Good afternoon, everybody. Thank you for joining. First of all, I appreciate—I recognize quite a few of the names on the list and recognize that. I appreciate that you—some of you will have heard the background before, but for the benefit of those who have not, I'm going to explain a little bit about what we do before I launch into the results themselves. By way of introduction—oh, here we go. I'm Louis Hall, CEO and founder of the company way back in 1999, and Cerillion was formed as an MBO from a—well, at the time it was a large U.K. software house called Logica. Andrew, quick introduction?
Yeah, sure. Thank you, Lou. I joined the company just over three years ago as CFO. I previously spent seven years working at a company called Videndum. Before that, I was at Smith's Group, the FTSE 100 company, and I started my career out at Deloitte.
Great. Thanks, Andrew. Before we get into the detail of the half-year results, I'm going to give you a little bit of background. As I said, Cerillion was formed as an MBO out of a U.K. software house called Logica. Essentially, we MBO'd the telecoms products part of Logica. It was a very small part at the time, but a unit that we could see had tremendous opportunity. In terms of what we do, we provide telecoms companies with CRM, charging, and billing software. This is software that enables them to monetize their network assets and connect with their customers. Oops, I've got the right slide here. If you look at where we sit in the chain, let's come back. Our software is called in the industry BSS or an OSS software.
It's essentially the bulk of the enterprise software that fulfills that function of connecting telecoms businesses' network infrastructure assets with their customers. We're talking about everything from defining the products they sell to their customers across all different service types, whether it's mobile, fixed, broadband, TV, etc. The software that supports the sales channels, whether that's retail outlets or call centers or online self-service mobile apps, etc. The software enables telcos to onboard their customers. Once those customers are onboarded, the software talks to the network side of things and connects those services, makes them live on the network.
Once those services are live, the software that monitors in real-time customers' use of those services, manages their balances and so on, has this customer got enough credit to start this internet session, for example, or to download this streamless TV program or to make this call. That is all happening in real-time. At the end of a billing period, the software that sends out bills, that then manages receivables, collects payments, etc., etc. A very broad range of software. We sell this, as I said, of different modules addressing different areas of that functionality within the backdrop of an industry-standard framework that is a global standard. Wherever you are in the world, in the world as a telco, you will be working to these same standards.
Truly global business makes it relatively easy for us to sell internationally, which is where most of our business is. Dropping back to higher level—sorry, if I get this to work. There we go. We have around—oops, sorry, I am jumping around with the slides going on here. We have around 75 customers across 45 countries. We initially had private equity backing. We then IPO'd in 2016 at a market cap of only about GBP 22.5 million, and we have since grown the market cap over 20 x to a market cap of around GBP 500 million. That is a very quick overview as to what Cerillion does and where it came from. I will now talk a bit about the highlights.
Obviously, if people have more questions at the end on what we do and how we do it, obviously, happy to talk in more detail if people have specific points they want us to answer. Now moving to the highlights for the six months to 31st of March. First of all, revenue and earnings were a little lower than the same period last year in H1 2024. Really, this is a timing issue to do with where software subscription agreement extensions and renewals land. Last year, the majority of these extensions and renewals occurred in H1, whereas we anticipate this year that most of these will land in H2. In terms of the overall shape of the year, we're confident that we'll meet market expectations for census forecasts and grow by around 12% and meet the earnings forecast.
I just want to put a couple of reasons why we're confident about this that underpin that confidence. First of all, the major new contract we signed with a new customer for $11 million in January. This was one of the national telcos in Armenia. Because that was signed in January, not a lot of that revenue will have had time to be recognized in the first half, which ended at the end of March. We'd expect a reasonable portion of the services revenue to fall into our second half of this financial year. Now, the project is actually underway, and we're doing the services work. Also, the license revenue for this customer, for this project, we would expect to land in the second half as well. That will make a significant contribution.
We'll just pull out as well that we recently signed in April, just after the end of the half, a new implementation service agreement with one of our large European customers to support the migration of a new mobile base that they've acquired in their home market. That GBP 8 million deal will also contribute services revenue in the second half. Of course, having signed that a few weeks after the end of the half, that increased our back order by 23% year-on-year, as opposed to the 7% increase we actually reported at the end of March. Just the timing when that deal landed made quite a big difference to our back order increase.
The other interesting thing about this deal is that what it does not include is the extension of the existing subscription agreement with that customer to extend the right to use the software to the much larger base that they will now be supporting. This migration of the new acquisition in their home market will more than double their customer base. That will lead to a significant expansion of their subscription agreement. We would expect the license element of that, which would be quite a substantial amount of that, to land in the second half too. In addition to that, our new customer pipeline remains steady, remains strong. It is up a little bit, 3% year-on-year. We are confident we will see some traction on that pipeline in the second half too.
Whilst new logo customer wins that we close now in the second half will not make a major contribution to the second half because, obviously, by the time we start projects and recognize license, we are going to be more into the first half of next year. They are really important for powering next year. That is still important. I think the flip side of that is we are not dependent on any major new logo new customer win to achieve our targets for the 2025 financial year. Of course, all of that has given us the confidence to increase intrinsic dividend by 20% for this period. I am now going to hand over to Andrew, who will talk about the KPIs in some more detail and the rest of the financials.
Great. Thank you very much, Louis. As Louis has said, revenue in the first half was down slightly on the prior period. This was due to the timing of license revenue against what was a relatively strong comparator. As Louis has flagged, we are currently expecting this to be a year of two halves with higher revenue in the second half of the year, driven by higher license revenue, which overall should lead to year-on-year growth. I think we've just lost the slides, but I'll carry on talking and hopefully the.
We're working to get those back now.
The slides should come back shortly. At the same time, the decline in profit was actually slightly lower than the decline in revenue. We did achieve favorable operating leverage. This was driven by a number of factors. The biggest factor was actually the fact that we achieved higher day rates on some key implementation projects. I'll talk about this and the other key factors in a bit more detail on the next slide. What this effectively meant was that the adjusted EBITDA margin remained very strong at 47.7%. Maybe if we just sort of go back to that slide. The adjusted EBITDA margin remained very strong at 47.7% despite the lower level of higher margin license revenue that we recognized in the first half of the year.
In terms of recurring revenue, you can see we do have a track record of recurring revenue increasing over time. In fact, recurring revenue overall has been growing faster than total revenue. Over time, we have been becoming a higher quality business. The growth that we saw in the current period of 8% mainly reflects higher levels of support and maintenance revenue recognized in the period. As we have flagged in previous presentations, this would be higher if we included the impact of term license renewals. In terms of the back order book, the graph in the very middle, this increased by 7% up to GBP 50.2 million at the end of March. After taking into account the GBP 8 million services contract that Louis referred to earlier on, this actually increased by 23% to GBP 56.5 million as at the end of April.
This number is made up of two main components. Firstly, GBP 46 million of orders that have been contracted. We're flagging that we estimate around 45% of this balance will be recognized as revenue within 12 months from the end of April. On top of this, we've got GBP 9.6 million of annualized support and maintenance revenue. New orders in the period came in at GBP 19.6 million. This includes the impact of the new major contract with a new customer for $11.4 million, which we announced in January. Most contracts continue to be with existing customers. I think it's interesting that these continue to offer significant upsell potential. In terms of net cash, you can see we do have, again, a very strong record of cash growth. Cash continued to build in the period.
It was up to GBP 31.2 million as at the end of the period. As Louis said before, this has enabled us to increase the dividend per share by 20% up to GBP 0.048 per share. Now turning to the financial highlights, we've already talked about the impact of total revenue. If we look at the impact on a segmental basis, you can see that we did recognize 24% more services revenue in the first half. This was more than offset by lower software revenue due to the timing of license revenue recognized. You can see the very high margins that we reported in the period. The gross margin was actually slightly higher at 80.6% versus 80.4% in the prior period. The adjusted EBITDA margin, again, holding up very strongly at 47.7%.
I said before that the main factor behind this was the higher day rates that we achieved on some key implementation projects. The story behind this is that effectively, we have been operating more efficiently as an organization than we did in the prior period, where we did experience a couple of overruns on some major projects. These were not repeated in the current period. We also experienced a favorable mix in terms of services as well, with a high proportion of time and materials work, where we typically get a better day rate as well. That was one factor. We also capitalized more development costs. This reflects the fact that we invested around 30% more effort into R&D in H1 2025. This meant that we were able to put more of those costs on the balance sheet as opposed to going through the P&L.
We also experienced continued favorable headcount mix. Whilst we did increase headcount across the group, and as you would expect, everybody got pay increases where they were eligible to do so. If you look at the payroll cost period on period, they were actually flat because we recruited more people, in particular in India, where the costs are significantly lower than in the U.K. We also continued to recruit strongly in terms of the entry-level roles at the graduate level. Overall, this favorable headcount mix was very beneficial from a margin perspective as well. The final factor I'd like to pull out is the impact of FX. We did see some element of favorable effects in the period too. Now I'd just like to look at cash in a bit more detail.
As usual, the table at the top shows a reconciliation of adjusted EBITDA down to free cash flow. You can see that despite the low level of license revenue, we did see a small increase in working capital. This was mainly driven by building accrued income due to the timing of revenue recognized versus cash received on some key implementation projects. As you would expect, because we recognized significantly less or slightly less license revenue than we did in the first half of 2024, there was a lower build in accrued income and therefore working capital. The second point to pull out here is that, as I said before, we did capitalize more development costs, so roughly GBP 900,000 versus GBP 600,000 in the prior period. This reflects the higher level of investments made.
The graph at the bottom shows a reconciliation of opening net cash through to closing net cash. I think the key point here is that the free cash flow that we generated of GBP 5.9 million was higher than the amounts that we spent on dividends, the amounts that we spent to purchase employee incentive shares, and a relatively small amount for lease payments as well. Okay, in terms of the detailed income statement, I think the key point here is that we do continue to invest in R&D, as I said before. The total investment was up over 30% on the prior period. It was up to just over 8,000 days that we worked in the first half. Operating expenses remained broadly flat.
The main reasons for that were increases from higher headcount, inflation, and higher depreciation and amortization were offset by higher capitalization of development costs, continued favorable headcount mix, and favorable FX. Finally, the next slide shows the consolidated balance sheet. Again, the key point here is that the balance sheet remains very strong. You can see net cash of GBP 31.2 million. We do not have any debt on the balance sheet. You can see there has been some increase in the period due to new leases that we have signed, in particular in our India office and in our Bulgaria office. This explains why the right-of-use assets and the lease liabilities have increased slightly from the prior period. Back to you, Louis.
Thanks, Andrew. I think really just to reiterate the general points at the beginning, we've spoken about the weighting of H1 to H2 and the fact that post the period end, back order is up significantly and other reasons why we have confidence in that H2 outcome. Also, equally important is new logo wins looking ahead to next year. We think we're in a good position there. I think in general, our model in the market remains pretty unique. We are providing a software as a service solution based on a pure product strategy. All customers have the same software. It's an off-the-shelf product. It works on day one in a market that has traditionally been dominated by big vendors that are looking to provide a much more services-heavy solution.
Where they're starting with some kind of product framework, but delivering a quite heavily tailored, bespoke solution to each of their customers. Of course, the big advantage we have in that market is that our approach, because it's based on standard product, is much more cookie-cutter. It's much more cost-effective. It has a much lower total cost of ownership, a much faster time to market, and it's much more flexible. I think that does put us in a strong position in a market that is estimated to be at around $70 billion a year at the moment by one of the main analysts out there. Huge market for us to expand into based on our current revenues, a well-differentiated competitive position.
I think all of that positioned us well to not just achieve the current year market expectations, but to go on growing significantly next year and beyond. I will pan back to Gareth. I appreciate there is lots of other detail we could have gone into on the company. If you do have questions on any of that, direct me to the area you would like to know more about, and we will do our best to respond. Thank you very much.
That's great. Thank you very much, Louis. Thank you, Andrew. We will now move to Q&A. Just to remind everybody in the audience, please be aware you can ask questions by typing in the Q&A button at the base of your screen, and management will attempt to answer your questions this afternoon. We've got a number of questions already being submitted. First of all, a couple around acquisitions. Could you comment on what kind of potential acquisitions you would be interested in? In the case of an acquisition, do you have a minimum hurdle rate that you would look to achieve?
It's a good question. I'm sorry, I'm trying to step back through the slides. Essentially, if you think about the diagram we show with our different modules that fit the industry standard BSS OSS map, the sort of acquisitions we're looking at are bolt-ons or tuck-ins that would be focused around businesses that can offer additional modules that we could bolt on around the edges of our current offering. Obviously, the benefit of that is that will bring with it telco customers we do not have today. A priority for us is to bring on board some bigger names, some bigger logos, and so on.
Adding boxes around the edge of this diagram is to not just bring additional product into our products that we can upsell, cross-sell to our existing customers, but to bring in new logos we have not already got that would create some opportunity for us to upsell the rest of our main product suite into those logos. I mean, one thing we did not spend any time on actually in the presentation was customers. We do have a very broad range of customers, as you can see from the logo. Oops, bear with me. The logo wall, some tier one names you will have heard of, but a lot of other telcos in the world that you will not have heard of.
We have everything from main network operators like Virgin Media, for example, and Orange, and so on, to MVNOs, MVNEs, messaging businesses, power companies that have become telcos in their own right, etc. A very broad range. The more we can diversify that range for acquisitions and create more opportunity, the better that is. Hopefully, that answers that question in a perhaps a little bit too expansive way.
That's fine. It's very helpful. Another question here. This one may be for Andrew. You've increased the interim dividend by 20%. If you meet market expectations for profitability and cash flow for the whole year, would you expect to increase the final dividend by a similar amount?
Yeah. I mean, I don't think at this point we can commit to exactly what's going to happen at year-end. I mean, we do have a progressive dividend policy in place. Therefore, we do plan to continue increasing the dividend. I don't think we can commit at this point to definitely increasing the final dividend by 20%.
It'd be fair to say it'd be a reasonable expectation.
Hopefully, but it depends on final performance and board approval.
Of course. Understood. Okay. Thank you for that. We've had a couple of questions around competition. Could you maybe touch a little bit on the competitive landscape and what's the difference between your vertical solutions, some of the others offered by the competitors? In particular, a couple of questions have been asked about Amdocs. Any sort of compare and contrast there would be very useful.
Yeah, that's a good question. Let's jump to that. Apologies, it does take a while for the slides to change. Oops. Apologies. I was trying to get to a slide which actually breaks out the different types of competitors. Essentially, we've got four main groups of competitors. The large independent software vendors, these people on the right-hand side, called on the right-hand side. That includes Amdocs, who are the biggest vendor in terms of revenue in this market, about $5.5 billion turnover, I think. They're not the biggest company in the market. Obviously, others like Ericsson and Oracle and Nokia and so on, bigger companies, but they have the biggest amount of business in this market.
The differentiator there is that what I was saying earlier on, we are a product one-size-fits-all solution, off-the-shelf, as opposed to vendors like Amdocs that offer a much more tailored, bespoke solution. His strategy is to maximize services revenue, whereas we want to be a software business, a product company, and our strategy is to minimize service revenue to a large extent and focus on the recurring software revenues. Of course, as I was saying before, that gives us a much faster time to market, lower long-term cost of ownership, and a more flexible solution that not just is flexible on day one, but can be upgraded seamlessly going forward, which is much harder with a bespoke tailored solution. You have then got the network equipment vendors. That is essentially Ericsson, Nokia, and then Huawei and ZTE from China. Ericsson do have a competing product suite.
They have a product suite in this space. It's been around for a long time. It was put together through acquisitions. I think it's not really a focus for them. Software is not really a focus for Ericsson right now. The focus is more on the network side. They have got a big customer base and a lot of salespeople around the world. They are a competitor. Nokia are a partner. Nokia don't have all of the boxes on that diagram we looked at. They sell some of our boxes as part of their suite. For example, a big project we have with Orange in Egypt for the new Egyptian capital city, all the telecoms infrastructure, the smart city infrastructure, if you know what smart cities are all about. That was a contract through Nokia.
With ZTE and Huawei, we do not really see them in Europe, obviously, or North America. They do compete with us in Asia-Pacific, for example, but they are not as—I mean, they used to win on price. They are now not as cheap as they used to be. In fact, they are not cheaper anymore. We do stand our ground against them. Again, they are not really offering a product approach. We tend to win on that gambit. That is our differentiator. There are some small independent software vendors. I say small, more our size, or at least of the same order of magnitude. They tend to be more regionally focused. There are some in India, particularly, that are price-focused. Their USP has just been the cheapest.
They're typically playing more in emerging markets where we're not really trying to compete in emerging markets for all sorts of normal reasons. It's hard to get paid. It's hard to get the work done, etc. We don't play in those markets. There's another kind of a newer set of competitors that are really kind of multiple vendors getting together to offer a, so this is multiple SaaS vendors, software as a service vendors offering a software as a service approach like we do, but with different components for different functions. You're talking about the old-fashioned best-of-breed approach, but on a SaaS basis. Of course, it comes with all of the baggage and the problems of the old-fashioned best-of-breed approach where you're talking about vendors whose modules aren't designed to work together necessarily.
You need to have an integrator come in and put all that together. You need to keep it in step over time as different vendors evolve their roadmaps at different rates and different directions. That is a much more complex, more expensive, higher risk, longer time to market, less flexible solution in the end. There is lots of differentiation there. That is really a quick summary of the competitive landscape. I think also it is probably, I mean, just someone asked about Amdocs. If you look at the Amdocs margins, the lower margins and the higher service revenue content do kind of highlight the difference between our model and their model.
That's great. Thank you for that. Yes, there were a couple of other questions which kind of related to some of the things you've mentioned as you've gone through that competitive landscape. One of them in particular was about Egypt. You mentioned the Egypt project. The question was, how is the smart city project in Egypt progressing, and have you received positive feedback?
Yeah, I mean, absolutely. I mean, the smart city itself, it's called ACUD, the Administrative Capital for Urban Development. It's not a they haven't settled on a good name yet. There is a lot of the core of the city that exists in the desert. A lot of buildings have gone up, government buildings, accommodation, mosques, community centers, and so on. It is a real project. There are real people living there who are being billed through the Cerillion Nokia infrastructure. What's interesting is, because it's a smart city, it's not just about telecoms billing. We're also doing billing for all of the utilities. Power, water, gas, smart meters, and anything else that's sold as a municipal service, like, for example, car parking, cinema tickets, public transport, all those things that have to do with the municipality are being managed and billed through Cerillion.
It's mostly complete in terms of a project. Really, it's now a case of waiting for more people to move in. One of the challenges the Egyptian government has is getting people to move from Cairo, which may be noisy, crowded, congested, but people are somewhat reluctant to move out into this completely new place, which I guess is sort of inevitable. Our work is mostly complete, and feedback so far has been very positive.
That's great. Thank you. You also talked about the services proportion of your business relative to some of the competitors. We have a couple of questions about the services revenue. First of all, is there any aim to reduce the proportion of services revenue over time? Secondly, for the services revenue, how much would you say is recurring, and how much is related to new implementations?
I think I'll just say very quickly that our goal has become more of a software, more of a higher software content business, to increase the proportion of software revenue over time. I think as we add more customers and more subscription agreements, then the concentration of services revenue, which is a lot of which is derived from implementing new customers, will become a smaller proportion as the customer base expands and grows. I'll let Andrew handle the detail part of that question.
Yeah. In terms of the split, I mean, we do disclose that the services revenue in the half was GBP 10.3 million. At the same time, we disclose the recurring revenue element was GBP 8.2 million. The recurring revenue is made up of support and maintenance, managed service, and third-party hardware and hosting revenue. I mean, those two splits are relatively equally spread. I think, as I said before, what we do not include in our recurring revenue is the impact of the term license renewal. We have a sticky customer base. A typical contract for us is a five-year term contract. At the end of the five-year contract, we would fully expect the customer to renew. At the period at which the customer renews, we can then recognize that license revenue upfront all over again. That is not currently included in the recurring revenue.
Effectively, the metric that we're showing as recurring revenue is very prudent. I think one could argue that the true recurring revenue is higher than GBP 8.2 million.
Okay. That's great. Thank you very much, both. Slightly different angle of question. Could you please elaborate on what kind of operational improvements you were referring to when you say they helped improve day rates for implementation projects? Are you able to give any specific examples?
Yeah. I mean, let me start on that. I mean, I think the key point here is that most of our new implementations are based on a fixed price. We have a fixed price, and then we have to perform the work until that is done. Now, clearly, when we're pricing a project, we have to put an estimate in place as to how long the project is going to take. Sometimes a project takes slightly longer than we're estimating. Sometimes it takes slightly shorter. I think the key thing that we're saying here, the reason the day rates have improved period on period, is that in the prior period, we did see a couple of projects which overran, whereas the opposite was true in the current period.
Actually, there was one project in particular that has been underrunning, and therefore, we're able to recognize slightly more revenue against it. The way we recognize the revenue in each period is on a percent completion basis. Therefore, if the project takes a shorter time than we're initially expecting, that leads to a greater degree of revenue being recognized, and therefore, the day rate goes up. I think that's probably the best specific example I can give on that.
Okay. That's great. We had, in some ways, a related question. How strong is your pricing power on recurring revenues? I.e., do they outpace inflation around the world?
Yeah. In terms of pricing power, I mean, I think the key point here is that we do have indexation built into most of our contracts. What that means is a typical contract over five years. In each of the five years throughout the contract, we are able to increase prices in line with inflation. Now, when it comes to renewing the contract, as at the end of the period, only you could argue we do have a decent amount of pricing power because we have a very sticky customer base, and it can be difficult for customers to move away from us. At the same time, it's really important for us to build long-lasting relationships with customers.
I think we do treat customers fairly, and we typically put prices up in line with inflation in order to make sure that customers do stay with us for the long term. I think there are a lot of examples of that. If I look at our customer base, there are some customers that have been with us for over 20 years. We do have a good track record of looking after our customers. I think that has worked in our favor in the past.
That's great. Thank you. We've got a couple of questions about evolving technologies. Maybe it might be worth answering them sort of in sequence. First of all, how are you integrating cybersecurity into your software solution? Then separately, what would you say are the risks around potential AI developments from some of your competitors?
Cybersecurity is something which is integral, really, to everything we develop. We are compliant with ISO 27001, for example, on security. There are a lot of fundamental principles that we have to have our developers adhere to. Of course, that's just an ongoing huge focus, I think, to be honest, in most software businesses. The way we write software is governed by that, but also the way it's deployed. When we're hosting the solution, obviously, the firewalls that we have in place and all of the deployment protections are also key. We have regular testing with external specialist third parties who do penetration testing to see if they can break our firewalls and find weaknesses in the software itself that may lead to cyber attacks. That's very much baked into everything we do.
AI, I think AI is really important in two ways. First of all, it does offer a big opportunity to increase productivity. We are using AI tools in software development, but not just in the writing of software, but also generating the tests of software we write and then running those tests. The next step along that journey is to have AI actually correct the errors it finds in the software it tests that to some extent is written itself. There is massive opportunity for big productivity improvement in software development. I think we are already at the point where really nobody writes a new line of code in Cerillion anymore. You cannot say to AI at the moment to these tools, "Write a new charging platform," but you can tell it to write modules within that charging platform.
It is quite fascinating the power that is there to, as I was saying, I guess, to get productivity improvements. Of course, the other side of AI is AI features in our product that we offer to customers. We have been really heavily focused on that. In Barcelona, at the big Mobile World Congress trade show, we are demonstrating some of these new features that are already in the product. For example, the ability for a telco customer to query either verbally or in a sort of natural language request to a chatbot, "Why is my bill higher this month? It is $50 more. What is going on?" The intelligent assistant we have developed will then go through the calling history, the calling patterns, the charges, and so on, and come back with an intelligent answer. For example, you were roaming in Barbados because you were there on holiday last week.
Oh, of course I was. Fair enough." That sort of thing may sound simple, but it's actually a more powerful, a better answer than most call center representatives will be able to give because they just don't have the ability to scan all the data fast enough. Of course, it's an answer that doesn't need a person to provide. You get a productivity gain. You get a customer service gain. These things are really important to telcos. I think we're at least as far ahead as anybody else is in this. Any of our competitors and a lot of our R&D does get spent on AI on the product side.
Okay. Thank you. Another question about customers and in particular new implementations. Generally speaking, how long do implementations on new customers take?
It depends a lot. To be honest, the biggest limitation here is the speed the customer wants to move at. Because we're starting with a product that works out of the box on day one, it's much quicker than most of the competition. I mean, there's a lot of work to do. We still have to configure all the products the telco is selling. Most of the time, we're not dealing with greenfield. We're dealing with existing customer bases and migrating those. We've got to map all the products the telco customer currently sells. Also, products they're no longer selling, but they have existing customers using still. We have to model all the workflows around the business to how to get customers on board, etc., and deal with all the customer service issues that might arise. We have to then do data migrations.
We've got to typically extract data from multiple different systems to consolidate onto our central platform, multiple legacy systems converting into one new system. We have to do all of the kind of network integration, network integration testing, and so on. That is normally at least nine months. It can often be 12 months, 15 months, sometimes 18 months. Of course, sometimes telcos will want to do this in multiple phases. They'll want to do maybe the fixed wire base first, then add the mobile customers, or do it by brands or whatever. It does vary, but typically, it's about 12-15 months.
Okay. Thank you. Another question. How do you select and incentivize local managers in new countries? Most importantly, how do you make sure they then act with integrity?
We have three main operating bases: London, Pune in India, and Sofia in Bulgaria. We have not got lots of different operating units to worry about. We have salespeople distributed more widely, but obviously, they are all reporting into a central oversight function. I think in India, we have head of India who has been with us for 22 years or so. Fundamentally, all of the kind of financial, if the concern is about how do we control the financial integrity, the financial control is all centralized in London in terms of payments over certain thresholds and all the rest of it. The same is true in Bulgaria, which is our newest office. We have only been active for about three years, but it is a relatively small team so far. Again, all of the kind of financial transactions are effectively managed out of London.
I think, does that answer the question? I'm not sure I've quite hit on the right vein there.
No, I think it does. There is actually another one, which, depending on the answer, may also be related to people or certainly to operations. What's the major cost inflation that you're currently noticing across the business?
Yeah. I mean, in terms of cost inflation, I guess more widely, if you look at our cost base, the biggest cost that we have across the business is headcount costs. Now, if I break that down, as Louis said, the main regions where we operate are in the U.K., in India, and Bulgaria. I think it's fair to say that within those operating areas, the largest push for inflation has been in India over the past few years. I think at the same time, I think it's fair to say the inflation that we have seen in India has been lower this year than in previous years. I think that's partly due to the sort of redundancy levels that we have seen in some of the large tech companies.
When I joined about three years ago, I was surprised that if you made five offers, maybe on the first day, you might have sort of one or two people actually turning up to work in India. That has definitely changed. I think that the people have less bargaining power in general, and therefore, we are seeing a reduction in inflation in that region.
Okay. That's great. Thank you. We've got a last couple of questions already being submitted. Just as a reminder to the audience, if you do wish to ask a question, please do type it in the Q&A tab at the bottom of your screen. The last couple that we've got, one of them relates to Starlink. Do you see Starlink as a major disruptor of the telco market, and could it have an impact on Cerillion?
Good question. I think it could be a disruptor of the telco market in that it may take business away from mobile business, away from some of the major providers. It probably does not really affect us because at the end of the day, it is just another transport mechanism. Whether the network is satellite or terrestrial or whatever, we work with every kind of network: mobile, fixed, TV, broadband, satellites. For us, it is just more of the same. I mean, will low-Earth orbit satellite networks significantly disrupt telcos in general? I would say the jury is still out, to be honest. At the end of the day, there is always going to be a latency constraint and a bandwidth constraint. Although more satellites closer to the Earth makes that less of an issue, it is still hard to see that being as effective as terrestrial networks based around fiber backhaul.
We'll see. I think the jury's out.
Okay. Understood. Thank you. The last question we have is, do you still have the ambition to double revenue or size within the next five years, and how confident are you in achieving that target?
Yeah, absolutely. I mean, that's certainly what our target is to do that within the next three to five years. I think we do have to execute. We have to close deals. We have to keep expanding. I think we have a well-differentiated position, and that's certainly our goal.
Yeah, that's great. Thank you very much. This is the end of the webinar. We won't be taking any further questions. Louis, if you've got any final closing remarks, please give those to the audience, and then we'll finish the webinar.
Great. Just thank you all very much for joining. Appreciate your time. If anybody has any follow-up questions, feel free to submit those. If you already know us, obviously direct or through your broker or through Gareth and Progressive, and we'll be happy to follow up with you separately if that would be helpful.
That's great. Thank you very much, Louis. Thank you, Andrew. For the audience, please do submit the feedback forms that you will receive from us. It's very useful for management to get that feedback. Please do take the time to fill in that feedback form. Thank you very much for attending. This is the end of the webinar.