Good morning, ladies and gentlemen. Welcome to the Bango PLC final results investor presentation. Throughout this recorded presentation, investors will be in listen-only mode. Questions are encouraged. They can be submitted at any time via the Q&A tab that is just situated on the right-hand corner of your screen. Please just simply type in your questions and press send. The company may not be in a position to answer every question it receives during the meeting itself. However, the company can review all questions submitted today and will publish responses where it is appropriate to do so on the Investor Meet Company platform. Before we begin, we would just like to submit the following poll, and if you could give that your kind attention, I am sure the company would be most grateful. I would now like to hand you over to the executive management team from Bango PLC.
Paul, good morning, sir.
Good morning, everybody. Thank you very much for taking the time to join us this morning. I am joined here by Matt Wilson, our new CFO, who joined us at the start of the year. We will spend quite a lot of time going through the results of 2024 in quite a bit of detail. Sukey is here as well, who will guide us through the Q&A at the end. Thanks to everybody who submitted questions in advance. We have had a great number of questions in advance. We will try to answer all of those at the end. If you have any as we go through, please do add them into the online tool.
Today we are going to go through the reminder of the business model, a bit of a summary of 2024, then I will turn it over to Matt to go into the financials in quite a lot of detail. I want to talk about the DVM opportunity and some of the signposts we have seen in the market for the size of the opportunity and how we are playing in it and how we are progressing against our ambitions in that particular market. We are then going to talk about where we have been investing. A lot of our investment in R&D CapEx, which has started to come down, sort of peaked in 2023, was down in 2024, will decrease further in 2025 and 2026.
I want to really give you a bit more flavor of where we're investing that and why we're investing that, and the benefits that we're seeing from that investment. Then we'll talk a little bit about the outlook, how you can find out more about Bango, then we'll move on into the Q&A. A lot to go through, so please do continue to ask questions as we go through. Just as a reminder, if you're ever visiting one of our Bango offices, you will see this everywhere. We're here to be the place where people subscribe to give power and control for customers. Our values are THRIVE. You can see those at the bottom there, power everything we do, in terms of how we strive to behave and how we go about driving ourselves towards that vision.
I thought that was a nice link before we dive into the results to talk about the ESG, well, certainly the E and S aspects of ESG. I think the social ones are, for us, it's all really all about people. We've had a very big spike in the employee base, as you can see from the graph. The line on the graph on the right-hand side shows you the DOCOMO Digital acquisition, we went from over 120 people, and really peaked to 360, then just come down continually from that. You can see that in the graph on the right, which is basically our employee engagement survey. It's something we do annually.
The line shows you the number of respondents, and we always get in excess of 95% of employees responding to that, which again, is a really good sign of employee engagement. What you're seeing is throughout that period, we've continued to have really best-in-class positive responses. That's the blue bar, in terms of people who respond positively to the 80-odd questions we ask them once a year. Then declining number of negative answers. Obviously, the gap between the two are people in that neutral zone. Throughout the journey that we've been on, we've continued to maintain really strong employee engagement. That's important for recruitment, it's important for retention, and it's important for the productivity. Because fundamentally, engaged employees give discretionary effort, which allows us to do some pretty extraordinary things. We also like to support the charities that our employees support.
We supported 11 different charities last year. Those are all employee-led activities. We don't have a company charity that we support. We support what our employees are passionate about, and that's really a core part of our values here at Bango. Finally, on the environmental, we've been talking about measuring our carbon emissions for a number of years now. We recently changed to a new service, which we think is really the best. It allows the lowest cost way of measuring the emissions and our carbon footprint in a great deal of detail, a lot more detail than we'd ever had before. We're driving to that net zero in 2040, and we have a whole host of actions to drive down our carbon footprint as we continue to grow the business.
There's lots of detail online in our sustainability report, as well as in the annual report, which was also published today. Just a reminder of our strategy. There are four key pillars to our strategy of growth. You'll see these featured very heavily in the annual report and in the CEO statement. Firstly, on EXPAND, it's all about dominating the distribution of subscriptions through the telephone channel. You can see we've made some great progress on that, as we'll talk about as we go through this presentation. EXPLORE is about looking at new verticals. The next one with Continente in Portugal at the end of last year. Portugal is a large high street retailer, adding Disney+ into their loyalty program. I think it's a great example of that. There's a healthy pipeline building behind that in both financial services as well as in banking, as well as in retailer.
Increasingly, we're seeing content providers look to create their own bundles. People putting complementary content with their own content. SiriusXM in the U.S. have just launched a bundle with Fox Nation. They put that Fox Nation excellent service alongside their music streaming service to create a unique content offer. The Digital Vending Machine is really at the heart of all of those transitions and all of those different mechanisms of bringing bundles to market. The nice thing is, for all of those use cases, it's exactly the same products. When I talk about the investments, when I talk about the features, you'll see it's exactly the same no matter what the vertical is. The ENHANCE, this is probably, I think, the least developed. We're in very much in that experimental phase.
If you think about it, we sit between content providers and distribution channels, and we sit between the two, and we bridge the platform that bridges those two worlds together. We're paid in effect by the people at the bottom, by the reseller at the bottom who's distributing that content in most cases. Really what we have is a whole host of valuable data in the platform, and all our experiments in this enhanced focus are really how can we monetize that data? How can we use that data to deliver more success, to allow more subscriptions to convert, to allow more people to upgrade, to allow content providers to better target customers?
Lots of opportunities to, in effect, look to monetize both halves of the marketplace with exactly the same product just with using some of the information that's in that product in a different way. Very much experimentation certainly for the future, but really for us, a key area of development and one of the reasons that Marisa joined us as Chief Product Officer is to really spearhead the work in that particular area. Finally, on EXTRACT, I'll talk about managing the payments business for cash profit. We still continue to see growth. We still continue to add new routes in that business, but it's a lot more disciplined now. We will turn off routes that were not profitable.
We'll go through some more detail of that later, but it's really a very disciplined approach to that payments business, which, pulling the dot on digital acquisition, is really at a scale where it can be a significant cash engine for the business. Again, Matt will cover more on that as we go through the presentation. Just a reminder, we do have, and we'll talk about these as we go through the presentation, is two halves of the business. We have the payments business, which we book in that transactional revenue stream. Carrier billing is the best example of that. You purchase something, and you pay for it on your phone bill, whether it's a bag of PokéCoins here from the Google Play Store or whether it's some physical goods from the Amazon.co.jp store in Japan. Really doesn't matter.
You go and you make that purchase, rather than pay for it on a credit card, we charge it to your mobile phone bill. We act as a bridge between those global content providers and the telecoms who basically have a bill on which that charge is placed. On the other side, we have the Digital Vending Machine. That's that bridge in the subscription world between content providers and people who want to distribute and resell those content providers. That's where the Digital Vending Machine comes in, and we'll spend a lot of time talking through that as we move through the presentation. Both of those fundamentally benefit from a very simple and single value proposition.
Certainly, the benefits and the technology differentiators we have in the product go way beyond this, but if you boil it down into the simplest words, it's that you connect once and you access many. We bridge worlds between content providers and channels and telcos, and you connect into us once, and you get access to everybody on the other side. We make those connections very quick and very seamless, and that's fundamentally what we do. We're bringing those two worlds together, and that's the same for the Digital Vending Machine, and it's the same for the payments business. On the Digital Vending Machine, we ended the year with 110 content providers.
If you look across the DCB and DVM together, we have 115 content providers and 125 people, if you like, on the sell side of that platform, mainly telcos, but obviously moving beyond that, who are connected into us to get access to all those different content providers. We can launch all those people very quickly, and I'll talk a little bit more later on about why that's possible a little bit later on. That's the core value. Connect once, you access many. We can bring these services live very quickly indeed to market. 2024 was a really strong year. I think a year of both growth and optimization. If you look at the CAGR for the three revenue, both the group revenue as well as the divisional split, see we've, over the last few years, really CAGR has been around mid-30% range.
You start to see in last year a big step-up in EBITDA, you see that in the graph on the right-hand side. Matt will talk a little bit more about why that is, but that's really because of the operational leverage we have in the platform that drove that growth in the top line, coupled with some disciplined cost management allow it to drop to the bottom line. We'll talk a lot more about profitability as we go through this deck. The Digital Vending Machine obviously gives us this recurring revenue stream. We started to report ARR a few periods ago. That saw good growth of almost 60% to $ 14 million last year. A part of that within that, again, we started to report a new metric, net revenue retention. That's the growth in the same corporates of customers from one year to the next.
Anything above 100% means those customers are growing. You see the 125% of that is existing customers start to support more subscriptions and to climb through those license tiers. We'll show a few different splits on how you can interpret that as we go through the presentation. We've got nine new DVM deals in 2024. As we wrap up at the end, I'll talk about what we've seen in 2025 so far, which is certainly an acceleration, but continuing momentum in terms of bringing new customers onto the platform, as well as content providers. Again, we added a whole host of new content providers who are using the Digital Vending Machine to distribute their subscription services to a variety of channels. A really strong year. I'd say a combination of both growth and optimization.
You see that big step-up in EBITDA, I think that's a good point to turn it over to Matt, who can both introduce himself and also walk you through some more detail.
Thanks, Paul. Good morning, everybody. A really exciting time for me to be joining Bango, thrilled to be here and looking forward to walking you through the results for fiscal year 2024. Overall, as Paul outlined earlier, I think we can consider it a transformational year for Bango, delivering double-digit revenue growth, a significant EBITDA improvement of 139%, and a strong second half performance, I think, that underscores our continued progress and resilience. Starting with the top line. Momentum continues. Group revenue up 16% on the year, and a compound annual growth rate of a little under 40% over the last three years. Annual recurring revenue up 59% on the year to $ 14 million, and 10 new partners starting to generate ARR in fiscal year 2024. As well as new business growth, we saw continued growth in our existing customer base.
By way of reminder, we measure this through our net revenue retention metric. That's the ARR from existing customers at the end of the period, divided by the ARR from those same customers at the beginning of the period. In keeping with last year, we saw this metric continuing above 100%, with 125% delivered in FY 2024. Looking at each segment in turn, starting with transactional, a strong year overall, up 11% on fiscal year 2023, and 14% at constant currency. The transparency, I've broken out the revenue bridge here into the various components to provide a little bit more color on the performance. As we continue to optimize and improve the profitability of the portfolio, we exited some loss-making routes, and you can see that in the second bar on the chart.
During the DOCOMO Digital acquisition, we acquired some routes with quite low margin and a high cost of sales. These routes grew by $ 3.8 million during the year, which was much higher than we forecast. Important to note, though, because these routes have very low margin, they have minimal impact to Adjusted EBITDA. Growth in what we call the core transactional business, which is basically everything aside from those exited routes and the higher cost of sales routes I just mentioned. Gross margin there is typically a lot higher, circa 90%. We saw good growth there in fiscal year 2024, growing by $2 million. We experienced some FX impacts, particularly in Japan, on the back of a strong U.S. dollar against the yen, and you can see that in the penultimate bar. Overall, a strong year for the transactional business. Moving on to DVM and one-off revenue.
We saw growth from both existing customers and new contract wins, up 28% overall. If we exclude the audiences business, which we discontinued in Q1, underlying growth was an impressive 40%. We ended the year with 27 DVM customers, 24 of which were generating ARR. Based on feedback and interest around our customer mix, I've included some cohort analysis for ARR, which you can see on the right-hand side. The top chart breaks those 24 DVM customers down by size. At the low end, we have customers generating less than $ 500K ARR, and the top end for ARR greater than $2 million. I think the key point really to bring out here is the ARR increasing as the customers move through the license tiers.
You can see that with the appearance of the yellow bar in fiscal year 2024, where we now have two customers generating ARR greater than $2 million as their subscription growth increases. The bottom chart shows ARR by contract date, customers added pre-December 2023 in the purple bar, and the turquoise bar, those added in fiscal year 2024. As we covered with the net revenue retention, you can see the strong growth in that pre-December 2023 cohort. Moving on to the next slide, looking at costs. I think 2024 marked the year where we were very disciplined on expenses. Again, to support transparency, I tried to isolate our core administrative expenses, which you can see in the blue highlighted line in the table, because I think this is a better indicator of the controllable expenses in our business.
It removes items such as D&A, capitalized R&D, and exceptionals, which can effectively distort the overall picture. As you can see, between fiscal year 2023 and 2024, we actually delivered a $7 million reduction in core admin expenses over the period. Part of that is driven by people costs as the rapid scaling we undertook following the DOCOMO Digital acquisition has since been rationalized back. You can see in the bottom chart, which shows the average headcount over the year, decreasing from fiscal year 2023, and we expect it to further decrease this year, by another 14% to around 200. Outside of the core admin expenses, you will see increased D&A expenses. That is natural. As you will recall, we have done a lot of investment in capitalized R&D. As that begins to generate revenue, it will start to amortize.
We had $ 4.2 million of exceptional items over the period, $2 million of which was a non-cash impairment, and those relate to assets within the audience business, which as I mentioned earlier, we ceased in Q1 last year. Moving on to the next slide, and covering sort of R&D CapEx. Overall, capitalized R&D reduced by $ 2.3 million, and we expect further reductions this year in 2025, and then further reductions in 2026. I think by 2026, we will be targeting that capitalized R&D being 20% of sales. We have seen the complexity of the DOCOMO Digital integration and I think Paul has touched on that previously, which has effectively extended the timeline. Migrations, I am pleased to say, have now largely completed, with 98% of traffic acquired having been migrated to the Bango platform.
As a result, we expect that transactional R&D CapEx to be $1 million going forward from FY 2026, and that really supports healthy cash generation in that segment with Adjusted EBITDA margins approaching 50%. Really is sort of the cash generative engine of our business. With development of the core digital revenue machine now complete, continued investment in the product is there really to underpin the future growth. Whether that be new features to increase revenue per user, for example, the customer Cx interface, or increasing the stickiness of customers through enhanced offer management, or really reducing the internal effort to integrate on the Bango side. We can do that through the Sandbox environment that we have created. Pulling that all together for the period, strong Adjusted EBITDA increase for the year, up 139% on fiscal year 2023 to $ 15.3 million.
You will see in the gross margin line, we saw a slight dilution there, and that was driven by the higher mix of those higher cost of sales routes in the transactional business I talked about earlier. Excluding these routes, the transactional business as a whole will have gross margin of approximately 90%. We saw a benefit in other income last year of $ 2.2 million. This relates to the recovery of costs from NTT DOCOMO for periods pre the acquisition of DOCOMO Digital. It is very difficult to predict what these will be. We do expect there will be other rechallenges in the future, but not at the same level that we saw in fiscal year 2024. Following the strong revenue growth and overall disciplined cost control, EBITDA margin now improving to just under 30%.
Given the operational leverage we have in the platform, we expect to further build on that going through this year and into FY 2026. Overall loss for the year narrows this year to -$3.7 million. It's a $ 5 million improvement on fiscal year 2023. I expect Bango to generate a profit at that bottom line from fiscal year 2026 onwards. Moving on to cash flow. We ended the year with a reduction in cash of $ 300,000 versus the prior year. The step-up in cash flow from operations driven by that increase in Adjusted EBITDA, as well as the positive inflow from working capital. That working capital benefit coming from the timing of receipts and payments within our transactional business. That's for routes where Bango effectively sits in the middle of the payment flow between telco and merchant.
As that cash moves through, we take our revenue share. We expect that inflow that we achieve in fiscal year 2024 to fully unwind in fiscal year 2025. Last year, the business used this working capital benefit to fund cash flow with limited buffer. I think this year, a real part of my initial focus has been to really proactively address that, and really de-risk the position. The new financing facilities that we've put in place, and I'll talk about that on the next slide, have supported that approach. The increase in cash flow from operations was then largely offset by investment in R&D, which continues to be the main use of cash, although, as we've said, that will come down as we move through fiscal year 2025 and fiscal year 2026.
From a financing perspective, we made two repayments on the existing shareholder loan with NHN, the GBP 3 million overdraft facility we have with Barclays remained undrawn at the end of the year. Cumulating together, net debt improving to $ 1.8 million, which is a $ 2.2 million reduction on fiscal year 2023. As I mentioned just now, a large part of my early focus at Bango has been on the capital structure. I'm pleased to announce two new financing facilities in conjunction with our fiscal year results. One, an enhanced loan facility with NHN, as well as a new revolving credit facility with NatWest, and that will replace the existing overdraft with a much larger committed facility. I think both financings demonstrate the strong confidence in Bango's business and strategic plan, both from our shareholders and our banking partners.
I'm pleased to say this really materially strengthens the balance sheet as we move through FY 2025. Under the enhanced loan from NHN, the existing loan will be topped up by $ 2.85 million, essentially returning the loan to the original balance when the loan was put in place. Those loan repayments will be deferred until the end of 2026. The interest rate on the entire balance will move from 6% to 7%. In conjunction with that, new warrants will be issued, and the warrants on the existing loan will be canceled. Under the NatWest facility, we've secured a multi-currency, three-year RCF of $15 million with a step down to $12 million in year three. That facility will obviously come with standard financial covenants, those will be tested from September 2025 quarterly. We're confident with having significant covenant headroom in all those metrics.
Those facilities will really provide significant flexibility to accelerate the initiatives we want to put through this year and into FY 2026. Lastly from me for the looking ahead. With a much stronger balance sheet and committed financing, my priority is now turning to revenue growth and really driving that profitability. In terms of where we are to date, core transactional revenue, again, that's excluding those high cost of sales routes we talked about earlier, currently trading in line with expectations. The strategic investment in the Digital Vending Machine and overall growth in super bundling are driving a strong sales pipeline with plenty of opportunity this year. I'm pleased to say DVM is on track once again to really deliver high double-digit revenue growth for this year in line with consensus.
The high cost of sales routes and transactional have had a more volatile start to the year and are currently behind expectations, and that's driving some of that working capital outflow I sort of touched on earlier. I think it's really important, again, to stress, given the low margin profile of these routes, there is very limited impact to Adjusted EBITDA. As a result, we expect to report Adjusted EBITDA for 2025 in line with consensus. Those are progressing with a series of efficiency initiatives this year, and those are expected to deliver a $1 million upgrade to fiscal year 2026 Adjusted EBITDA versus consensus. We've taken actions to reduce costs of sales and improve gross margin. The high cost of sales routes that remain will be under continuing review to see if we can further optimize.
I mentioned earlier, R&D CapEx is also forecast to reduce versus consensus, $ 0.5 million this year and a further $1 million next year. As the bulk of our expenses are people, there are restructuring costs associated with those efficiencies, which will have a cash impact this year. Those exceptionals, coupled with the unwinding working capital and also the one-off costs that we've incurred, in conjunction with putting the financing facilities in place, will result in a cash outflow in fiscal year 2025. However, the NHN loan and the revolving credit facility with NatWest will provide more than adequate headroom to manage that transition effectively. These efficiency savings and continued revenue growth underpinned by a strengthened balance sheet will result in significant cash generation as we go into fiscal year 2026. We can really look forward to the future with a lot of confidence.
With that, I'll hand back to Paul.
Thanks, Matt. Let's zoom in on the DVM opportunity in a little bit more detail and just sort of try to give some signposts of sort of where we are and where we're going. If we step back and look at the broader subscription economy, we can see that economy is continuing to grow at a six-year CAGR of just over 6%, with an estimation of 3.5 billion paid digital subscriptions by 2029. A clearly massive market. What's important for us is there's an even faster growth in the portion of those subscriptions that are being bundled. This is some data from Omdia, and you can see the growth of bundled subscriptions, and just through telcos is actually significantly higher than the growth of the overall subscription market.
You can see almost 10% CAGR with almost over 600 million telco subscriptions bundled through telcos alone by 2029. That's really where the Digital Vending Machine platform fits, is the bridge between people who have that subscription service and those that want to bundle it and distribute it through a channel. That's entirely our market, is that growth in bundled subscriptions. Here this is just talking about telco. Obviously, we include the other channels as well. It goes significantly higher than that. If we step back and look at why is that happening. This is from some data. If you follow us on LinkedIn or anything, you will have seen the various reports that we publish.
Probably the place to mention I would encourage you to join and register for our new Investor website where you'll get automatically notified by some of the work and some of the research that we do, which is primarily to support sort of the commercial activities. I think as investors you'd find it very interesting. One of the surveys we've done is understanding why people buy subscriptions and what they buy subscriptions for. You can see here there's an increasing number of people buying multiple subscriptions through channels. They're doing that because they want a better value, they want more control, they want to make it easier to sign up. Of the 5.4 average subscriptions that people have, two of those are now coming through a channel.
You can see that significant change versus some of the research we did a year or so ago. Now, almost two-thirds of customers are buying multiple subscriptions through a channel. I say that's where the Digital Vending Machine sits. It's that bridge between subscription services and channels that want to distribute those subscription services. The nature of the bundles is changing as well. This is also good news because certainly, the value and the importance of Digital Vending Machine increases as you go from left to right. If we look historically, bundles were, I think, very simplistic. You bought a particular mobile plan, and you got one service free as part of that, and we call that basic bundles. It's a very simple thing.
You sign up for this, you get Amazon free as part of that or Amazon as a discount as part of that telco subscription bundle. Really not much choice, very fixed offers, but clearly giving the customer value add. That's a lot of what we do already in basic bundles and the Digital Vending Machine supports it. Where it really starts to come into its own as we move into what we call multi-party bundles and into super bundling. A multi-party bundle is a bundle where maybe there's choice, maybe there's multiple different pieces of content put together, maybe there's some discount that you can then add on additional services on. It's giving consumers more choice about how to create these bundles, what services they pick, how they put them together, which level of subscription people go.
Not forgetting Netflix now is not just one subscription. There are multiple different tiers, from an ad-supported tier to a regular tier, to a premium tier where you get sort of the 4K and the Dolby Atmos, et cetera. Multiple different tiers. Giving the customer that choice about how they put those together is becoming increasingly important. We call that multi-party bundling. The final stage is that move to super bundling and think of that as all of the app store for subscriptions, Optus SubHub being one of the best examples where you can go in, pick all these subscriptions, try them easily without having to give your credit card details, experiment with them, maybe pause them, bundle them together, get discounts for buying multiple ones. It's sort of the ultimate choice for bringing all your subscriptions in one place.
As I say, while the Digital Vending Machine serves all three of those, the value that we can provide and the differentiation we have really ramps up as you go from left to right. That's what we see happening in the telco bundling market. We take that telco bundle chart that I showed a while ago and look at how do we think that's moving. If you look at the conversion from basic to super bundles, you can see at the moment today, most subscriptions are simple bundles, but there's an increasing growth of these either multi-party bundles or super bundle subscriptions. We see that growth being really, really high, close to 80% CAGR. I say that's where the Digital Vending Machine really adds significant value. An indication of where we are. We've talked about these hundreds of millions of subscriptions.
If you read through the annual report, we've said this previously in the past, today on the Digital Vending Machine, we're generating tens of millions of dollars of revenue from tens of millions of subscriptions. I think it's the first time we're giving you a bit more detail and flavor on the number of subscriptions that are managed through the platform. We ended 2024 with about 15 million subscriptions on the platform. We're now just in excess of 18 million. You can see we're in that tens of millions in terms of the number of subscriptions that are managed through the platform. These are active subscriptions. This is whether the user is actively using that. In most cases, the way subscriptions are managed, if you pause a subscription, it still counts towards a subscription license tier.
I think it's best just to talk about the active ones, because I think that's a better and clearer metric. Quite often, in the subscription license tiers that operators have, they count subscriptions in other states other than active. You can see we've seen really good growth over the past 12 or 18 months, taking us into that close to 20 million active subscriptions. If you think about our ambition, and I think we've talked about it before, is at the moment we're doing tens of millions of revenue from tens of millions of subscriptions. You can sort of see that we want to do hundreds of millions of revenue from hundreds of millions of subscriptions. That ambition, I think, is supported also by the likes of telcos like Verizon.
You can see quotes here from their CEO of different investor conferences, in particular, talking about how they want 50% of their customers to be on myPlan. If you look at the size of Verizon's customer base, it's about 100+ million customers, of which probably 20 million- 30 million are probably eligible for the myPlan type services. We're talking tens and tens, if not hundreds of millions of subscriptions just from one operator alone.
You can see when we talk about our ambition to get to hundreds of millions of revenue from hundreds of millions of subscriptions, you can see that operators equally have ambitions that very much support that, and we're in a really strong place of sitting at the heart of all these different subscription services, supporting some of the world's largest telcos, like Verizon, who have really lofty ambitions in terms of generating more and more of their revenue and delivering more value to their customers by bundling third-party subscriptions together. A bit more on investment for the future. I talked about this a little bit at the start. One of the things that the Digital Vending Machine does is make the bridging of these two worlds easy. This is a simplistic example that shows on one side, on the right-hand side, what a content provider looks like.
You can see most content providers have multiple APIs. Some have a mixture of synchronous and asynchronous. They upgrade and update all the APIs regularly as they add new plans, add supported tiers, et cetera. It's very complex to keep pace and also to do the first integration. What the Digital Vending Machine does is because we access that connect once, access many, it allows us to do that work once and make it available to hundreds of telcos without any effort on the telco part. It really is that connect once, access many. A lot of the investment that we've been doing so far in building the core platform is to make this as simple as possible.
We can upgrade one content provider, and all our telcos immediately get access to all the new features that are available with that new telco API. The telcos don't have to integrate to these six, seven APIs and do bespoke integrations on a per content provider basis. They can do one integration to Bango and get access to the 100 and so content providers that are on the other side of the platform. That's really the core platform or the core benefit of the platform, as I talked about earlier. You see that in terms of the ability of just to bring services to market and bring services to market very rapidly. That's the value that the platform delivers. Increasingly, we're moving and extending, and certainly where the R&D CapEx is focused now is on managing that subscription over its entire life cycle.
If you think right at the start, if you operate in a bundling or a multi-party bundling or a super bundling platform, how do I find the content providers? How do I get them on board? We build things called partner portals, you can go in and self-discover content providers that you want to add to your bundle. Our eDisti program, where we act as a distributor for those services, is a core part of that, and already we've launched a couple of services this year where partners have self-discovered themselves without us necessarily having to do any introductions or any sort of ongoing discussions. You bring that partner in, how do you integrate it? How do you test it? As Matt mentioned, one of our drivers for R&D is where we can reduce the effort that's required on our side.
Tools like a Sandbox, that people can self-certify into the platform, can self-test, can self-connect, are really, really important about driving down the overall cost that Bango has of maintaining the Digital Vending Machine and of bringing on new partners. Then we get into the complex world of offers and how these bundles are put together. I'm going to delay that a bit to the next slide because that's where, really, a lot of the value and a lot of the intelligence really sits within the platform that we've delivered over the past 12 months or so. But right at the end, you have the maximizing growth.
That's how we can use the data, the dashboard, and the insights that we have in the platform to make sure that subscriptions are renewed, to look in the future using AI for things like automatic and proactive renewal messages. Have you targeted the right customer base? There's lots of opportunity to be using the data along with AI in the platform to really ensure we can maximize the growth. It's not just about bringing the offers to market, it's about how do you make them successful once they're in that market. The offer management and offer orchestration is extremely complex, and I've tried multiple times to create a simple slide and gave up. I think really that's because what that feature does is take something complex and try to make it simple. Let's take something that intuitively seems very simple.
You want to take a single telco subscription, I don't know, so many minutes or so many gigs of data per month. I want to offer my customers with that, this unique piece of content, I'm going to bundle Netflix and HBO Max together, I'm going to give them that as part of that tier or package. Then I'm going to allow them to pick one or more other things that they can add on top. Sort of perks that they can bundle on top. Sounds pretty straightforward, sounds like the sort of thing that should be very easy to do. But, if you break it down into the constituent components, it gets very complex very quickly. Simply activating all those at the start could be in excess of sort of 30 different API calls.
A very complex workflow because you've got to create the telco subscription, you've then got to create the Netflix subscription, the HBO Max subscription, and then the perk subscription. It's a lot of work for you even sort of get that set up. What we've done with offer management and offer orchestration is make that a lot simpler. It's a templated approach, you can describe that offer, publish that offer to Bango, then we take care of all the underlying orchestration. We've taken all that complexity away from the telco's back office system and absorbed it within the Digital Vending Machine. The benefits of that, it allows these services to launch a lot quicker. The faster we bring these services to market, obviously, the faster we start generating recurring revenue, and the faster subscriptions count towards the license tiers.
It allows us to reduce the effort, because actually helping an operator through these 30 different API calls and the different error cases is very, very complicated. It allows us to actually reduce the effort of bringing these services to market, at the same time as bringing them to market faster. It creates a very sticky relationship, because historically, that complex business logic that might have been done in the telco's back office is now done within the Digital Vending Machine. It creates a very sticky long-term relationship with the customer. That complexity goes way beyond just the initial integration. It goes into the life cycle of that subscription. What happens if I want to change my Netflix to the premium tier? What happens if I want to change the perk? What happens if I want to pause a subscription?
What happens if I want to change my overall mobile plan? Every event where there's sort of any change has a ripple effect with these multi-party bundles. They're a great consumer offer, but the complexity of managing those, if you're not careful, can soon become overwhelming. That's a lot of the work we've done over the past 12 months or so, continued to do, is to take these complex scenarios and make them extremely simple, so you can deliver incredible consumer value very, very quickly indeed. That's where our CapEx and our R&D investments have gone, and that's where it goes moving forward. The sort of core platform investment now is pretty minimal. I'm talking about sort of $1 million for payments. It's probably a couple million for the Digital Vending Machine.
The rest of that investment is on these features that either generate new revenue, create a stickier customer, allow us to get services faster, or reduce our own integration effort. That's entirely where our focus is. We're coming to the end, and then we'll move into Q&A. If you look at sort of outlook, you will have seen a lot of this if you've been through the trading update we put out before today. The bulk of the DOCOMO Digital traffic is now migrated. That transforms the profitability of that payments business, allows it to be the cash engine it is for the business. As I say, it's taken us longer than we'd like or thought to have done those migrations.
Now, that's largely behind us, and we'll really sort of unlock that business in terms of being the profitability and the cash generation engine for the business that we know it can be. The core transactional business is growing as we would have expected. As Matt said, some of those higher cost of sales routes that we acquired from DOCOMO Digital, there's a volatility in those, so they're sort of currently behind expectations. As I said, there's minimal impact to EBITDA because of the margin nature of those routes. We continue to look at how we can take additional cost of sales out of the channel, and how we can potentially restructure those routes, which may result in less revenue, but at a higher quality sort of margin. On the Digital Vending Machine, we've had a great start in 2025.
As I said earlier, we've done nine new customers in 2024. That's about the average over the last two or three years, is about nine a year. We had 27 customers. We started sort of back end of 2020. So about nine or so customers a year is on average what we've done. Already in the first five months of this year, we've added six new customers. That's a great step forward, a big acceleration in that business. I've talked previously about my frustration has been the long sales cycle. I think we're now benefiting from deals that have been around for a while, now coming down to the bottom of the sales funnel. We're also seeing a sort of increased momentum. If I sort of pick three different geographies and sort of give three different examples.
In the U.S., we've had for a number of years now three out of the top five telcos. We're now at six out of the top eight. That's a great position to be in. The U.S. is an interesting market. You sort of have five very large providers and then a large sort of Tier 2, Tier 3 group of sort of about tens of millions of customers, who require sometimes regional broadband providers or regional cable providers, that are looking to use bundling increasingly to reduce sort of churn and stop people cutting the cord for their cable TV subscription by putting third-party subscriptions on top. The launch we did with Altice, having Disney+ as part of their package with Hulu a few weeks ago that you will have seen in the RNS is a good example of that.
The U.S. is a great position now in that sort of Tier 2, there's lots of opportunity for us to expand in that marketplace in the U.S. Moving into Asia, we signed our first contract ever in South Korea. We will lead in telcos. They're coming over here to Cambridge to do a formal ceremony announcement over the next week, so you'll see more details on that over the next week or two. I think that's really, really important. Korea is a market in many ways like Japan, and Japan is a market we've been in for a long time and through building up trials and developing slowly, we've built a very strong and dominant position in Japan. Korea and Japan share very similar traits. It's a very difficult market to enter.
Once you are in and you've proven as a trusted partner in that market, the business opportunities, given how people pay for goods, how people subscribe to services are really, really exciting. For us, that's a fantastic position to be in, and certainly, NHN as a shareholder, have been very supportive in terms of helping us developing that business in Korea, and we're really, really proud and pleased to get that first sort of contract over the line. Finally, in Europe, which is I think my biggest personal frustration, I think I've shared that in the past. We've seen the first sort of green shoots, if you like, and deals have been in the sales funnel for some time. We signed our first deal in Eastern Europe at the back end of last year. We had a deal in the Benelux region in the start of this year.
I'd say that the first green shoots of a well-developed sales funnel in Western Europe are really starting to flower, hopefully we'll see more success as we move throughout the year. As Matt mentioned on big focus on efficiency, profitability, sort of cash generation, you'll see we upgraded 2026 EBITDA versus consensus by $1 million, we've taken $500,000 out of our R&D CapEx this year, $1 million out of CapEx in 2026. You add that sort of EBITDA increase with the reduced R&D CapEx, you can see obviously the cash generation of the business is significantly improved, 2026 is really a year of significant cash generation. Finally, we really appreciate everybody joining today. There's lots of other ways you can keep in touch.
We've launched a new investor website called InvestorsHub, where you can log in, ask questions, see questions other people have asked, comment, like things. It's much more sort of social media engaging. It makes it easier for us to manage and answer questions. Please, do encourage you to join that with the way of staying up to date with what's happening. It's also a good way of engaging with us in a very simple and easy-to-use way. Really encourage everybody who's on the call to join that. A lot of what we do for sort of our commercial marketing activities, as I said earlier, I think is very relevant for investors, a lot of that is through LinkedIn. You can follow Bango, you can follow me. I post on videos and comments and posts and things like that.
There's a lot of all the research and the core research that we produce at Bango is on the Bango LinkedIn. LinkedIn, while those channels are really for us to help us drive business, I think there's a lot of value in there for investors as well. I'd encourage you all to take a look at those. With that, Jake, I'll turn it back to you, and we'll head into Q&A.
Perfect. Paul, Matt, thank you very much indeed for your presentation this morning. Ladies and gentlemen, please do continue to submit your questions just by using the Q&A tab that's situated on the right-hand corner of your screen. Just while the team take a few moments to review those questions that have been submitted already, I'd just like to remind you that a recording of this presentation, along with a copy of the slides and the published Q&A, can all be accessed via your investor dashboard. Guys, as you can see there, we have received a number of questions that were both pre-submitted ahead of today's event, as well as those that have made their way through throughout your presentation this morning as well. Firstly, thank you to all of those on the call for taking the time to submit their questions.
Sukey, at this point, if I may just hand over to you to chair the Q&A with the team, and if I pick up from you at the end, that'd be great. Thank you.
Great. Thank you. Thanks, Jake. A big thank you to everyone who submitted questions. We're always keen to engage and improve understanding of the business, a big thank you. We've actually had so many questions, some covering very, very similar themes that I've grouped. We 'll ask those to Matt and Paul now. However, just a note, if you do feel that your specific point from your question was missed or that it wasn't answered to a degree you'd like, then please use the Q&A top right-hand side and make sure you ask and you get the answers that you want. I'll start with a question around cost. I'll start with you, Matt. It's around the DOCOMO Digital acquisition and further action on costs that we're likely to see with the acquisition or other areas.
Sure. I think in the presentation I touched on that core admin expenses metric, w hich we saw really good reductions last year between 2023 and 2024, a little over $7 million, and R&D CapEx also reducing in conjunction with that. I think that core admin cost metric, as I mentioned earlier, is a better way of viewing the cost base. Obviously, admin expenses as it's reported in the statutory accounts can get distorted by D&A and those non-cash items and hence core admin cost is a better reflection. I think with the efficiency initiatives that we plan to put through this year, we expect another $ 2 million-$3 million savings in that core admin expenses in FY 2025 with a further $1 million of savings in FY 2026. Our R&D CapEx over that period, I expect to reduce by $3 million.
You can see the real emphasis being placed here on really driving profitability and ensuring we're deploying capital and resources in the appropriate areas.
Thanks. On DOCOMO Digital acquisition, Paul, there's questions around how the acquired business is developing and when the benefit will drop through to the bottom line.
Good question. Well, firstly, I think you start to see it dropping through to the bottom line. If you look at the step-up in EBITDA in the second half of last year, I think that becomes very evident. I understand it's been sort of a gradual thing because we've gradually been reducing costs over the period since we did the acquisition. As a reminder, we did that only at the back end of 2022. It was an acquisition we did for customers. It was never a technology acquisition. We did it particularly for the customers that DOCOMO Digital had, but specifically for that long-term engagement with NTT DOCOMO, who's the largest operator in the largest carrier billing market in the world. We did it for scale.
It immediately made us the number one integrator for the Google Play Store, the number one integrator for Amazon, and we were already the sole integrator for Amazon DCB in Japan. It gave us that, as I say, that exclusive relationship with NTT DOCOMO for DCB of online services in what is a huge market. That's what we're doing. The plan always was to take these customers and move them onto the Bango platform. I think it's certainly taken longer than we said at the time or than we anticipated. Most of those are not technical challenges. It's logistic and coordination challenges across the hundreds of routes that we've now migrated onto the Bango platform. Once we're on the Bango platform, the cost of any incremental transaction is obviously smaller. That's why we get traditionally a very high margin from the DCB.
Outside of where there's sort of this channel arrangement that Matt talked a little bit about earlier with these sort of high cost of sales routes. Really the cost of that acquisition, if we look at where it is, it wasn't an acquisition where we paid tens of millions to acquire the business. The cost of the acquisition has been on the inflated cost base that we've had since that. It's not a huge acquisition. It's not like there's a big one-off cost for the acquisition that appears on the balance sheet. The cost of this acquisition has really been the inflated expense. As I say, you saw our headcount go from 120 up to 360. You'll have seen in Matt's slides, we're going to average this year at around 200.
That means we'll exit the year at around 180, which is around the right sort of size for us moving forward. It takes time to do that. You can do that at a certain pace. I think what we're getting, w e've started to see that come out. It's maybe come out a bit more gradually than people would necessarily have seen. Certainly, you can see it in the EBITDA step-up. You certainly see it as increasingly moving forward that the cash and the profitability generated from that transactional business, we get towards sort of 50% EBITDA margins, very minimal CapEx. You'll certainly really start to see that moving forward. I did appreciate it taking a bit longer. For sure, it was absolutely the right acquisition.
I think in hindsight, maybe we didn't explain it necessarily well or explain some of the challenges at the time, but it's been absolutely the right thing to do to give us scale at that payments business.
Question on outlook and when greatness will be fully achieved by Bango. The nub of the question is when we will see real profits, not Adjusted EBITDA.
Do you want me to start that one?
Yeah.
I think I mentioned in the presentation when we were going through the P&L as a whole, I think the key message really is that that bottom- line profit, we expect Bango to make a profit from fiscal year 2026 onwards. Obviously, we saw a big narrowing in the loss this year, which will continue. Profitability will emerge, we expect, in FY 2026. I think with the reductions in both OpEx and CapEx, I just touched on cash EBITDA is set to be positive in 2025, and then we'll see a meaningful step-up change in FY 2026 in that metric when the efficiency initiatives we're putting through come to fruition.
Great. On communications and progress, there's this feeling of useful communication around deals and planning financial forecast is difficult. Going forward, what are the expectations around this and improvements you're looking to add?
On the deals front, I think we share your frustration. Obviously, we did nine deals last year. We didn't do nine announcements. That's quite often because we're prohibited from doing that by the partners, b ecause they're trying to launch a competitive service and differentiate themselves in a competitive market. Somewhat reluctant to highlight the means by which they're doing that to the market. I sort of share your frustration. We've had some debates about whether we should announce without names. I think that our consensus is that's not the right thing to do. We tend to bundle those numbers up into the results presentation. As I say, with Korea in particular, you'll see a lot more detail on that particular win in the next week or so as they look to make that public.
There's lots of news flow around going over the subscriptions on the LinkedIn channel, I'd encourage you to follow those. I think we have, over the last sort of year or two particularly, got a lot more transparent on that, a lot more KPIs. We added ARR in, we added net revenue retention. Matt gave a lot of a breakdown today on the different financials. We had a cohort analysis for ARR. We have the transactionals displayed and growth from the different elements of the transactional business, then show the subscription counts. I hope you see and appreciate that we've added a lot more visibility into the business over time, because we appreciate that's important. We have to do that at a pace that makes sense.
Some of it is commercially sensitive, we make it available at a pace where there's enough scale in it that it sort of avoids highlighting any particular customers or revealing accidentally some customer status. There's definitely been noting progress on that, you'll see that as an increasing trend moving forward. If you've got any particular thoughts or feedback, the InvestorsHub is a great way to send those in. On access to forecasts, there are two analysts that cover Bango. It's Canaccord and Singer Capital Markets. I appreciate that you may not have access to those because of some of the rules in the U.K. I think there is a summary of the reports on certain of the Research Tree websites. You can see part, if not all, of the reports if you are a Research Tree subscriber.
What we do try to do is put the consensus on the website. You'll see the consensus on the website across those two analysts. We try to give as much as we can, we'll look to continually add and improve that. But, u nderstand that could be a frustration.
A number of questions around share price. How do you explain the valuation difference between Bango and Boku ? That's for you Matt.
Sure. I think if you take a step back, I think some of the share price performance will be market driven. It's well documented that the overall AIM market has had its challenges, obviously there's an element of it that's sort of self-inflicted on Bango. I think in the past, we've done a lot of upfront investment in the Digital Vending Machine, but it's about finding that balance between long-term investment and sort of near-term stability. I think it's my job really to help focus the company, ensuring we've got that balance right and deploying, as I said before, the capital and resources in the right areas.
I think with a strengthened balance sheet and these sorts of measures that we've put in place with the financings, I think with the financings, it's important to also add that the revolving credit facility, this is not a term loan. The business just hasn't suddenly taken on $50 million of debt. It's a facility that we can draw as and when we need to, it provides a lot of flexibility. With those both in place, we've got significant liquidity headroom and really the financial firepower to address what we want to address going forward. Again, to touch on those efficiency initiatives, they are there to deliver cash generation in future years. Putting it all together, we really see an uplift in that cash EBITDA, cash generation going through FY 2026.
We've reflected that, I think, in the forecast and what we're saying around the benefits to Adjusted EBITDA in next year.
A related question, do you consider AIM to be the best listing for Bango falling the overall AIM market?
Good question. I think matters don't get any performance. It's certainly not helped. I think the uncertainty we had over IHT around the budget certainly didn't help, and obviously the IHT benefit was reduced in the budget. I don't think any of those helped. At the same time, regulation on the main market has been reducing. AIM has not reduced. If anything, it's got slightly harder in terms of the regulation. I think its future is unclear. I think there's certainly some structural changes I think needed to best position AIM up, which used to be a great market for fast-growing companies to raise capital, and that's certainly why Bango listed on AIM in the first place. Certainly, that's one of the areas that Ray, as chair, is specifically spending some time on. He spends a lot of time with the QCA.
He was down at the street a few weeks ago in a forum that was reported on the news of discussing the future of AIM and some of the structural changes that are required. In the meantime, we continue as a board to look at other listing alternatives. We don't have any firm decisions or plans at this time, but we continue to look at the options. At the same time, I would say Ray is spending quite a lot of time helping people understand some of the benefits and getting engaged in discussions of can AIM make a better market.
With the current share price, has Bango had any recent takeover approaches?
I think it links the last bit where you've seen lots of companies come off AIM and sort of go private. I think we firmly believe that's not in the best interest of anybody, any of our stakeholders, shareholders, employees, or customers. We really don't think it's in their best interest. As a management team, we're not supportive. Another thing, I think it's pretty easy to take yourself private at a 30%, 40% premium. I think we're significantly more of the value than that. The DVM is making massive progress. We have a balance sheet to draw. DCB will start to generate significant cash. We're not interested in a 30%, 40% increase on the current share price. That doesn't increase anything. None of us are here for that.
We're here to deliver something great and big and exciting. I think at the moment, AIM is where we are to sort of deliver that. I guess you get to the other side. Well, what if there's a hostile takeover? I think hostile takeovers are quite rare if you judge most private equity companies. They're all about bringing the management team along with them. Takeovers that are not supported by the management team do happen, but are rare. I think we're in a fortunate position where we have a strong shareholder register with both the founders, NHN, as well as long-term holders like Herald, who have been very supportive of the business. I think that provides us an element of protection.
Matt, a question on why are receivables so high?
Good question. I touched on it a little bit, I think, during the presentation. For certain routes in transactional, Bango sits in the middle of the payment flow. Payment passes from telco to Bango. Bango takes its revenue share, that then passes on to the underlying merchant. The receivables is naturally inflated because of that dynamic.
Paul, this one's you. Why were annual results published later than in previous years?
Good question. I think it was a combination of factors. Firstly, I'm going to blame Matt. We had a new CFO. We also appointed a new auditor. BDO, we appointed a new audit partner. That always creates challenges in any business, where an audit will know the first audit is always a little more problematic because they get to know and sort of understand the business. On top of that, we had a new audit chair on the board . W e appointed Tony Perkins, who took on the Chair of the Audit Committee . We sort of had the three key actors, if you like, in delivering the audit or something new, so it's taken a little more time.
Obviously, as part of Matt coming on board, one of the first things we focused on was delivering sort of a stronger balance sheet, and that's what he's really been working on. We felt it made sense to bring it all together at the same time, and that's resulted in why you're seeing it sort of all and actually, alternatively like best of breed and explain all in one go rather than sort of dribble out sort of bits and pieces as we go fast over time. A combination of all those parts together.
A number of questions on competitors. Who are the main competitors? How do we differentiate, and do we have pricing power?
Good question. Let's take a key part of the business. I think if you look on the payments business, I mean, there are two global players in that market. It's ourselves and Boku. I think our footprints are very different, but there's certainly some overlap, but we are the two global players that I think are in that market who've got the various bits of M&A to sort of consolidate those positions. It's a very stable market. There is growth, there is things that happen, but it's a very stable market, and that's why it becomes sort of a cash engine certainly for us moving forward. On the Digital Vending Machine side, our biggest competitor remains sort of people trying to do it themselves.
Certainly a lot of those simple bundles, remember I talked about that, we move simple to multi-parted, super bundles. A lot the simple bundles were done by telcos themselves because it's one content provider, it's one set of integrations. Where the value of the DVM really shows is when you want to go from one to two to three to five to 10 bundles. That's where we really differentiate. We're seeing increasingly some of those DIY subscriptions migrate onto the platform. We did actually 500,000 subscriptions migrated onto the platform a couple of weeks ago with an operator who previously done directly, but they wanted them all integrated on a single platform. That continues to be, I think, the biggest sort of competitor.
Also the Amdocs out there who've rebranded the sort of MarketONE solution for subscription bundling. I think they're the people we see most, and like I said, the only people I would say have a product to compete with us. I use that word a little bit loosely because I think we are very much a product company. We don't like doing customization. We don't invest on things specific for customers. We build products, and we build features for everybody to use. With Amdocs, obviously, their business model is quite different. It's very much more of a professional services customization model, but that continues to be where it is. I go back to the T-Mobile story I may have told before. T-Mobile in the U.S. is a combination of Sprint and T-Mobile. They were two of Amdocs' largest customers.
They merged, they became Amdocs' second biggest, one of Amdocs' largest customers sort of overnight. We competed head to head in an RFP process for bundling, we won that, and we integrated back into the legacy Amdocs stack. I think why do we win is because of the moats that we have and the sheer number of content providers that are connected into the platform. Also the advanced features that we have, like the user interface, like the software management. Those are really differentiating features.
Praise for our D&I commitment. However, given the hostility of the Trump government to this concept, are you concerned it will cause problems for the U.S. business?
I think the answer is no. I think Trump's anger is aimed elsewhere, especially over the last sort of 12 to 18 hours or so. I think fundamentally, with all the initiatives we've done what's right and what's the right thing to do. I think some of the trend in the U.S. is people who've over-rotated and are doing things that are sort of detrimental to the business. We've always done the right thing that's right for the business, and a big part of that is diversity and inclusion. It's a core part of our values. We believe a diverse team is the best way to get the best results, so we're focused on delivering the best results.
I think we've sort of managed that, I think in the right way that's right for the business rather than it just being something nice to put on a website to claim to investors. For us, it's about doing the right thing for the business.
Matt, one for you. There's a question on the high percentage of development costs and how that percentage of revenue that takes up. How many years will it take to recover the cost of the margin earned?
If we take a step back, the Digital Vending Machine is a relatively nascent product. It came onto the market in 2020. I think what Bango's been very good to do is obviously invest upfront and underpin sort of that future growth. There's been a lot of investment in this product. As Paul mentioned before, initially that investment is to establish the core product. Now, it's about investing to really fine-tune increased revenue per user, improve efficiency on our side to integrate new partners. That's what we expect to happen. We talked a little bit about the revenue opportunity that we see here. There are big secular trends towards subscriptions, towards bundling, towards super bundling, and all that outlined in the past. What we see is the ARR opportunity here.
We can see a payback and a return on that CapEx once that comes through. Naturally, as the product develops more and more, then the pipeline increasingly begins to build. We've seen that this year with the momentum in the early customers that we've added to the platform at the start of the year. It's a gradual process. As I said, going back to what I said before, it's ensuring that we just get the balance right in terms of that investment, future versus near term.
There's a follow-on question about what the steady state segment spend for future actually be, sectors going forward.
I think with the migrations completing, we touched on that a little bit earlier around from fiscal year 2026 onwards, we see sort of the capitalized R&D for the transactional business being roughly $1 million per year. That's really sort of driving healthy cash generation in that segment. On the DVM side, if you were to strip back all the investment for sort of future growth, I think Paul touched on it earlier, you're looking at mid-single-digit spend. The key thing here is really sort of future-proofing that future growth. That's why the investment is higher than that.
I think it's fair to say this, building on that, the reason the DVM margins are high is because we've invested to build a product that's suitable for everything. Because we're investing to build a product that's suitable for everything, under rules we capitalize it and then amortize it sort of out over time under the financial rules. We're not building a product for one customer. If we did, there would be expense and the cost of sales would be higher, and the margin wouldn't be at the level it is. I mean, if you look at the structural finance of that business, the reason it's high margin is because of the investments and because of the technology we've built within the platform itself.
Just a follow on that note, how much cost is expensed each year in R&D that does not qualify for capitalization under IAS 38?
Look, we will capitalize anything that qualifies under the standards to capitalize and anything else we will expense.
On the loan announcement. How did you determine that the total cost of the NHN loan is a good deal for the company and unrelated shareholders?
Let me do that, then I'll sort of turn it over to Matt. I think, the NHN have been a supportive shareholder, clearly as a result they're a related party. We did a full analysis with a NOMAD on were the terms fair and reasonable for all the shareholders, and we strongly believe they are. I think the question talks about the warrants and 2.5% of warrants. I think there's a couple of interesting points to point out. Firstly, obviously, the warrants with the previous loan were canceled. The new warrants when exercised will result in a cash inflow into Bango, and there were no arrangement fees with the loan, no sets of fees. There's no covenants, no early repayment fees. It's very flexible, and that's very, very unusual in term loan financing.
When we looked at other solutions or other opportunities for term loans, the terms of the NHN loan is much better for the business. I think for us it was a no-brainer. It was absolutely the right thing to do. As I say, we went through a very detailed analysis with our NOMAD and it's ultimately been fairly reasonably looking.
I don't have very much to add to that. This ultimately it's a fixed cost at a rate that's very reasonable when you compare it to other financings, with a base rate and a margin.
On the 110 content providers that are connected, are we seeing any network effects or increased stickiness amongst telco partners, given that there's 110 now as a result of the deeper integrations with that cohort?
Absolutely. In fact, while we've been doing this is about three or four years now, once the customers have launched subscriptions, we've never seen a churn from a telco. We just don't see customer churn once these services are launched. Like you say, they're very sticky, they're very deeply integrated. I think for us, with the network effect, it's actually sort of in the referrals. I think we are known now as a company of super bundling. I reach out on LinkedIn and people would say, I've been talking about bundling, and everybody I talk to mentions Bango. Can we discuss it? That's not an unusual thing to happen at both trade shows as well as through direct mails. Continente in Portugal, I think another great example of that was a referral from Disney. There's definitely a network effect from within this.
We have telcos who very much support our business with also other telcos who reference calls. Fundamentally, the nice thing about where we sit, it's in everybody's interest that there's a platform there that makes that bridge between the subscription world and the channel easy. It's in everybody's interest to make that platform successful. There's a natural network effect, and I think that allows us to build this business and grow it with a relatively modest sales and marketing investment. For sure, there's absolutely a network effect.
Matt, back to you. What's the risk of the weakening U.S. dollar to Bango's results?
Good question. There's obviously quite a bit of volatility in the dollar at the moment. In terms of our cost base, obviously a lot of it is people, a lot of it is based here in the U.K. There is naturally some impact between the U.S. dollar and the pound. I think it's also important to highlight there's a natural hedge given the number of currencies that we earn in the regions that we generate. It's a balance. We currently don't hedge, but it's something that we always keep under review and continue to evaluate.
Back to the high cost of sales routes. You mentioned at the beginning of your section, can you provide more info on them, what they are, how much of revenues do they account for?
If you have, we call them routes in terms of telco and merchant, Bango obviously takes a revenue share in that transaction. In the high cost of sales routes, Bango actually sits in the middle of the payment flow. Cash moves from telco to Bango to merchant, as I sort of talked about before. We acquired these routes when we bought DOCOMO Digital. The margin on them is very low. Again, picking up what I said before, the gross margin in that core transactional business is typically very high, sort of circling 90%. The margin here is very low, and it's driven by a number of factors. I think in these particular routes, there's also a cost that we incur working with third parties who help us deliver that service into those markets. That is a primary factor that drives a higher cost of sales.
One on revenue growth. Revenue growth in 2024 was much lower than the CAGR figures in the presentation. Given the consensus are low still, how do you see growth over the medium term?
Let me try that. Let's talk about the digital revenue stream, b ecause that's really where the growth is. The payments business is stable and grows as the market grows, and the real growth is in that Digital Vending Machine business. As Matt said, that segment that we reported in included in 2023 a full year of the audiences business. It only discontinued in Q1 2024. If you strip that out, the growth of that DVM business was 40% that year, which is higher than the sort of the CAGR over the previous time. I think, again, we'll do a better job of making that sort of growth clearer. The thing is, in a business that's quite early on in the stages, by default, lumpy, one or two big customers can make a material impact.
A big customer delaying a launch by two or three months can make a big impact. There's lots of lumpiness at the start. Eventually, as we get the scales over the next two or three years, the law of averages will help out, and we'll start to see more predictable growth. For us, DVM is a growth engine. We see easily double-digit growth in terms of DVM moving forward. In terms of numbers in the market, I think we've tried to guide somewhat on the conservative side, knowing that this level of lumpiness and the impact of one or two customers shifting a quarter can have a big impact on the overall financials. Ultimately, that will play out once we get the law of averages and scale things out.
Question from a relatively new shareholder. Simply put, what is the business model? How do you make money from transactional business and DVM? What CapEx is involved for new clients on DVM, and are there exit costs for you and recovery? Just before you answer, I would point you to the annual report. There's some really good content in there that breaks out the two business models. That's quite a huge reference point, too.
I think there's lots more on the Investor website as well. I'll try and answer all these sort of very quickly. The transactional business, we make money by taking a percentage of the retail price for processing that transaction. That's anywhere from 1%-3% for digital goods, down to sort of about 30% for physical goods. That's on the transactional business. On the DVM, we charge the retailer a license fee. There's a setup fee for the initial integration. There's a license fee. That license fee is scaled based on the number of subscriptions we manage. Not the number of users, but the number of users times the number of services that they have, and that's done in sort of tiers. That's what creates this recurring revenue model business. There's really no new CapEx for new clients.
That's not the way we operate. Like I said a while ago, we're not a services company, we're a product company. The reason it's CapEx is because we're building a product. That's what the accounting rules say we treat it, and it's amortized over time. We build wherever, even if it's something specific for a customer, we will try to build it in a way that gives everybody benefits and build it in as part of the overall product roadmap. There's no CapEx for new customers. Contracts for DVM are typically three-plus years, many of which auto-renew. Some auto-renew for a year, and automatically roll over. Some renew every three years for a further three years. It really does vary. We're saying we really never experienced any churn. All our contracts have continued to renew once those services have been live.
Question on the need for such a large financing funding facility. Given the heavy listing on the platform that's been done, why is CapEx not slated to fall more significantly?
I think on the funding facility, this is a revolving credit facility that Bango can dip in as, and when it needs to. While the size of it is $15 million, it's not the case that Bango draws on the facility for $15 million year one. It's there to provide the flexibility to do the things that we need to do over the course of the next couple of years. Then in terms of the CapEx point, I think a good analogy is akin to we've built the core product. We've built the car, essentially. It's about fine-tuning that car to optimize the car. We talked about the three drivers there around increasing the stickiness of customers, increasing the revenue per user, and reducing the integration effort on Bango side. That's where that CapEx is being deployed.
Back to the U.S. and Trump. Do you see any impact from U.S. tariffs?
I can take that one. The answer is no. The tariff policy targets primarily physical goods, and it's not targeting digital services. No impact to Bango.
What are the growth expectations, which you expect to meet and then beat in 2025 and 2026?
I think we'll make sure that you'll see those on the website. They're certainly from a top line from a DVM perspective. Conservative, but you'll see the consensus we publish on the website.
Matt, can you discuss how you see free cash flow developing over the next two years?
We've touched on a lot of the themes during the course of the presentation and how we expect to see meaningful cash generation in FY 2026. I think I'll guide people to sort of the broker coverage and the consensus forecasts that are in those reports.
Last question. Have you considered increasing the equity research coverage of Bango?
As we said, we have two analysts covering at the moment. Obviously, our broker, Singer Capital Markets, and obviously Canaccord. Getting increased analyst coverage generally comes with a fee. We will talk to, and we do talk to lots of analysts in the hope that they will initiate coverage on, but more than likely trying to initiate a business relationship and come with a fee. The various companies that do research are very much focused on private investors, but again, it's paid-for research. I think it'd be interesting to hear your feedback on that, and maybe that's one to comment on the new InvestorsHub. We've never thought that's the best use of our resources is to pay somebody to do research necessarily on us. There's a question of how independent that is, although all the analysts will say they're completely independent.
It's not obvious that's the best use of our resources. Interesting to hear your feedback. As I say, we do try and make the consensus available. As I say, if you read Research Tree, you will see at least part, I think, from both the Singer Capital Markets and the Canaccord reports.
Great. Thank you. I'm aware we've run over slightly, but thank you for letting us get to all the questions. I will hand back to Jake now to wrap up.
Perfect, guys. That's great. Thank you very much indeed for being so generous with your time and addressing all of those questions that came in from investors. Of course, if there are any further questions that do come through, we'll make these available to you immediately after the presentation has ended. Paul, perhaps before, really now just looking to redirect those on the call to provide you with their feedback, which I know is particularly important to yourself and the company. If I could please just ask you for a few closing comments just to wrap up with, that'd be great.
Thanks again for everybody's time. Obviously, we've overran. I really appreciate you staying on. I really, really appreciate the questions. We really do appreciate the investor engagement. I know sometimes it feels like we don't want it, but we really want the questions. We really want the feedback. We really want your thoughts. That's why we've launched the InvestorsHub. That's why we spent the time here this morning. We're really excited about the business. DOCOMO Digital acquisition has certainly distorted the financial profile for the business over the past few years and made the story maybe a little difficult to explain to the equity market. I think we're coming way beyond that now. I think the progress in Digital Vending Machine is clear. We've got Matt here, who's a great addition in terms of driving that strong financial discipline.
Introducing the new strength in the balance sheet is great. The sort of first step, focus on profitability. Couple all that together with the market opportunity, we're really excited about the future. It's a really interesting space that we operate in. The sales floor looks really exciting. We've seen that acceleration of deals already in the first half of this year. W e look forward to seeing what else will be delivered over the next few years. Thanks again for your time. Really do appreciate it, and please stay in touch through the website.
Perfect, Paul. That's great. Thank you once again for updating investors this morning. Could I please ask investors not to close this session as you will now be automatically redirected for the opportunity to provide your feedback in order that the management team can really better understand your views and expectations. This will only take a few moments to complete but I'm sure will be greatly valued by the company. On behalf of the management team of Bango PLC, we would like to thank you for attending today's presentation. That now concludes today's session, good afternoon to you all.