I'm joined today by my colleague, Ninnie Karlsson, the CFO of Aira. I will start by doing a company presentation and talk about our latest growth initiative, Aira for Fortnox. Ninnie will cover the financial highlights from the third quarter, and we will end with a short Q&A session. If you wish to ask a question, you can do so throughout the presentation by using the Q&A button in your Zoom toolbar. Feel free to ask questions all throughout the presentation, and all questions will be answered at the end of the call. All right. A short introduction to Aira. Aira is an AI company. We're a young company. We were a spinoff from my previous business, Upsales, and we listed and became an independent company in June, a few months ago.
Our business is about helping our clients, particularly small businesses, find new opportunities, find new customers by using data in smart ways, and also understanding the customers they already have. We're building this on two decades of experience. We are an AI-native company, delivering the service both as a mobile app, but also as a traditional web app. Our latest initiative, which was communicated a few weeks ago, is the Aira for Fortnox product, which I will talk a little bit more about, why we built it and how we see the opportunity. The first Aira product was commercially launched in April, and we have since launched more products. We realized after the summer that we had more opportunities than we could pursue. In the third quarter, we chose to focus all of our efforts on one of them, which is Aira for Fortnox.
Why is this? Aira for Fortnox is a sales and sales intelligence product that it runs itself, and it lives inside the existing accounting system the customer is already using. It's the same engine, it's the same technology that we've been building for more than two years now. We build it inside Fortnox, and we use all the data that the customer already has in their existing accounting setup. Why this is so important, we believe that the ERP or the accounting system is the key. The data that resides in there is the key to get all the valuable use cases you want from a tool like Aira. Because the data you have in your accounting system is typically high quality. No one is sloppy in their accounting system. So you have the right data about your customers.
You have all the historical data about your sales. By deploying Aira as a product that lives within Fortnox, we are able to set everything up for the client in a matter of minutes. It requires literally zero effort from the user. The only thing the user has to do is to click a button, and Aira gets access to all the data inside Fortnox. We can set everything up, and we can start delivering value to the user within a matter of minutes. Our experience tells us that if you can remove as many of the thresholds as possible to start getting value, to start trying out, to start using a piece of software, that's the key to getting scalable go-to market. This is a huge market. Fortnox has 600,000 clients.
We estimate that around 20% of them have a need or have the kind of business which requires a tool like Aira. It's a huge market. I'm going to talk a little bit about why no one has reached them yet. The absolute majority of these businesses have not bought anything. They're still stuck in manual routines with spreadsheets and so on. From my experience of running my previous business, I know that the very smallest companies with less than 10 employees, they simply do not have the time to go through the hassle of googling products, setting up trial accounts, trying, evaluating different kind of softwares, and learning new software. They are too busy.
Also some experience from when I've been doing this for the last two decades, it's like no matter how much you target larger clients, no matter how much you change your targeting and your marketing and everything, 80% of the leads you get are still small clients with less than 10 employees. If you don't have a super scalable product, you have to disqualify 80% of them on price or complexity. These are some of the learnings from the last 20 years of running similar businesses. The market is huge. We estimate that this is an addressable market of SEK 2.2 billion per year based on the 20% of the 600,000 Fortnox clients.
We believe we can reach a significant share of this market. The model continued to scale as well, because if you go to Denmark, there's a similar company like Fortnox in Denmark called e-conomic. If you go to Norway, you have another one called Tripletex. Both of them are equally dominant in their markets as Fortnox is in Sweden. There's also a company called Xero in the U.K. My team, we have been working with both Fortnox, e-conomic, and Tripletex, and Xero in the past. We've built integrations to these systems. We have more than 10 years experience working with the Fortnox ecosystem. We really see a huge opportunity, and the perfect team to execute on this opportunity.
How are we going to reach? If we want to get this product out there to 3%-5% of Fortnox's total customer base, what's the path of getting there? What we will do, we will, again, build on what has worked in the past based on all our experience. It's a combination of outbound sales and doing a lot of events. It's a little bit different or very different, I would say, from how most SaaS companies do this. Because most SaaS companies, they argue that if you're going after smaller clients, you cannot reach efficiency with direct sales. Our experience says the exact opposite. If you do it right, it's the most effective model in reaching a large share of this market.
Selling the product through the Fortnox Marketplace is key to this launch, because it doesn't require any setup, any API keys, any new passwords, any contract. Nothing, just a click of a button from the client. There are more than 10,000 companies in the Fortnox partner ecosystem, typically accounting firms. We will leverage the partner network to help us reach all of these clients. What I like about this model is that it scales very well. You can set up a sales team, where a sales rep can close two of these clients per day. You ramp a new sales rep in a matter of weeks, because the cycles are so short, and the product is so straightforward. We estimate that the average ARR per customer is around SEK 18,000 . So a client will pay SEK 699 per month per license.
We estimate that the average client is just above two seats or two licenses. This is based on more than 10 years of working with the Fortnox ecosystem. These are estimates, but we think we are pretty spot on. Again, by deploying this model, we reach a very high effectiveness in terms of cost of sales vs new ARR. We recover our sales and marketing investments in less than four months. This model, it scales like nothing I've ever seen in my 20+ years of being in this business. That was the short recap, a short summary of what we're seeing. I'm going to hand over to Ninnie now to talk a little bit about the financials from the third quarter.
Thank you, Daniel. In the third quarter, our development and commercial focus was on Aira for Fortnox while Aira and Aira Connect kept shipping. ARR at the end of the quarter was SEK 1.85 million , up SEK 0.08 million in the quarter. Net sales for the quarter were SEK 0.5 million . EBITDA came in at SEK - 3.1 million , which is an improvement from SEK - 4.6 million in the second quarter. Our cash position at the end of the quarter was SEK 9.9 million . Looking ahead, I want to highlight two things. Aira for Fortnox launches commercially on the 1st of November, and we expect meaningful revenue from Aira for Fortnox already in the fourth quarter. That will extend our runway in the near term.
All right. Thank you, Ninnie. We will head into the Q&A. Again, if you wish to ask a question, just use the Q&A feature in your Zoom toolbar. All right. We have a question from Johan about our previous launch. "Earlier this year, you said that the market for Aira would be different from Upsales, more global, with a rollout taking place through partners and therefore scaling well. The Fortnox integration sounds like a fantastic opportunity, but also like a more local strategy reminiscent of Upsales' early years. How has your view of Aira's market changed over the course of the year?" Okay. The first product we launched was a standalone sales tool with access to all the data. The data is what sets Aira apart. We realized quickly that that product was difficult to scale.
What we saw in the early feedback from we still have customers paying for that product, but what we saw early was that the most interest came from clients already using an existing sales tool like HubSpot or Salesforce. That is why we pivoted in, I think it was in June, and we built Aira Connect, which is the product we have been selling since. Aira Connect is a data product that connects to existing CRM and sales tools to make them better. We have traction there, so I think there is an opportunity to build a great business on that product. But when we started looking at if we connect to the ERP instead of an existing CRM, we can deliver 10x, 20x, 30 x more value to the user with zero effort. That was the main reasoning why we are so enthusiastic about this opportunity.
My team has, again, more than 10 years experience of building these ERP integrations. Why we did not launch ERP integrations in the first product was because, again, the thresholds are slightly too high. If we want to sell a scalable product to 10,000, 20,000, 30,000 companies, and it requires every one of them to go into their accounting system to pull out an API key to set up an integration, they will not do it. Less than 10% will do it. So, we decided not to go that path. But when we realized that we can build something with the access to ERP data on top of the Fortnox Marketplace, so the customer does not have to do anything. They get all the value from using that data, but they do not have to sign up for anything.
They do not have to set up integrations, find API keys, set a password. We just realized that this is the perfect product to sell. That is why we are so enthusiastic about this. Again, the model scales. Sure, the first product is local, but a local market with an addressable market of SEK 2.2 billion is an amazing opportunity, and it scales to more markets with e-conomic in Denmark and Tripletex in Norway. Basically, any market that has a dominant player selling accounting software to smaller teams, also with decent integration abilities, like the Fortnox Marketplace. So, the total runway for this initiative is almost infinite. That is my view. Another question, what has the main channel been to find the SEK 1.8 million ARR?
What we are doing now with the existing product, I would say we are doing marketing and advance and different kind of digital initiatives. But selling a product like the Aira Connect, which is SEK 100,000, SEK 200,000 in ARR per client, that is a sales motion. So you generate leads, but you need to build a sales team, a traditional sales team that finds opportunities and builds pipeline and closes deals, which scales. I have a 25-year track record of doing exactly that. But the model of selling to this Fortnox market, which it is also a sales motion, but much more simple kind of sales motion that scales a lot better and a lot easier. Right. Let us see if we have. We do not have any more questions right now. I will give it a minute or two. Okay, there is one more.
Is the value delivered with the new product the same, and is the main difference how customers get started? No. The basic idea of what we want to deliver, we want to help our clients understand their existing customer base. We want to use that data to generate new leads, and to help them just work more effectively. That's the core idea of all of our products. The difference being is that when we integrate to an existing CRM like HubSpot or Salesforce, the data in any CRM is typically very low quality because it's entered manually by sales reps. We deliver value to those clients, but there's a limit to how much value we can deliver, and it also requires some effort from the user.
I think that's the big difference with building on top of an accounting system is that you know you can build a product on the assumption that the data will have 100% quality. More customers will get more value, guaranteed with less effort. That's what I like about this model. Again, feel free to type in your questions in the Q&A feature. Okay, we seem to have no further questions. That concludes the call for today. Thank you, everyone, for joining, and have a good day.