Welcome to the Cerillion interim results webinar. All attendees are in listen-only mode. At the end of the presentation, there will be an opportunity to ask questions. At any stage, please type your question using the Q&A button. This webinar is being recorded. I now hand over to Louis Hall, co-founder and CEO, and Andrew Dickson, CFO. Louis, over to you.
Many thanks, Tamsin. Welcome, everybody, to the webinar. Thank you for taking the time to listen to us this afternoon. I appreciate that some of you will already know the story. For the benefit of those who don't, I will go through the introduction to the company and, you know, what we do, who we are, and so on. I'm Louis Hall, CEO and founder of Cerillion. We founded Cerillion back in 1999 as an MBO from a company that was then Logica, which was quite a large U.K. software house at the time. We went out to the market, raised private equity funding, and invested in building the business over the next 15 years or so. Then exited private equity when we did the IPO in 2016. Andrew, would you like to introduce yourself quickly?
Yeah, sure. I started back in February 2022, so coming on for 18 months ago. Before that, I worked for Vitec for about seven years, and I started my career at Deloitte, where I did my accountancy qualifications.
Thank you, Andrew. What does Cerillion do? We provide the enterprise software layer that sits between telecoms businesses, customers, and their networks. What does that mean? Let's have a little look at that in a bit more detail. This is a diagram of the different product modules that we provide, and essentially, I won't go into this in a lot of detail, but we provide all of the software that enables telecom businesses to create the products they sell to their customers. Bearing in mind that most telcos these days, telecoms companies are selling what we call quad play. That's a combination of mobile broadband, fixed wire, and TV services, in quite complex bundles. That requires some quite complex software.
We also provide all the software that enables telecoms businesses to onboard their customers, whether that's through CRM in call centers or retail outlets, or through self-service where customers onboard themselves or through our mobile apps. We also provide software that then connects those services to the networks, that monitors usage of those services, puts charges for those services onto bills, collects payments, manages collections and so on. Again, other peripheral modules sit around this area. This is absolutely at the core of everything that telcos do. It's not nice to have. It's mission- critical. I guess our USP in the market is that we provide this already pre-integrated together.
Some vendors provide parts of this product set, individual modules or one or two of these modules, and then they have to be integrated together with other vendors modules. That's a complicated, expensive, relatively risky process. We provide all this on day one out of the box, as a service, as very much Software as a Service model. You don't need to have your own hardware or your own IT department. We provide this as a service, working out the box on day one. In terms of the business model and revenue model, if you like, we sell customers a subscription service. That subscription fee covers licensing for the use of the products. It covers the support and maintenance of those products.
It also covers what we call managed service, which is us operating the solution, as I was saying before, on the customer's behalf, and also us hosting that solution either in public or private cloud. In addition to that, we will also, for a new customer, provide implementation services to put that solution into production. Because these are large solutions and they're essentially digital transformation projects across the whole breadth of the business, those projects can consume somewhere between 3,000 and 6,000 man-days of effort. They typically take nine to 12 to maybe 15 or in some cases even 18 months to complete. Having said that's quick in our industry. It might sound like a long time, but that's a very short project for this kind of engagement.
Mainly because we're starting with the all customers having the same product. There is no change to the product when we deliver it to a new customer, or at least there's not very much change. Very rarely we're writing new code. All the work we're doing is around understanding requirements and then doing configuration and data migration. In terms of the revenue model, in addition to a sort of a quarterly SaaS subscription fee, we're charging an implementation services fee to put the software into use. That is recognized over the term of the project. If the project lasts 12 months, we recognize revenue for that over that 12-month period against project milestones.
For the subscription fee, although the customer sees one fee, internally, we break that out into a license fee, a support maintenance fee, a managed services fee and a hosting fee. The license fee we recognize per the IFRS 15 accounting rules once the software is installed, so right at the front of the engagement. Typically, these are five-year term contracts, so customers paying subscription over five years, we're recognizing the license element of that as soon as the software is installed. The rest of those price elements, the support managed service and hosting, we recognize on a straight line basis over the duration of the contract term. Moving on. Great. Okay. We're headquartered in London. We have about 320 people.
We have about 100 in London, 200 in India, in three different offices that we have in India. We've recently opened a delivery center in Bulgaria, in Sofia, where we have about 20 people today, but we're growing that office quite rapidly. We opened this office just to give us a bit more diversity in terms of addressing some of the overheating in our U.K. and India locations. Having said that's obviously calmed down quite a lot since last year, but it's always useful to have alternative sources of resources. We found a good, a good seam of a rich seam of resource to mine in Bulgaria. That's working out well.
It gives us more people who have E.U. passports. A lot of our business comes from Europe and, you know, that's useful in terms of having easy travel. Ease of those people working in the rest of Europe. Today we have around 80 customers in 44 countries. That's everywhere from Australia through to North America. We also have sales presences in Sydney and in Singapore and in Brussels, and we're in the process of putting a on the ground sales resource in North America.
In terms of the high level, some high-level numbers, the split between software and services is usually These are first half numbers, but usually a bit more than half of our software revenue comes from software, and a bit less than half comes from services. This is fairly consistent. Equally, a bit more than half of our revenue comes from Europe, normally. Europe is in our view, by far the most dynamic telecoms market. And the other half is usually split roughly between Asia-Pacific and the Americas. This particular half, we had a fair amount of revenue in Middle East, mainly deriving from a big project we're doing for Orange, to provide software for the new telecoms infrastructure in the new Egyptian capital city.
We do have very sticky customers. Once our customers are on board, they tend to stay with us for a long time. This chart is showing that 81% of our customers have been with us for more than five years. We have customers in our portfolio who have been with us for more than 20 years. It's a big process to move these to different platforms, move customers to different platforms, so they tend to stay once they're on board. Try and advance the slides here. Okay. I think if, you know, Without going into mass amounts of detail, the market drivers behind Why do telecom businesses spend all this money and go to all this effort to have these solutions or to change these solutions?
I think that there are three main drivers. One is revenue growth. There's a huge amount of investment going into, at the moment, going into 5G rollout and fiber rollout. In order to recoup those investments and better monetize those investments, telcos need sophisticated software to help them build better product bundles to get closer to the customer and so on. Revenue growth is a key factor. You know, what we, what we're seeing at the moment is two of the big drivers in this market are digital customer experience.
How do you Not just how do you save cost by getting customers to serve themselves, but how do you make it more attractive for customers to join your service, to come on board as one of your customers rather than the next telco's customers, the competitor's customers. I think that's all about making it easy to use those self-service tools or those mobile apps that enable new customers to do that.
The other thing that we're seeing, which I've already mentioned, I guess, is that this need to be able to build more sophisticated product bundles, to be able to add loyalty schemes on top of that, and then layer discounting schemes on top of that, both in business and consumer, is a key to, again, generating more revenue to recoup the investments made in these huge infrastructure investments that telcos are making. Of course, all the other factors still apply. Obviously operational efficiency, if you can consolidate a number of legacy platforms onto a modern conversion platform that can cover all the different services and brands as our platform can, that's a driver. Of course, technology is always a driver in telecoms. If anything, technology is the driver in telecoms.
Telecoms businesses will always have to be able to offer the latest and greatest services if they're gonna stay in business. That's a factor, too. Of course, not all legacy enterprise software can cope with all the new infrastructure technology. That in itself can drive change and customers to move to new platforms. I just wanted to draw attention to a recent Gartner report for some external validation. Gartner, one of the big business analysts, they cover telecoms very widely.
This is a report that came out just last month in April. It was a vendor survey which we included in, but it also surveyed the telecoms businesses and asked telecoms companies, "What are your priorities in choosing enterprise software?" The software that we provide. Priority number one for the market was digitization of sales and support channels. That's all about this improved customer experience. Priority number two was support for new product types and business models, which is the second thing I was thinking about, i.e., it is all about creating more sophisticated product bundles to earn more revenue off the same basic services. It's interesting to get some external validation from Gartner on what we're seeing at the core base.
In terms of the markets that we're in, so the thing to know about telecoms is that, you know, not all telcos are like BT or Vodafone. There are all kinds of different telcos. You know, we have customers, you know, who are typical telco names you might recognize, like Three in the U.K., KDDI is like BT in Japan, Proximus is the BT of Belgium, and so on. But we also have customers who are not really telcos at all, like, so in the U.K., Neos is the telecoms business of SSE, the power generator.
In Denmark, Norlys is another big electricity company, the second largest power generation in Denmark, but they're also the largest supplier of TV and broadband services to the telecoms market. They've recently, just a few weeks ago, announced that they're buying Telia, one of the big Danish mobile operators, and are going to integrate that into their business as well. You know, that's not a telco you might think of as a telco. Of course, there are all the MVNOs, the brand-based telco businesses. Think about things like Tesco Mobile, for example, we call those Mobile Virtual Network Operators . We have, you know, Emergency Services Networks around the world.
They all have their own telco operations that need all of these software modules. Again, without going into vast amounts of detail, there are all kinds of different types of telco. Just going back to the previous slide, the market. You can segment the market in lots of different ways. You can also look at it in terms of just scale, so, you know, Tier 1, Tier 2, and Tier 3 , and so on. I think for the largest telcos, we are able to break into those through niches. Either by providing one or two of our modules and not the whole product suite, or by providing the whole product suite to one or two of that telco's brands. That's the way we break into the big ones.
The sort of more mid-size, smaller ones, we can provide the whole suite to all the brands. There are different connect points and different places that we can operate in that market, which gives us a lot of diversity. In terms of channels, we also work through the channels. Channels are a lot less important than when I started out in this industry. I think part of that is driven by the SaaS revolution. Because software vendors are providing software as a service mainly now, you know, the concept of going through a third party to have access to the service is not quite as obvious. You know, certainly telecoms businesses, we've seen want to go direct a lot more to the software provider.
However, Nokia is an important channel for us in that Nokia can access some of the markets that it's harder for us to get to on our own, particularly in the Middle East. You know, hence the deal with, we won with Nokia, the Egyptian capital, which we're currently working on, and that's a very interesting project. That would've been hard for us to win with Orange without Nokia being involved. Also, we do work with the big systems integrators, so people like Infosys, Capgemini, CGI in Canada and so on a sort of fairly opportunistic basis. They do give us access to leads that we wouldn't necessarily always see.
They're also potential pools of resource to help us with projects on some of these larger deals. We've spoken a bit about customers. Just point being, we have a very broad and diverse customer base. And we can certainly handle questions on that later on if people want to know more detail. In terms of the competitive landscape, well, there is competition, unfortunately. Keep us all honest. Competition splits into three main groups, the large independent software vendors, so, people like Oracle, for example, who have product suites in this area. And we tend to win against these much larger beasts because we're very focused on, as I was saying before, our product solution. Every customer has the same software.
If a customer wants features we don't have, we won't give them a customized version. We will either introduce that feature through our own R&D program, or in some cases, customers will fund R&D to have that feature, produced, if it's not something we think that we want to invest in or something that we will do at a later date, but not right now. So a fair amount of R&D is customer funded. And that's a useful source of product development. That customer then does not have any IP rights in the software or any royalties. That feature is a product feature that all the other customers will benefit from. That's an absolutely fundamental principle in our business model. Whereas the...
You know, most of these large independent software vendors are providing quite heavily tailored solutions. They'll start from some product framework or a solution from a different site or different customer. But they will then get into quite, you know, heavily tailoring that solution, which obviously takes a lot more time, it's higher risk, it's more expensive. Then the middle group of competitors are the, what we call the network equipment vendors. That's Ericsson, Nokia, ZTE, and Huawei, the main ones in the world. The Chinese vendors are not in our current main markets, Europe and North America. They're still strong in the emerging markets, but we tend not to do very much in emerging markets. Nokia is a partner, so they're not really a competitor.
Ericsson are a competitor, but have somewhat lost focus on their software business. So we're not seeing as much of Ericsson as we would normally do. And then there are some smaller independent software vendors of the same sort of order of magnitude size of ourselves. But they tend to be more kind of regionally focused or focused on particular types of telcos. So we don't generally compete much against the smaller ones. Okay. So a little bit of. Just a quick couple of points about the six-month period we're reporting on, and then I'll hand over to Andrew to go through that in some detail.
You know, we've had very strong trading, I think in the first six months of the year. We have hit record highs on all the key KPIs. You know, Andrew will go into the details, but as you can see, revenue, EBITDA, adjusted PBT, net cash, all up. What I'll just spend a couple of minutes on is the orders and the sales side of things. We achieved GBP 15.3 million or a 40% increase in new orders, compared to the same period last year. This is still lower than the high we had in 2021. The reason for that is that new orders in 2021 were heavily skewed towards the first half.
Total new orders for 2021 was something like GBP 30 million. Whilst GBP 15.3 million is lower than GBP 23 million, we're pretty confident that we'll exceed the total for 2021 in 2023. A couple of points about that new order figure. All of that, all of those new orders came from existing customers. The interesting thing is that we signed two quite substantial deals with existing customers of a scale that, you know, a few years ago, would have been considered to be big new customer wins. One of those we announced at GBP 10 million, the other one we announced at GBP 6 million.
Just to explain, before I go on, how GBP 10 plus GBP 6 doesn't equal GBP 15.3 in our new orders numbers, we don't include any support and maintenance sales. That's because when we take new orders into backorder, we only include a year of support and maintenance contracts in our backorder. That's just so that we don't distort our backorder. Otherwise, you know, we've got backorders full of contracts that are getting. Sorry, revenue that won't get recognized for five years or so. So that's why there's a slight discrepancy. Of course, we've done other existing customer sales than those two big ones, and obviously, that's a different number.
Just to give you some color around that, the interesting thing about those two quite substantial existing customer sales is that they weren't particularly large customers, and they certainly weren't to our largest customers. One of those customers had been a customer for more than 20 years. That's, you know, really interesting and gives us a lot of confidence that we can keep this growth rate growing and keep momentum moving when this customer base has this potential to deliver these kind of deals. What was in those sales? They were a mixture of things. Some of it was license capacity expansion, where our customers are growing their bases.
As customers take on more customers themselves, so more mobile users or TV customers or broadband customers, whatever, they pay higher fees. Our subscription fees are directly linked to number of end customers. So some of that was just growth, but also customers establishing new brands, new bases, and some of that was one customer, one of those customers that bought additional modules from us. They hadn't bought all of the modules up front and went back now and bought the rest, most of the rest of them. That not only generates digital license fees, but it also generates additional services revenue, services sales to put those modules into production. In addition to that, we had some Evergreen, sale of Evergreen.
Evergreen is our subscription program where customers pay a quarterly fee, and that gives them automatic access to upgrades on a regular basis, rather than those becoming a once every two or three year exercise that then becomes quite disruptive. We're finding a lot of traction in that with existing base, not to mention, of course, new customers who tend to take that automatically. In addition to that, there was some extension of the term agreements that also generated more order value.
A sort of corollary point to this is that what's interesting as well is that we're seeing the opportunity to convert older customers into more of the SaaS model through selling them not just Evergreen, but managed services, for example, so that we take over running the software from all the customers' IT departments. I think, you know, we'll see more of that, more of that going forward. Okay. I think Sorry, one final point before we, before I hand over to Andrew. In terms of new customer sales, I'm hoping these slides are gonna advance fast enough for... No, it's gonna take too long. I'll just speak about that for a minute.
It's not on this slide. We do have a slide on the sales pipelines further along. Our new customer pipeline, as in new logo customer pipeline, is up by 23% to a record GBP 212 million. I think that we will convert some of that. Oh, here we go. Yeah. Thank you. As you can see from this chart, this is our unweighted new customer pipeline. We're pretty confident we'll convert some of this in H2. We'll also have new customer sales coming through as well. Again, that should help this order value to be, you know, at quite a high level, I think, for 2023 as a full year. That is not a forward-looking statement. Could we go back to the KPI slide, please, Tamsin? One more. Great. Okay. We're very positive about the sales outlook. I'll hand over to Andrew now to go into some more detail, give some more color on the numbers. Andrew.
Thank you very much, Louis. As you can see from the graph in the top left-hand corner, revenue growth was very strong in the first half, increasing by 27%. That follows a 26% increase in the first half of last year, really continuing the very strong trend that we've been seeing in recent years. Moving over to the right-hand side, it's interesting to see that adjusted PBT grew even faster than that, this grew by 46% in the first half. Clearly we're benefiting from favorable operating leverage, as we gain those incremental revenue as the incremental revenue comes through. Part of the reason for that very strong operating leverage in the first half of this year, was a lot of the extra revenue came from license sales.
As most of you will know, any incremental license sales will drop all the way through to profit at 100%, 'cause there's very little additional cost associated with them. Moving to the right-hand side, you can see that net cash performance remains very strong. Net cash was GBP 23.6 million at the end of the half. There's a detailed slide we'll come onto in a minute, which looks at cash performance in a bit more detail. In terms of new orders, Louis already talked through the key points on the new orders. There's nothing else I wanted to add to that. If you look to the graph in the very middle in terms of the backorder, you can see the backorder remains very strong at GBP 43 million.
This is made up of two components. First of all, GBP 34.7 million of orders that have been contracted but not yet recognized. Essentially this is, you know, orders that we've got in the hopper that ready to be turned into revenue, and gives us very good visibility into revenue over the next few months. On top of that, we've got GBP 8.3 million of annualized support and maintenance revenue. Again, that GBP 43 million, you know, is very well covered, gives us very good coverage for revenue for the rest of the year, and also going into FY 2024. Graph to the right of that shows the recurring revenue run rate. This is increased by 34% up to GBP 13.1 million.
Again, the increase in recurring revenue run rate has been faster than the increase in revenue of 27%. This is showing that over time, Cerillion is becoming an increasingly higher quality business. Part of the reason for the increase in the recurring revenue run rate was the increase in managed service run rate, which you can see on the graph on the bottom left-hand corner. This increased by 80% over the period from GBP 2.5 million up to GBP 4.5 million. We, you know, we expect this growth will continue at these sort of rates into future periods. In terms of adjusted EPS, this really mirrors the adjusted PBT graph as you would expect, and just demonstrates again the very strong operating leverage that we have got in the business.
The final graph in this slide shows the dividend per share increased up by 27% to 3.3 pence per share, which , demonstrates our continued progressive dividend policy. In terms of the financial highlights, as I said before, revenue increased by 27% in the first half. The table here shows the breakdown of that revenue between software, services, and other revenue. As you can see, the increase in the period was fully driven by an increase in software revenue, and that was mainly due to an increase in license revenue being recognized in line with IFRS 15.
Another way of looking at that, the recurring revenue in the period was GBP 6.5 million, so that makes up 32% of the total revenue balance, which again was two percentage points higher than in the first half of last year. Again, reflecting back on the previous slide, over time, you know, Cerillion is becoming a higher quality business. In terms of margins, gross margin increased by three percentage points to 81.5%, and the Adjusted EBITDA margin increased by four percentage points to 48.9%. Again, the majority of this increase was driven by the higher proportion of license revenue that was recognized during the period. At the same time, there has been some increase in operating expenses, which we'll see on the next slide. This reflects mainly increased investment in headcount.
We're continuing to grow our headcount in order to ensure that we can continue to grow the business at these high levels into the future. As you'd expect, there also has been some element from higher inflation falling through, as well as other elements, including higher sales commission. Finally, you can see our very strong cash performance there. Again, net cash increasing by 43% up to GBP 23.6 million. The next slide looks at the cash performance in a bit more detail. The table at the top shows a reconciliation of Adjusted EBITDA down to free cash flow. What you can see there is that there has been some element of an increase in working capital in the first half.
This is linked to the higher proportion of license revenue that we recognized, and this is something that we did flag as part of our FY 2022 results. As we are recognizing more license revenue, in line with IFRS 15, we have to recognize that revenue up front. On a typical five-year contract, the customer will pay in equal installments. As we recognize the revenue up front, we have to recognize more accrued income on the balance sheet, but that will then unwind over the contract term. In terms of CapEx and net interest and tax paid, these balances were broadly in line with the prior period. This is the key story for getting to free cash flow of GBP 5.8 million. 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've generated is a lot higher than the amount that we spent on dividends, on lease payments, and also from unfavorable FX. Consistent with prior periods, we have generated net cash. In terms of the consolidated income statement here, I've already talked through the increase in revenue and also the drivers for the increase in margins. Really just four additional points I wanted to mention here. First of all, is that we continue to invest in research and development. Over the full year, we expect to invest around 12,000 days into R&D. From an accounting perspective, in the first half, we capitalized half a million GBP worth of development costs. This was broadly the same as the amount of the amortization charge that went through the P&L.
Net-net, there's been no overall benefit from capitalizing these development costs. Secondly, as you can see, there was an increase in operating expenses of 18%. This was much lower, only 13% after stripping out the impact of foreign exchange. The fact that this was well below the increase in revenue of 27% shows that we continue to control our cost base very closely. Thirdly, in terms of the depreciation and amortization balance, you can see there the GBP 1.6 million includes half a million pounds relating to amortization of acquired intangibles. This balance stems from the IPO and is now fully amortized.
Therefore, going into the second half of the year and in future periods, this balance will be closer to GBP 1.1 million, so around about GBP 500,000 lower than we've seen in the first half. Finally, on this slide, as we had anticipated, there was an increase in the effective tax rate. This increased from 14.2% up to 19.4%. The 19.4% reflects our best estimate for the full year tax rate, and the main reason for the increase is the increase in the U.K. corporation tax rate from 19% up to 25%. In terms of the consolidated balance sheet, I think the key point here is that the balance sheet remains incredibly strong. You can see net cash of GBP 23.6 million.
There's no debt on the balance sheet as this was repaid in full two years ago. Overall, there was an increase in net assets of 38% from the prior period. In terms of the consolidated cash flow statement, this is mainly shown here for reference. You can see a reconciliation of Adjusted EBITDA at the top of GBP 10 million through to our closing cash balance of GBP 23.6 million. I think the key point here is that cash generation remains incredibly strong. You can see that cash has increased by 43% from the prior period, but I think most of the key points have been covered on the previous cash slide, so I wasn't planning on saying anything in addition to that here. Back to you, Louis.
Thanks, Andrew. That's great. Thank you. I won't say much more to just sort of summarize that your strong order book and strong pipeline, particularly the new customer business, gives us a lot of comfort about the future. I think, you know, in terms of outlook, we're pretty well positioned. That's really it from us and I guess back to you, Tamsin. Floor's open for questions.
Tremendous. Thank you very much. If you have a question, click on the Q&A button and type it in. We've only got Louis till 2:00 P.M. Andrew might be able to stay around for a bit longer. We'll kick straight off. What do you intend to do with the huge cash pile? Are share buybacks a possibility?
I think we buy limited amounts of shares to fund the staff Save As You Earn share option scheme and the LTIP for the senior management team, but they're relatively small amounts. We're more likely to We're more likely to use that cash for acquisitions. In acquisitions, we're looking for bolt-ons or tuck-ins, adding product that fits around the edges of our suite, where we can bring in product that we can upsell to existing customers, or equally bring in new telco customers we can cross-sell our existing product set too. That's more where the focus is. Also, it's worth saying that as we win larger customers with larger deals, there's more focus on our balance sheet and, you know, we do need to be able to demonstrate that we have a strong balance sheet. It shouldn't be all of that cash is available to be disposed of, so to speak.
Thank you very much. You talked about three classifications of competitors. It sounds as if the competitors most similar to Cerillion are the smaller software vendors. How good is their service versus your own? How likely is it that one or more of these will do Cerillion and become real competitors by taking a high quality offering to a wider audience?
That's a good question. I think we do punch above our weight. We don't compete much with the smaller vendors. We compete really with the bigger ones. It's a hard one. I think part of it is that, you know, if you go back to our DNA, we were founded out of what was a large software business. From day one, we had a lot of process and procedure and a sort of quality ethos that in a lot of smaller startups, you just wouldn't find, because of our MBO background.
I think also it takes a long time to get established in this market. It's taken us a long time to, you know, to have this breadth of references and trust that we have. You just have to have done a lot of successful projects to achieve that. It's not impossible, and other smaller vendors might spring up, but there's a big entry barrier. They've got a hurdle to get to that point.
Thank you. Are you currently working on a new module to add to your package, so to increase sales to existing customers?
Well, we're constantly refreshing the modules, and we're not working on a completely new module at the moment. You know, we've recently updated and upgraded our campaign management, which we call CPQ, Configure, Price, Quote, which is a key part of the business sales process. In such that customers can now use Cerillion to replace Salesforce, for example. We have customers who would have bought Salesforce, but they looked at our campaign management CPQ and said, "Well, what's the point of having Salesforce? We'll just use this for campaign management."
That's an interesting development. Done a lot recently on loyalty schemes, improving that functionality in our product. Of course, at the moment, we're looking at AI and the new features that can provide to give us differentiation in the market. Not just on the productivity side, which is obviously a key part of AI, but, you know, what new features can we bring in that make use of the AI capability that's now available.
How will AI be part of your business going forward?
I think AI will be a key part of production. What we're seeing now is tools that can write software modules and then test software modules. Of course, AI makes mistakes, so it's important that AI can not just write software but also test it. Now, as you are users, you've got to think about software systems as almost like Lego models. Let's say we're building the Eiffel Tower out of Lego. You know, AI will not be able to necessarily put all the bricks in the right places to build the Eiffel Tower if you just said, you know, "Build the Eiffel Tower." Of course, probably not a great example because AI could look at a picture of the Eiffel Tower and 3D diagrams and probably do that.
You know, software can be a more complex thing to design. What AI can do is it can make the bricks. If you think about, you know, maybe that model is 1,000 bricks, and each of those bricks being a piece of code, it takes maybe five days to write, and five days to test. The AI can do that work, that's a lot of the kind of, you know, the coding that. Well, I'm trying to think of a polite way of putting it, the detailed, you know, production work. Not manual work, but, you know, it's sort of lower- skilled work than the design part. Y ou know, that not having to spend all that effort on just the basic code cutting and testing will save us a lot of effort and increase productivity significantly. It will have a big impact.
Tremendous. Thank you very much. The current business focus is in the telco area, and as previously stated on the last presentation, you see plenty of opportunity for growth in this industry sector, which is a positive position to be in. That said, do you see a benefit of pursuing additional market sectors with similar needs where your existing solution could be tweaked to fit, so this would be a potential future growth strategy, or would it need internal resource, and would that be a challenge to pursue?
It's a good question. I think when you look at the share of the market we currently have in telco, you know, I would be surprised if it was much more than 1% of the market. There's such a vast space for us to go into that right now our focus is on that and achieving greater momentum in this market. I think you could certainly adapt our software to other utilities, electricity, water, gas and so on.
In fact, some of our modules are used in other industries. Really for the big transformation projects, you need a lot of domain expertise. We're experts in telco. We can go and do a digital transformation project in the telecoms business and, you know, turn everything over. We don't have that domain knowledge or experience in electricity, water, power, gas, and so on. Whilst the software could probably do most of the functions, we don't have the people who have the knowledge to make that happen.
Great.
That is more something we would look at through acquisition, I think.
Great. I'm aware of time. Perhaps one more question before you go, Louis, before Andrew takes over. Do you encounter customer reluctance arising from fear of conflict in acting for potential competitors to existing customers?
I'm not sure I got that. Sorry. Do we come across?
Do you encounter customer reluctance arising from a fear of conflict acting for potential competitors to existing customers?
Yeah. Yeah. Sorry. I get it now. Not really because we're talking about software that's hugely configurable and, you know, even though every customer of ours has the same product, they will have vastly different configurations of that product. you know, everything from the products that are defined, the workflow processes that defines the sales processes, the front end the customers see, that's all configurable. Also, you know, what the CRM screens look like. It can all be configured differently. It generally is to suit individual customers' needs, but all within the same code base.
Thank you. You probably need to go now. Shall I carry on with Andrew?
I can do another five minutes. If there's a question.
Tremendous. What's the size of the addressable market?
We don't have a reliable figure on that, and it depends a lot on, you know, do you include China or not, and Russia's, the Russia or not, and so on. There are different. If you go out and look at Gartner and some of the other analysts, they'll give you vastly different range of figures. We just don't quote a figure. It's a huge market. It's a huge market. I think, you know, you'll get different ranges from, you know, GBP 200 million a year to GBP 1 billion a year, whatever. It just depends what you look at.
Thank you. What's the gross margin you'd expect going forward? i.e. half one appears to have benefited potentially from higher than normal license sales.
One for Andrew.
Yeah. That's absolutely right. The gross margin in the first half of 81% has definitely benefited from the high proportion of license revenue. I think going forward into the second half of the year, we would expect the proportion of license revenue to be slightly lower, and therefore, we might expect the margin to be slightly lower. You know, we're not expecting it to be significantly lower than the sort of levels that we have reported for the current period. I think, you know, looking over to the medium to long term, as we continue to win larger contracts with larger customers, we would expect the number of subscribers in our deals to continue to increase. Therefore, over the long term, we would expect the proportion of overall software to increase, which should drive our higher margins as well.
Thank you. Is there a concern that service revenues fell in half one?
No, I think that there was an awful lot of services work going on last year on new customer implementations, so we're doing five at once at one point. It's not really a concern. There's always timing around those things, you know, and sometimes we're busier than others with new customer implementations where most of the services work happens. I think it's really just a timing thing.
Tremendous. Thank you. What proportion of software revenues come from upfront license recognition versus ratable recognition? Thank you.
To give you a rough idea, and this is all very round numbers. Typically, if you take a, you know, GBP 10 million up a new customer deal, roughly 25% of that will be upfront license. This is a five-year term SaaS deal, and half of the rest of it will probably be something like half the rest of it will be project implementation services that are recognized over the term of the project, so typically 12 months. The other half of that half will be the ongoing SaaS services over the five-year term from once it goes live. That's a very rough rule of thumb way of thinking about it.
Tremendous. Thank you. To what extent do you consult with current customers as part of the process of developing the product? What are the pros and cons of this approach?
We consult with customers a lot. We have twice yearly, what we call webinars where what we call them?
Customer forum.
Customer forum. We have another word for it, though. Anyway, we have a essentially customer user group twice a year. We go through in detail the roadmap and what's on the roadmap for the foreseeable future. We showcase new developments, but also we ask for feedback on what customers think we should be doing and what concern, you know, what particular needs they have that they feel we might not be addressing yet. That's a key part of our model. I don't think there are really any cons in consulting customers.
Tremendous. Thank you. You mentioned that part of your revenues are from North America, but you're building the team in the U.S. Can you expand on your plans for the U.S. and the benefit you anticipate from expanding the U.S. team?
We're not building a delivery team, I mean, in the U.S., but we're putting some sales resource on the ground. I mean, bear in mind, we do sell into North America, but we've done it based out of London most recently. I think it's just about getting a little bit closer to that market by having boots on the ground. I think the challenges with the U.S. market, particularly as opposed to North America in general, is that the telcos tend to be very large or very small. Too small is too small for us, and the very large are quite hard to break into. You know, there is more diversity, I think, in that market than there has been in the past. We do believe it's worth investing more into trying to build a bigger customer base in North America.
Thank you. Analysts are forecasting earnings per share growth for the next two years to be significantly below the last two years' growth. Are you experiencing a slowdown in growth or just being cautious...
Mm.
in your forecasts?
I think there's a. As people know us will know we are very conservative, and we have, you know, the analysts reflecting our conservative approach, I think. I think all the analysts have said there's upside risk in the forecast. You know, in broad brush, We're not expecting the current growth rate to slow. So to the extent that's not reflected in the forecast, that's us being conservative and being a little reluctant to push the pedal to the floor until we see, you know, the next new customer deal come in. I think it's, again, it's a timing thing really.
Thank you very much. Last question. How is inflation affecting costs now, and how do you see this going forward?
Andrew?
Yes. If we look at our cost base, I mean, the vast majority of our cost base is made up of payroll costs. You're absolutely right, there has been an element of payroll inflation coming through in the current year. And if we look at the spectrum of headcounts across the world, we've got people, you know, the majority of our people are in the U.K., Bulgaria, and in India. Earlier this year and last year as well, you know, the Indian market in particular was particularly hot. There was, you know, a fair amount of inflation coming through in India, but we are seeing that market cooling now. Therefore going forward, we would expect to increase pay costs by a lower proportion than we have done in the current year.
Thank you very much indeed. That's the end of questions. Louis, do you have any closing remarks?
I'd just like to thank you all for joining, and perhaps just reflect for a moment on general market. I mean, I know there's been some negative press from some players in tech and software and in telecoms. Also, you know, there are companies that are, I think where companies have solutions that are mission-critical, then there is strength in this market. Not just us. If you look at, for example, at IQGeo, you know, they're absolutely booming in this telco space. I just urge people not to attach too much significance to one or two announcements.
Tremendous. Many thanks. Thanks for giving us a bit of extra time. Thank you also, Andrew. To everyone listening, you'll now be taken to a webpage to give feedback on the presentation. If you can't complete it now, you'll get a follow-up email. We'd be really grateful if you could take a few minutes to complete. Many thanks for joining. This is the end of the webinar.