Here’s why Citi says crypto prices have been weak recently
Columbus A/S reported a challenging third quarter in 2025, with a 7% decline in revenue and an 18% drop in adjusted EBITDA, leading to a 3.05% decrease in its stock price. The company maintained its full-year guidance despite these setbacks, indicating confidence in a potential market recovery.
Key Takeaways
- Columbus A/S’s revenue declined by 7% in Q3 2025.
- Stock price fell by 3.05% following the earnings report.
- M3 business line showed 8% revenue growth and improved margins.
- The company maintained its full-year guidance despite financial challenges.
- Market conditions remain cautious, with European markets facing difficulties.
Company Performance
Columbus A/S faced a tough quarter, with significant declines in revenue and EBITDA. The company attributed these results to challenging market conditions, particularly in Europe, where Sweden and Denmark experienced slow recoveries. Despite these challenges, the M3 business line performed well, with an 8% increase in revenue and improved margins.
Financial Highlights
- Revenue: Declined by 7% year-over-year.
- EBITDA: Declined by 18% (adjusted).
- Contribution Margin: 23%.
Market Reaction
Following the release of the earnings report, Columbus A/S’s stock price fell by 3.05%. The stock is trading near its 52-week low, reflecting investor concerns over the company’s financial performance and market conditions.
Outlook & Guidance
Columbus A/S maintained its full-year guidance, expecting revenue of approximately DKK 1.7 billion with an EBITDA margin of 7-9%. The company anticipates a potential pipeline conversion in late Q4 2025 and Q1 2026, which could bolster performance.
Executive Commentary
CEO Søren Krogh Knudsen emphasized the company’s focus on practical AI implementations and strategic growth markets. "We are seeing a big lineup of decisions for November, and it will be important for us to win our share of that," he stated, highlighting the importance of upcoming market opportunities.
Risks and Challenges
- Continued market caution and postponed investment decisions by customers.
- Challenging conditions in European markets, particularly in Sweden and Denmark.
- Slight decline in the Data and AI business line.
- Need to maintain efficiency and manage workforce adjustments effectively.
Q&A
During the earnings call, analysts inquired about the expected timing of market recovery and the company’s strategic initiatives. Columbus confirmed a focus on converting its pipeline in late Q4 and Q1 2026 and emphasized its strategic hiring in AI and growth markets.
Full transcript - Columbus A/S (COLUM) Q3 2025:
Conference Operator: Good day and welcome to the Columbus Interim Report Q3 conference call. At this time, all participants are in listening-only mode. After the speaker’s presentation, there will be a question-and-answer session, and to ask questions through the phone during the session, you will need to press Star 11 on your telephone keypad, and you should hear an automated message advising you that your hand is raised. To withdraw your question, please press Star 11 again. Alternatively, you can submit a question via the webcast. Please be advised that today’s conference is being recorded, and I would now like to hand you over to your speaker of today, Søren Krogh Knudsen. Please go ahead.
Søren Krogh Knudsen, CEO/Presenter, Columbus: Thank you very much, and good afternoon to everybody. Welcome to this webcast where we will be presenting our financial results for Q3 of 2025 and the year-to-date 2025 results. As always, I will be going through briefly some financial and operational highlights first. I will then hand over to our Group CFO, Brian, who will take you through the financials in some more details and finishing off with the outlook for the full year of 2025. We will have a short Q&A session at the end of the call. Q3, we saw a declining revenue of 7%. It is driven by market conditions in Denmark and Norway that remain pretty challenging for the time being. We are seeing some uptick in our pipeline, but for now, we still have a lower-than-usual activity in the market.
In Sweden, that has been subdued, actually, market-wise for a couple of years. We’re starting to see the recovery kicking in. In the U.S., we’re actually enjoying a continued positive trend, and we are structuring around that to maximize the growth we have there. In Q3, our EBITDA declined by 18% when we adjusted for other operational income and expenses. This is primarily due to a capacity adjustment, which we’ll comment on later in the presentation. Overall, this reflects the impact of a weaker revenue, which we see mainly deriving from a slightly lower operational efficiency than what we’re usually operating at. This, again, is what the capacity adjustment will address as well. Our contribution margin ended at 23%. That’s broadly in the same level as in Q2 of 2023.
Sorry, in Q2 of 2024, where we had 23% when adjusted for these extraordinary redundancy provisions that we had now, which are at DKK 11 million. I think we can say this demonstrates that we have a pretty strong and stable project execution. It leads to very profitable projects. We have very little in the way of guarantee and rework, and we have high customer satisfaction remaining. It is mainly about having enough activity that is the cause of the lack of top-line revenue. On the positive side at the end here, our second-largest business line, M3, continues to deliver solid performance and increasing performance. We have been working a lot with this over the past years. In this quarter, we are seeing revenue growth of 8% in Q3 of 2025.
We have a contribution margin that’s up from 13% in the same quarter last year and is now sitting at 20%, which is starting to get in the range of where we would see a well-performing business unit. This journey and improvement journey of M3 is by far not over, but we’re really seeing some results now, which we’re very pleased with. We move on to market and operational highlights. I think we need to start by saying that we are still seeing this continued caution, which we call it in the market. It’s basically a lack of demand. We’re seeing that customers tend to think twice about everything. They tend to postpone investment decisions. They tend to chop investments up in smaller bites than normally. However, that being said, that slowly also leads to a catch-up effect for many of the customers because the.
Long-term ambition for them remains the same. That leads me to the second point, we are starting to see the benefit of staying very, very close to our customers. Despite this lack of demand, we have not strayed from our customer focus. We have not strayed from our industry focus. In 2025, we have continued to develop the investment plans with these customers. There are signs that this is starting to come through to fruition. In Q4, which we are in now, we have a pipeline which has a fairly high number of large contracts that are up for final board approvals at customer sites, and it will be an important thing for us to achieve a good win rate here. Yes, also, a comment on the right side of this slide, which tells you something about the development of our efficiency.
In Q1 and Q2, you can see we’re operating at 62-63%. Which is not, it’s pretty respectable, but it can be a little bit higher. It can be 65-66-67 even. You’re then seeing this drop in Q3, which leads to the revenue shortfall that I’ve just alluded to. We’ve reported to you here in the light blue. This is important to say this is part of Q4, not Q3, but this is October in isolation. You can see that it’s risen from 58-61%. That is before we see any uplift from this capacity adjustment exercise because those people that we have unfortunately have had to exit the company are still on our salary books. We still report them as part of the efficiency figure. That number in total leads to a 2.5% increase or improvement in efficiency as well. Yes.
We see, again, in summary, our win rate when we have something to bid for is intact. We do win when decisions are being made. We’re staying very close to these customers. We’re seeing a big lineup of decisions for November, and it will be important for us to win our share of that. In terms of winning, for those that are not so familiar with our business, once we have won a large contract, there is a ramp-up phase which consists of usually we have some sort of explore or prepare phase, which is done by a slightly smaller team. Then the bulk of the team, the construct part of the team, will enter shortly after that. An important thing is also the timing here, how much of this can benefit Q4 and how much of it will flow into Q1 of next year. Yes.
Let’s go to the next slide, please. Okay. Again, we have adjusted the capacity, as I said, overall 89 headcounts affected, but it is not to say that we have only downsized. It is also a shift of competence. We are actually hiring quite actively for different types of competence profiles that we need going forward. Very much of that is within the AI space. Also, as you can see, some of our geographies are growing. Of course, we continue to hire there, and in other pockets, we’ve had to do adjustments. It is not like we have a hiring freeze, but we are adjusting where it makes sense, and we are building out and investing where that makes sense. The last point here on slide 7 talks about the engagement level of our workforce. We are paying a lot of attention to this.
Obviously, 2025 is a tough year. We started with the strategic review. We are seeing a demand shortfall in the market. We have had to do our capacity adjustment exercise, and we have really committed to being transparent to our workforce, telling them what our strategic considerations are and keeping them in the loop. I believe we are seeing a very big benefit here when it comes to a very recently conducted employee survey that shows us that we still have a highly engaged workforce. We are well above the benchmark that we always compare ourselves against. It is important to us that we keep investing in having a workforce that is motivated, that is engaged, and has full belief in the future despite going through a slightly tough period. Going to slide 8.
I want to talk to you in the next three slides about the adoption progress of AI and what is actually happening. We’re starting with a little bit of a classic hype cycle here, and I think I’ve spoken to most of you about this one before. Dunning-Kruger cycle. You know that at that very peak, H4, you have the height of expectations, which is typically then followed by the trough of disillusionment because things take longer time than experienced. I think there is some learning here. If we start by going to the far right, you can see that the AI personal boost, which I’m sure you’re also experiencing, actually raced through the entire curve fairly quickly. We are now all benefiting from that work-wise and also in our private lives, and it has tremendous benefits. This is pretty old news.
Now, the H2, the AI professional news. Boost, is somewhat different. The main difference to the personal boost is that we start to give our Copilots or whatever we are using access to corporate resources. That gives a whole other level of professional boost because we now can access things that are just not on the World Wide Web, but all the company’s internal repositories. This is done by many, not by all. I would not say we are fully there yet, but we are certainly in our company, but also many of our customers, we are reaping those benefits right now. It comes to the slightly more complicated things. I do not think we are very far down that journey, right? It is the H3 and H4. H3 is a true functional boost with semi-autonomous processes. Automated workflows. And.
Leading ultimately to H4, which is this true agentic, fully autonomous behavior where an entire workflow, an entire role description is now handled by an agentic entity. We have not seen yet at all the full ROI on large scale for customers. Everybody can come up with a few good examples. The large-scale adoption is yet to be seen. I would say the impatience from customer side is growing. There is less appetite now for long 10, 5-year visions. There is a lot of appetite for return on investment, proof of concept, something simple and truly working that we can scale on. If we go to the next slide, I have got two examples of how that could look. This first one is an example of an organization, global organization with a fairly complex setup of.
Benefits and compensation plan, a lot of variable incentives linked to the company’s financial KPIs, obviously subject to local law, employment law in the countries where the employees are employed. This has led to enormous amounts of man-hours being poured into basically analyzing, providing updates, where are we on these incentives, what happens if we need to change, how many are in each. How much accrual does a finance department need to make. Here, the solution has been basically developing in Copilot Studio a model that can track both the financial data that’s internal, obviously incentive data, that’s also internal, HR data, that’s also internal, and local law initiatives or legislation and handle that. Fairly complex. Maybe not the most exciting example you can think of, but it’s fairly complex, and there’s a lot of man-hours being saved. Learnings were it’s very positive.
The solution really works and can shift the mindset. A different learning was, we’re often talking about that the development of agentic entities will be democratizing. Everybody, every one of us will be able to set up an agent. The experience so far is that’s clearly not the case. You need highly specialized skills to do this still. Possibly, it will be more user-friendly with time, but it’s not as easy as marketing will make it sound. We also see hidden costs being triggered because the compute power required to do all of this, you have to be fairly smart how you set it up. Otherwise, you will rack up a pretty large bill in terms of the compute power you consume.
We also see the need for both governance and validation because if you set this up wrongly, you will get the wrong results and you will make the wrong decisions. As expected, but a very positive and beneficial case. Let’s quickly take the next one. I’ll go a little bit faster here. This one is a very simple one to understand. It’s an order handling flow. A lot of orders received by email. Previously, some manual handling in terms of entering these work orders into the systems and tracking them. You can automate this, simple Power Automate flow, and basically, you get a—you still have a human at the end of the loop to control and validate, but actually, the fault rate is much lower because you have less typing involved with it. It makes it then prone for.
Next wave where you can take your automation further. Simple example, but it generates cost saving at the same time. It’s much less vulnerable for peak workloads because, obviously, this one can work 24/7 over Christmas. If you have a higher order flow, it doesn’t really trigger a backlog or anything like that. It’s capable to keep up. Two simple examples for you, and I hope that sort of informs some of the many, many small cases that are being implemented on a weekly basis. With that, over to you, Brian, for the financials. Yeah. Thank you, Sean. And definitely interesting. I think. Even from an old conservative finance guy, I start to see that there is actually money in, and there is tremendous upside in this AI that we’ve been talking about for years now.
It is coming, and I feel it, and we also see it internally in Columbus as well. Okay. Let’s have a look at our Q3 first. As always, revenue per business line. First, Dynamics. Our biggest business lines count around 60% of our combined revenue. Of course, if they are having a hard time, we see it on the group, as Søren also mentioned. Again, it is Norway and Denmark that is where Dynamics is facing a headwind, and they end up with a -12% decline in revenue Q over Q. On the other side, as Søren mentioned, M3 is getting back on the growth path. That is very good to see. United States and Sweden are some of the countries where they see a strong uptake and new customers as well.
M3, which many of you might not know, they had an extremely strong position within the manufacturing business, basically around the globe. I think there is more to come from that business line, definitely. Digital Commerce, flat development. Data and AI, a slight decline, Q over Q. They have some project changes but are gearing up again in the coming quarters. We’ll be looking to the future. Good. Let’s look at the bottom line or the Contribution Margin for our business lines when we measure them on their profitability. Again, Dynamics is down. It is closely linked to the efficiency. Of course, the bulk part of our right side is done in Dynamics in the quarter. That will cost around 5 percentage points on their bottom line for the quarter. Still, they’re down, and they are working on getting back on track.
M3, Sean mentioned, very strong quarter. From 13% to 20%, 7 percentage points up. Strong projects and also improvement in older projects that we see that we have a good grip of the margin and how we drive the project, both with happy customers and also a happy supplier. Digital Commerce, flat development, Data and AI strongly linked to the revenue, a bit too low efficiency in the quarter. Also some new guys on board that just need to gear up and get some basic education in different areas. We do expect a comeback here the coming quarters. We normally do not talk much about it, but our order local business or EEM is actually driving a very strong, not small business, because they are bringing more money to the table as Data and AI and some of the others.
I think it is worth mentioning that they are gearing up and also growing when we look at the coming slides. Let us have a look per country or market unit, as we call it. Sweden, Sean mentioned it as well, slowly getting back, not over the top yet, but minus 1% Q over Q. It is good to see. It is our biggest country. Denmark continued to face some headwind from some very strong quarters as well the previous years. Still, we do expect more from our, so to speak, home market where we are sitting. We also feel like that will come. U.K. is down with 7%, actually the first time in four years that they have a negative quarter. It is not that we say, "Fair enough, you take a breath," but they do see some big shift in some major contracts.
If I do calculate it organically, because there have been some with the pound, it’s a 4% decline. Not that we say, "Fair enough, we keep up the pace," but we do expect them to be back in the growing path also in the future. Norway is still, and it has been the previous quarters as well, an area that we are facing some headwind. We do see, and Sean mentioned it, we had some strong pipeline out there, and some of it is definitely also in Norway. Let’s see. We do expect that uptake, definitely, in the coming quarters up there as well. U.S. is continuing up, lower value, but still on a strong growth path. Good. That was the quarter. Let’s have a look at the year-to-date numbers. Slightly same story. Dynamics, we talked about, year-to-date down with 7%.
The fast one, we’ll see that actually. In the last quarter, it’s getting a bit faster. We do expect that to even out soon and getting back on track. M3, year-to-date on a zero. They have, on the other side, seen an uptake the past quarters. So getting on the right side of the zero. Year-to-date flat development compared to last year. Digital Commerce, been through a very hard turnaround. Is getting back on track. Data and AI, actually, year-to-date, they are still on. A slight plus in a challenging market. Yes. On the margin side, year-to-date. Dynamics, 2 percentage points down. One if I addressed for the redundancy cost that they faced in Q3. Which is still not the level where we would like them. They are working hard on it.
Although on a keeping on a healthy level, as Sean also mentioned, we do earn money on the project and the work that we do. M3, growing. From 18% to 22%. Which is 4 percentage points, but 3, as we have written here. It’s actually even better. And that is good to see that they also start to get back. It’s, of course, helped by improved efficiency and some very well-delivered projects out there. Digital Commerce. Continuing the upward trends compared to last year, which was also very low. But it’s also good to see that we can turn it and slowly move back on the right. Level. Although it’s still a way to go, they should definitely be above 20%, and that is also in our planning.
Our EEM, or we call it other local business, is mainly what we call the enterprise information management systems, which is systems that make our customers smarter and more efficient and compliant in the way they work. Have seen a strong uptake. There is a bigger part of project revenue in this business line as well. They are actually both growing and improving their bottom line due to very strong delivery. Yes. Final slide before the outlook and questions, if any, is the market unit revenue year-to-date. Sweden, still down year-to-date 5%, as you remember, trending slowly upwards towards the zero or flat development. Denmark and Norway, if I take these two jointly, do face some headwinds, definitely Norway. We also start to see some of the pipelines in these areas of our business and these market units.
U.S. and U.K. is continuing a strong growth path, also in difficult markets, to be fair, in these two countries, but with a solid performance. We also do continue to invest in these countries to basically be able to deliver the growth that we see. Good. Let me end on the outlook slides. Based on the first three quarters, we basically maintain our full-year guidance that we announced in July. We adjusted it slightly downwards. That is DKK 1.7 billion in revenue, or as I think we wrote back then, around the level as last year, so plus minus 0% growth. EBITDA margin in the range of 7-9%. Good. It is back to operator or questions, I believe. Yeah. Thank you. At this time, we’ll conduct a question-and-answer session.
As a reminder, to ask a question via the phone, you’ll need to press star 11 on your telephone keypad and wait for your name to be announced. To withdraw your question, please press star 11 again. If you wish to ask a question through the web, please do so now. We have the first question. One second. Okay. Yep. We are going to go to our first question on the phone conference. The first question comes from the line of Yuezu from SEB. Your line is open. Please go ahead. Hi, Søren and Brian. Thanks for taking my question. I have two questions here. Firstly, on the current trading. Is it possible to give us an indication now how the current trade looks like? You talked about a solid pipeline, and we also talked about the improving customer activity. Have you seen those already start to.
Translate into more revenue for you already here in October? My second question is that you have done the right sizing. I was wondering if this is completed or we should expect more in Q4? Thank you very much for the two questions. The first one, are we seeing any effects already in the beginning of Q4 in terms of increased activity levels? The second one, are there further redundancies planned? Let me take the first one here. I showed on slide—on slide seven, as you could—sorry, on slide six, we have in October, we have a slight uptake in our efficiency level. I can see we have another upcoming question on that, which shows a 3% increase in efficiency. We are still not obviously where we want to be. We have the 2.5% coming from the capacity adjustment.
As to your question, the remainder when I talk about this pipeline are contracts that are due to be closed, which means for us, won or lost. As I said earlier, we have seen that even with less things to bid for, we are maintaining our win rate, but we still have to win these. They are to be won from today, where we already have one case, actually, that we have won today, through the end of November. When we talk about this pipeline, it is basically, it consists of two levels. There is one level, which I am personally very interested in. They are typically the biggest of bids. We also track the bulk of smaller opportunities that we work on.
The true effect of that is going to be seen towards the very end of Q4, which we are sort of approaching now, and then the ramp-up also going into Q1 next year. The true effect is yet to come, is the answer. The majority of the effect is yet to come. Your second one, we have no further major redundancies planned, but obviously, we are running a tight ship. We are performance managing the organization. We know that we need to look for where we have great resources that may have a little bit poor efficiency compared to what they will usually have, but they are contributing to our sales process. We want to hang on to them.
Where do we perhaps have resources that no longer fit the profile that we need so that we can also free up capacity for hiring new competencies that we need? Yeah. Thank you. Okay. Thank you very much. Thank you. Very clear. Great. Should I continue? Okay. I have two here on the screen. One is both from Michael Honoré. For several quarters, it has been difficult for you to raise your efficiency to much over 60%. It should be at least 70%. The question is, would it not be a good thing to cut into the staff? That is exactly what we have done. These are the 89 redundancies that have been made and carried out, and the DKK 11 million accrual that has been made for that exercise. That is basically having a negative effect on this quarter.
The 89, we find to be the correct figure. That is a balance between doing what is best for us right now, but also looking at what does it take for us to go back into growth. I might take the second one about the share buyback program that we are driving. You asked if we, or if we have halved it or something. What we do is basically we ask a bank to manage it for us within the fairly strict rules and calculation about what can you buy depending on some statistics from previous months or even quarters. I can trust you that we are not sort of, we do not manage it in that way. We ask them to do as much as we can within the rules they are. It is not, we do not manage that. Michael.
Trust me, I’m pretty sure we do what we can do until we reach the limit that we have stated that we would like to hit. Yeah. Yeah. The answer is we have not changed anything. That must also mean, Michael, that we have to perhaps double-check that this must mean that there is a change in the 20-day sliding average, which is what defines how much our share buyback program can do. We have not changed anything. Thank you. Okay. Any more questions? We have one more question on the phone. I’ll just put that through for you. We have another question here from Yuezu from SEB. Your line is open. Please go ahead. Hi, Way again. Just also on the question on the market condition, I was wondering, the higher activity. Customer activity you mentioned. Is this sort of a broader market.
Recovery, or is it sort of your sales effort-driven higher activity? That’s a very good question, Way. I will say to start with that I’m very proud of the sales effort that we have delivered throughout 2025. It takes a lot to stick with some of these processes that have now been delayed for more than a year and to keep the teams—and it’s not just a salesperson. It’s an entire engagement team that’s done designs, estimations, and basically wanting a result out of that, whether that be a win or a loss, but then at least you can move on. They’ve really stuck with it. We’ve never felt tempted to sort of go after new customer segments or smaller customers or a new industry vertical. We know what we do for these customers.
We know they appreciate it, and we know they’re strong companies that will come back. I think part of it has to be our sales effort. That being said, and this is purely speculation on my side, I think that a lot of our customers have been working through an initial—I was just about to say shock effect, but we’ve seen a lot of geopolitical, macroeconomic, trade policy changes compressed into very few years, perhaps starting with 2022, 2023, certainly 2024, and then culminating here in 2025. I think everybody, every responsible board and executive management team has felt the need to sort of take stock of where are we, how are we affected, how will this affect our long-term plans, short-term plans. As more and more of them—and again, it’s speculation from my side—feel that they have a firm grip of.
Understand some of the changes. Since their digital ambitions have remained the same, they tend to go back and revisit the investments they’ve planned. I think there’s also, to some degree, we’re benefiting from that. It’s hard for me to quantify how much comes from each camp. Okay. Great. That was very clear. Thank you. All right. Thank you very much. There’s no more questions on the phone conference at this point. Thank you. I think we’re six minutes past the hour. I want to respect everybody’s time. Thank you very much for dialing in. Obviously, we have a lot going on. We look forward to keeping you updated as we progress through this very exciting Q4 leading up to a fresh start on the new year. Thank you very much, and enjoy the rest of your day. Thank you. This concludes today’s conference call.
Thank you for participating. You may now disconnect.
This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.