaningful and -- can you put Siddharth on mute, please? Thank you. So we make -- we pay an FLDG cost, which while we have said is less than 5%, it still means that the business is not in that month, contribution margin positive. And yes, over time, we make a lot of trail revenue, which while I was answering Rahul's question, we sort of had a directional computation on. So overall, that business is significantly higher contribution margin.

I think the point that I was making earlier is that our FLDG cost despite the fact that it doubled this quarter, more or less doubled this quarter because you've seen from the disbursal numbers. Despite that, we were able to maintain our contribution margin. Obviously, going forward, we expect our revenue from merchant loans to grow faster than incremental FLDG -- than effectively FLDG growth because we are starting to get collection revenue from not only the old book, but also from the FLDG book. So we don't expect that incrementally from here, there should be a huge drag on contribution margin. But I'll just maintain the 50% to 55% range. I think there's just various factors where one shouldn't overly try to be precise about quarter-on-quarter contribution margin trends.

And I know too early on this one, but given that the DLG book has been there for now 5-odd months, how is it behaving compared to the earlier products? And I'm asking this because in terms of incremental disbursals, almost 2/3 are coming from this product. It's fair to assume that this product will become the largest part of ML disbursals going ahead, so.

So I mean the business is effectively the same, right, which is that merchants who display certain behavior are white listed and they can get loan from one of our partners. The DLG book may behave the same or slightly better or slightly worse compared to the non-FLDG book -- non-DLG book because of something which might be partners specific, right? So we have DLG relationship with some partners, but not with other partners. So that -- there might be certain partner-specific reasons why it may behave better or worse. As it stands, our DLG book that we have done in the last 5 months has done, I would say, noticeably better than the non-DLG book AUM that exists, right? So it is very much on track. If anything, it is doing slightly better, and we have really no cause of concern on the unit economics or the expected credit losses of that book.

Congrats again for a great set of numbers.

Next question is a follow-up question from Pranav Gundlapalle.

The question is on the personal loans. Can you just provide some color on what drives a lending partners' choice of going with a pure distribution model versus one where you are responsible for collections? Is it simply economics? Or is it risk? What drives the choice?

I think it might -- so in some cases, it's just segments, Pranav. So we have actually both models with multiple partners, right? So we have multiple partners who will do distribution business only with us because they are underwriting them a certain way, and they believe that their existing collection infrastructure can take care of those loans. And then there are -- then they'll also do collection-led business because they think that our collection capability may be much more efficient and much more effective for a certain base compared to them sort of figuring out or adding capabilities to their other collection capabilities. So it really is really an underwriting call that do you feel comfortable. So we'll work with them to say there's a segment that we think we will just have you distribute loans to because you don't really need our collection capability and there's other bases where you may be able to distribute -- give them loans because Paytm is able to collect from them.

Okay. So I'm just trying to understand if we should expect a meaningful change in the mix if, let's say, the operating environment improves from here? Or would it change if it deteriorates further from here? So I guess as long as the loan mix remains where it is, we shouldn't see a big change in that mix between one versus the other?

Yes. I think the idea is that you would go to partners and have both conversations. And I'm not quite sure like -- I'm not quite sure we have a very strong view that, let's say, we are sitting here 4 quarters from now and partners have more comfort on the credit cycle, will they switch more towards collection model or more towards distribution model? I don't necessarily think we have a strong view on that. But I think the idea is that work with partners and continue to find more win-win partnerships because our platform has a huge number of capabilities, distribution and collection and some partners may have...

Pranav, I also want to add that distribution is more like when we have extraordinary differentiation than, let's say, a Google Search would have. Otherwise, it's merely an advertisement because you're -- practically you are trying to generate the lead and then you're giving the partner something. So I'm personally going to say that we remain inclined towards higher-margin product, and it's sort of what we are showing and seeing even though distribution does make sense because we don't address all needs, for example, like larger ticket products. We don't put them in these categories. So there is a fair mix of distribution and distribution plus collection.

Distribution plus collection is the preference because it gets us more margin. Distribution-only does get us larger money per ticket. We, in both days, are trying to create methods or APIs between lenders and us so that it can be easier done than -- as neutrally third party as advertising on Google could work or some third-party place could work. What I can tell you is that we've been able to successfully create even in distribution, the integration, which is far easier and far better than someone else. So distribution opens up diversity and collection opens up high margin.

Next question is from Siddharth Gupta of Voyager Capital.

Congratulations for a great set of numbers. Most of my questions about DLG have been answered. I just wanted to -- I have a few quick questions. First one is, what is the quantum of DLG we want to meaningfully envisage over the next few quarters? We've gone from INR 225 crores to INR 350 crores, just to get a ballpark figure of where we are looking at. The second more specific one is basis our previous payouts and our rough estimations. What is the rough quantum of the UPI incentive we stand to receive in the upcoming quarter? And lastly, I just wanted to understand our road map on 3 of our products that -- now that we can add customers back on UPI, and we have also publicly stated a goal of going back to our Jan 2024 levels, which was, I think, around 9%- or 10%-odd market share. If we have a particular road map of how we wish to go down that route, do we wish to take a cash back-based route? Or do we envisage organic growth? And if you could shed some light on Paytm Money and insurance broking as a business because we didn't dive too much into it in the release that came out.

Yes, Siddharth, a lot of questions, but I'll go the most critical one, which is the consumer growth, which is -- because internally, we have now switched our gear towards the growth where the merchant growth and consumer growth, both are something that are very imminent and seeable. If you notice the number of consumer growth without marketing spend enhancement is very, very encouraging. I, in answer of a previous question, also suggested that there is an organic flywheel that we are seeing.

My and our attempt would be that we continue to do a product-led growth instead of big bang spend money led growth. And we do believe that there is a great opportunity of product improvement and product differentiation. We have a lot in queue going ahead in consumer side in terms of product streamlining. There were a lot of -- I tweeted and I learned a lot about what is not good. And actually, we knew it, a lot of it. So we were in due course, waiting for different, different level of regulatory clarity before we started to release out the product.

And I'm happy to say that those products lined up, will give us growth or retention, better retention, better growth, better reactivation. So that is where our primary opportunity on consumer growth will be. Money and insurance are still our work-in-progress businesses. Money is something that where there is an attention on mutual fund distribution because I think that mutual fund distribution is a great amount of number of customers that we are able to sign up because of -- as you are aware that -- you must have seen on social media and otherwise also that there is a clarity that revenues and trading revenues, thanks to F&O, et cetera, have declined.

So we have followed the same trend as the market has seen. So there is a lot to, let's say -- I would say, do there based on mutual fund, which is what our focus area is right now, number of SIP, number of mutual fund customers. That is what we are trying to say. And insurance-wise, it is an embedded insurance along with our health and auto insurance, which are very good bottom line contributors. So that is what we'll continue. And rest, DLG, again, back to Madhur, please answer.

Yes. Sorry, on UPI incentive, while we don't know what the exact structure will be, so it's hard to predict a number, but we do have about 15% to 20% growth in eligible GMV. And so if you look at our last year number and going forward -- and had to sort of take a ballpark guess at this number, this year's number, then it should be in that range. But I would hate to sort of just start blabbing out certain numbers of what we should expect in Q4. On DLG, Siddharth, we don't really have a target number. We believe in the model and this DLG is -- while it is a cost upfront, it is more than made up by the profitability on the merchant cash advance business or merchant loan business. So we think it's just, in a sense, an investment into getting that profit pool. So we don't really have like a number, which is to say, hey, we'll eventually cap it at this number. Of course, the Board-approved numbers have been disclosed, but I would expect that the Board would be constructive given the performance of this business when we go back to them for any additional approvals.

Fair enough. Vijay, that was a great initiative of meeting for 2 things on product improvement, and it was -- absolutely loved it.

Thank you.

We'll take 2 last questions, given paucity of time. Next question will be from [ Rushabh Sheth ].

I just wanted to check, are there any thoughts on the wallet business? Do you guys want to, at some point of time, start redoing it? Just wanted to kind of get your thoughts now that the whole Paytm Bank thing is behind us. If you have any thoughts or if you guys are planning something on that?

Yes, we want to do it. [Foreign Language] we'll take the next step. Idea is, [Foreign Language].

Got it. And any view in terms of the time line for this outcome?

As people who are on this side of the call, we have only actions that we can effort outcome.

Okay. All right. But based on that, you will decide how you want to go forward on that?

[Foreign Language].

Our next question is from Lokesh Manik of Vallum Capital.

Just a couple of questions. One is a...

Lokesh, sorry, you're not audible. Can you please speak closer to the mic?

Yes. Is this better?

Yes, please. Thank you.

Yes, Madhur, a couple of questions. One is on the FLDG [ lift ]. If you can share the AUM on the basis of which the financial services revenue are generated, do you think that would be a better way for us to track the performance of the financial services business? That is one. Second is if hypothetically, in the midterm, if you have to grow your FLDG book, would it be capped to the free cash that you're generating internally? Or would you want to dig into the cash pile, which is sitting in the balance sheet? Those are my 2 questions.

So on your second question, the sourcing fee that we make in the month of disbursal is [ not only ] the same, generally slightly higher than the DLG cost, right? And obviously, we make a lot more money as trail revenue. So we don't really -- so it's not like in the month of disbursal, it's negative business, and we are doing this, hoping that we'll make -- be able to make up for it later and then make more money. So that's not what is going on here. So that constraint would be a logical constraint to put, but it's not really the case here anyway. On your first question, a majority of the revenue that we make is upfront and sourcing fee, at least in the non-FLDG model, it was always that the sourcing fee was a larger chunk.

So that's why we had always pegged it to disbursal. I think we can consider whether AUM is a better number. I'm slightly skeptical, but we'll think through it. I think one of the things that we are driving at also is what is the revenue per customer that you make from the number of customers who have taken a financial service product. So one of the ways that we have also started to look at it is you have 6 lakh customers taking a financial services product and you made INR 450 crores. So what does that drive to long term? Of course, now some customers are higher ARPU -- some products are higher ARPU, lower ARPU. But broadly speaking, the leading indicator to being able to make more financial services revenue down the road is if more of your customers are availing it.

Yes. I think AUM is when you are the actual book owner, which we are not. So I would go towards per customer revenue sort of an ARPU from financial services, and they could take any of the 3 services, loan, credit or insurance.

Fair enough. That's a perfect representation of the product-led strategy that you're talking about. That's great. That's great.

Thank you so much. With that, we come to an end of this call. A replay of this earnings call and the transcript will be made available on the company website subsequently. Thank you all for joining, and you may now disconnect your lines. Thank you.

Thank you, everybody. Thank you, Pranav.