By Iryna Kozirevych
The success of an AdTech business depends on an entire range of factors; however, the most basic thing that you can (and should) do is to stay on top of the industry trends. Keeping an eye on them enables you to adapt your strategy accordingly and maintain a competitive advantage.
The AdTech landscape never stands still, with new tendencies regularly entering the stage. But, obviously, you shouldn’t follow each of them blindly – you must be able to distinguish pure gold from imitation. As always, we are here to help.
Today, we are interviewing Sotiris Oikonomou, the Managing Director of MarkApp, and Roman Vrublivskyi, the CEO of Attekmi, to learn more about four AdTech trends that are currently shaping the industry: AI, data collaboration, SPO, and CTV advertising. Are these trends something you should pay attention to or just temporary hype? Keep reading to find out!
Welcome to our interview series, and thanks for joining us today!
AI has been on the list of the hottest AdTech topics for the past several years, but does it really bring measurable value to the industry? Are any parts of the programmatic ecosystem vulnerable to AI hype without real ROI?
Sotiris: Well, AI definitely helps to improve targeting accuracy and the overall efficiency – these are just some of the applications. However, the value varies. Many programmatic platforms use AI to automate certain processes, but do not explain the logic behind those decisions, which makes results hard to verify. True value comes when AI-driven insights are fully transparent and can be aligned with business outcomes like conversions or cost savings.
Besides, many advertisers expect AI to do everything on its own, and that is a huge mistake. Let’s take campaign optimization and audience modelling as examples. Sure, AI can deal with these tasks, but it still requires robust, clean data and human guidance. Otherwise, it may reinforce bias, aim for impressions that never convert, and so on. That is just an example, but I think virtually any component of the programmatic environment can be affected by AI hype if you integrate or use the technology carelessly.
Roman: AI can definitely bring measurable value, but only when implemented with purpose and under human supervision. AI enables predictive modelling, more effective optimization, fraud detection, and much more. It helps businesses act faster and at scale. However, the measurable ROI also depends on how well AI is integrated with business objectives and strategy. AI is not a silver bullet that can fix everything independently. If your strategy is good, it will amplify it. If it is poor, things will get only worse.
I also think that many AdTech vendors label their solutions or certain features as “AI-powered”, even when there is only some basic automation involved. I guess that is especially true for predictive bidding and creative optimization. But the thing is that AI has turned into a buzzword. If there is no visibility into the model’s performance or clear ROI metrics, such products or functionalities are not likely to deliver the results you are in search of. Yes, AI can bring value, and it will keep bringing it. But we should remember that it is a tool, not just a word you can add to your marketing materials.
One of the most popular concerns regarding AI is that it may replace human experts. Is this relevant to AdTech?
Sotiris: Sure, it is. Today, integrating AI into your workflow is essential if you want to keep pace with how fast the industry is evolving. At MarkApp, we have already deployed five AI agents across operations, product, reporting, and client support. They don’t replace talent, but they amplify it, allowing our team to focus on strategy, optimization, and real growth. In this environment, AI becomes a force multiplier, and companies that ignore it will simply fall behind.
Roman: I cannot agree more. AI can help deliver tasks faster and more effectively, detect opportunities or weaknesses that a human may miss, and so on. However, it requires thorough human control. AdTech experts have no reason to be afraid, but they should learn how to use AI tools and keep an eye on their evolution – that is a way to deliver better results and remain competitive.
Data collaboration and clean rooms are gaining momentum as a privacy-safe targeting tactic. Do these measures ensure accurate insights or just add technical complexity? Can smaller companies benefit from data collaboration?
Sotiris: Just like in the case of AI, clean rooms and data collaboration can bring measurable results, but the value depends on execution. If the data entering the system lacks quality and structure, the output will be poor as well. In turn, if the data is of high quality, you can count on valuable insights and effective targeting. Some technical complexities are obviously involved, as you need to integrate the room with existing systems, secure the data, etc. But if everything is done right, the results are worth the effort.
As for small brands, these complexities, along with initial investments, may introduce a challenge. Building long-lasting partnerships is a way to solve it.
Roman: Since the whole industry is adopting the privacy-first approach, data collaboration and clean rooms are more than a trend. They allow us to work with audience data while respecting user privacy. If everything is done correctly, the results are far more accurate than cookie-based targeting ever was.
However, I agree that smaller companies may find it difficult to benefit from clean rooms, and that is not only about technical barriers or associated costs. To provide meaningful insights, clean rooms often require large volumes of data, and small businesses may simply not have them. I suggest collaborating with key partners and starting small. Clean rooms can be very effective, but it is always better to test the new approach before adopting it.
How do you expect the data privacy trend to evolve over the next, let’s say, three years?
Sotiris: I expect more innovations to enter the stage. AdTech players will adopt technologies like differential privacy to work with data in a secure way.
However, even being shaped by evolving regulations, the trend will not limit our approach to targeting and measurement. Instead, it will refine our methods so that we will still be able to personalize our advertising content without pushing users’ personal boundaries.
Roman: I would like to add that a privacy-centric approach will stop being treated as a compliance checkbox. This is already happening, and I expect this tendency to grow. The privacy-first model will become a significant competitive advantage in the near future.
At the same time, I guess, privacy regulations will keep getting stricter, while stronger consent frameworks are also likely to enter the stage. That is a logical evolution – we will never get back to the days when everything relied on third-party cookies and the Internet knew everything about you.
Okay, let’s talk about supply path optimization. How can media buyers ensure that SPO really reduces cost and fraud? Is SPO a must or just another trend that will disappear over time?
Sotiris: SPO is often defined as reducing the number of intermediaries, but from my perspective, it is more than just cutting supply partners. It is about choosing paths that can deliver predictable quality with minimal risks. How to make sure that this really works? As usual, conducting regular audits is the key. Compare the performance of direct and indirect paths to see what works better. Using ad verification tools is helpful as well.
To me, SPO is getting more and more critical since the supply landscape is becoming more fragmented. The trend is still evolving, but it is here to stay.
Roman: Definitely, SPO is not only about building direct partnerships with media owners. Instead, advertisers need a clear visibility into every step of the supply chain. This involves comparing fee structures, validating sellers.json entries, reviewing log-level data, etc. Such an approach enables advertisers to implement SPO in a data-driven way, which will help minimize both costs and the risk of fraud.
I do not consider SPO as a passing trend. Nowadays, CTV, DOOH, and audio ads are gaining traction, which makes programmatic ecosystems even more complex. Therefore, SPO is turning from a cost-effective tactic into a fundamental practice. In the long run, it will become a must-follow approach, especially for those who plan to scale their operations.
Since SPO is still gaining popularity, how do you expect AdTech providers to respond?
Sotiris: AdTech providers need to focus on transparency and verifiable quality. Just offering direct deals is not enough – media buyers need stronger fraud protection, path-level analytics, and automated recommendations for optimal supply connections. Providers need to prove the effectiveness of their platforms with data, not promises.
Roman: I expect platforms to offer more advanced inventory verification capabilities, as well as clear documentation. Providers with weak supply quality or unclear value propositions are not likely to survive unless they change their strategies and listen to their clients. That is exactly what we do at Attekmi – for instance, we released AdEx Enterprise, our ad exchange platform with advanced functionality, in response to our partners’ needs. We also create custom AdTech solutions to bring the unique and innovative projects of our clients to life.
The same applies to intermediaries. Since the SPO trend is not going to fade away, they need to find new ways to bring value, as well as ways to prove that.
What about CTV advertising? It is also on the rise, but measurement and attribution still remain a challenge. Can it be solved effectively, or will CTV marketing soon lose its popularity?
Sotiris: Despite certain challenges, I do not consider CTV advertising a temporary hype. It allows advertisers to deliver high-quality, engaging content – CTV ads are pretty hard to ignore. As for measurement and attribution, things will get better when device graphs get smarter and reporting becomes more consistent. CTV advertising will not lose its popularity; it just needs standardization. This will take some time to happen, but that is simply inevitable.
Roman: Clean rooms that we discussed previously can help, as well as greater transparency from the publisher side. Obviously, these challenges cannot be solved immediately and require collaborative efforts. However, we are already seeing some progress – in the meantime, probabilistic and deterministic measurements are becoming the norm. And yes, CTV advertising is not going anywhere. Linear TV still exists, but viewers are switching to streaming platforms. Besides, CTV marketing enables accurate targeting – this is something that linear advertising will never be able to offer.
More and more players are entering the CTV environment. Does this make the ecosystem more efficient or more fragmented?
Sotiris: Well, from advertisers’ perspective, more CTV players mean more inventory and a greater number of opportunities – that is definitely a benefit. But yes, this also means greater fragmentation. That is our current reality, but I expect different platforms and providers to collaborate on measurement standards and simplify access for media buyers. Until this happens, advertisers need stronger tech stacks to deal with the complexities.
Roman: I agree, these are two sides of the same coin. On one hand, competition drives innovation and opportunities. On the other hand, it is the power behind increasing fragmentation. Each new platform has its own data policies, measurement rules, and so on. That is something that contributes to the measurement challenges that the industry is already facing. What we need is consolidation, interoperability, and stronger shared standards. Over time, those players who cannot keep pace with the efficiency evolution will simply leave the market. That will enhance the CTV ecosystem and the programmatic industry in general.
Thanks! And the last question – in your opinion, what is one trend or innovation the industry is not talking enough about yet?
Sotiris: Contextual intelligence in a privacy-first world. The industry still focuses too much on IDs, while the real innovation is happening in content-level understanding across CTV and mobile apps. Advanced contextual targeting now blends metadata, behavioural patterns, screen type, and engagement signals to deliver strong performance without personal data. With identifiers disappearing and regulations tightening, this is becoming the most scalable and future-proof approach.
Through Pantheon, our contextual bidder, we see CTV and in-app inventory performing exceptionally well without relying on IDs – and the industry isn’t talking about this shift enough yet.
Roman: I would say that interoperability is exactly the trend that deserves more attention. We keep building new tools – ID solutions, AI-powered products, clean rooms, and so on. However, these solutions rarely work smoothly together. Standardization is essential not only for effective CTV measurement. The whole industry requires standardized ways for platforms to exchange data and minimize technical friction. From my perspective, interoperability drives ultimate value. As soon as it becomes a popular trend, the AdTech niche will start moving toward a brighter future.
Sotiris and Roman, thanks for sharing your thoughts – that was a very meaningful discussion! We hope to talk to you again soon. The AdTech industry keeps evolving, so there is always something to explore and discuss.
Follow our interview series – with the insights from leading AdTech experts, it will be easy for you to stay on top of trends and news, and, therefore, adjust your strategy and move in the right direction.
Ready to launch your own ad exchange solution in no time? Contact us.
