TMW #207 | Martech predictions for 2025

Welcome to The Martech Weekly, where every week I review some of the most interesting ideas, research, and latest news. I look to where the industry is going and what you should be paying attention to.

👋 Get TMW every Sunday

TMW is the fastest and easiest way to stay ahead of the Martech industry. Sign up to get TMW newsletters, along with a once-a-month complimentary Sunday Deep Dive. Learn more here.

Predicting the year ahead in Martech.

I know what you’re thinking: Predictions are stupid. Even TMW CEO Juan Mendoza thinks so… this is how he described predictions in TMW Special Edition | Three trajectories for 2024: Part 1:

“Predictions are for the weird fortune tellers, horoscopes, and people who blame having a bad day on ‘Mercury is in retrograde.’”

Wow, that’s a damning statement! So, surely, I wouldn’t go ahead and do predictions for 2025?

Wrong! I am very much going to do that. Let me explain why. 

Firstly, yes, predictions are in some ways destined to fail. They exist along a spectrum, with one end being the painfully obvious, and the other end being wild stabs in the dark. You either make predictions that are correct and therefore too easy, or you make ones that turn out to be incorrect and are just plain dumb. Every now and then, however, you throw a Hail Mary and it sticks. Much like sport, the outcome usually defines the success of the preparation.

But even within incorrect predictions come valuable observations. Sometimes the timing is wrong, so the predicted outcome arrives later than expected. Sometimes there is a Black Swan event or course correction that occurs in retaliation to a particular trajectory. Understanding the drivers of the current direction of travel is useful, even if there is a sudden left turn.

Take, for example, our three trajectories for 2024:

1. Customer identity is going to get messy
2. Generative AI will start to spiral
3. Cloud platforms will try and fail to get into Martech

In ‘Customer identity is going to get messy,’ Juan predicted that Google’s deprecation of third-party cookies (3PC) would fracture the customer identity landscape into a sea of different cookie alternatives (e.g. UID2.0, RampID, Panorama ID) and stress test whether first-party data really is the future of advertising. Well, he was right about it being messy: Google delayed 3PC deprecation under pressure from the UK’s Competition and Markets Authority and pivoted to a user-led approach, but has given no more information since then. Although the messy identity landscape that 3PC deprecation leaves behind hasn’t yet been tested in the way Juan predicted, it may play out once Google produces an actual plan for deprecating 3PCs.

Juan nailed the trajectory of GenAI in ‘Generative AI will start to spiral,’ predicting that LLMs would become table stakes, proprietary data combined with Small Language Models would become more valuable to marketers, and that scrutiny of GenAI would accelerate. As Juan covered in TMW #203 | The (A)ick, GenAI has become a challenging distraction for marketers and is emitting an increasingly pungent odor due to the sickening scale of investment and the lack of humanity that it is often leading to.

When it comes to ‘Cloud platforms will try and fail to get into Martech,’ perhaps a better way to describe this trajectory – with the power of hindsight – is to say that cloud platforms are dragging Martech towards them. As we covered in TMW #200 | Data Gravity, the warehouse-native approach to data management is increasingly positioning the cloud data platform as an ecosystem within which the Martech stack is assembled. The stack is moving to the data.

So, without further ado, let’s crack out the crystal ball.

2025 Predictions 

Vendors will vacate the CDP category en masse 

We already started to see this happening at the back end of last year. Most notably, Uniphore acquired ActionIQ and paired it up with Infoworks to create a Zero Data AI Cloud. Amperity positioned away from the category, rebranding as a Customer Data Cloud. 

According to G2, the CDP category is oversaturated with 260 vendors playing in the space . This makes it very hard to stand out in the category, which has created a massive product marketing challenge. Plenty of vendors have tried to differentiate with questionable derivatives of the CDP (Lytics’s Experience CDP, Census’s Universal Data Platform), but the key differentiator right now is composability.

As we covered in TMW #187 | The contracting data supply chain, cloud data platforms and marketing activation tools are also encroaching on CDPs. Whether it’s data lakes and warehouses rolling out identity features, or Customer Engagement Platforms developing data orchestration capabilities, CDPs are getting pressure from both ends of the data supply chain.

Expect lots of change in CDP land in 2025. Plenty of vendors will be acquired and re-positioned – or will re-position themselves. For the sake of survival, many will need to get out of the CDP category. Some sadly just won’t make it and will be sold for scraps.

Martech consolidation is coming

We have already published an essay on this one, so I’ll stick to the CliffsNotes.

There are a bunch of zombie Martech companies out there that took on large investments during the Zero Interest Rate Period (ZIRP) and haven’t been able to back that up with growth during the challenging recent macroeconomic conditions. Investors will be looking to recoup as much of their investment as possible and move on to higher growth opportunities.

This is likely to concentrate itself in the “torso” of medium-sized Martech companies that will look to survive through M&A. Either the biggest Martech companies will buy them out, they will merge with competitors or complimentary products, or they will be taken out of category by companies that want to apply their tech to non-marketing business challenges.

Even in the “tail” of abundant small and micro companies in the Martech industry, startup closures are on the rise. There is a very small barrier to entry nowadays in Martech, so perhaps AI-native apps will replace the startups that are closing shop, but we will nonetheless see plenty of churn.

Get ready for plenty of ugly exits.

AI agents won’t be a transformational technology in 2025

AI agents became the battleground product for big Martech companies in the second half of 2024, which led to a lot of posturing and plenty of splashy announcements from the likes of Salesforce, Microsoft, and co. But AI agents won’t be a transformational technology in 2025 for most legacy brands.

There is a lot of data out there to suggest we’re a long way off AI agents transforming businesses. First up is a mindset challenge: Brands are prioritizing “soft” targets for GenAI projects, and CFOs around the world will not let that go on for much longer. Secondly, GenAI adoption has been much slower than marketers expected. And finally, businesses are still just not ready in terms of data, technology, and personnel.

The other challenge is change management. Legacy companies move slowly – and particularly so when it involves taking risks like setting AI agents loose in their business. Only 7% of companies have a governance framework in place for deploying GenAI. Adoption of AI agents will be a slow burn for most companies, leading to efficiency improvements for precise, contained use cases in the first instance. Either that or marketers managing large, complex Martech stacks will focus on improving their current investments rather than splashing out on more AI.

Will AI agents one day become a common sight in businesses? Probably yes. Will they live up to the hype and transform how businesses operate? Not in 2025.

Compliance becomes a top-five skill for every marketing role

Compliance is already an important part of a marketer’s role, but as regulation that impacts marketing becomes both broader and deeper, it is going to be catapulted onto the list of one of the top five things that marketers look for when making hiring decisions.

In terms of data privacy, 79.3% of the world’s population is covered by some form of regulation. It used to just be marketers operating in Europe that had to worry about the GDPR, but countries around the globe are now adopting European-style privacy laws. In APAC, this has seen several countries expanding existing regulations. The big one is the US, which last year introduced the American Privacy Rights Act (APRA), although a change of president may put that on the back burner for four years.

But it’s not just privacy compliance. Marketers have a series of new regulations to navigate in 2025. Europe’s Digital Markets Act (DMA) and Digital Services Act (DSA) have started to bare teeth in the second half of 2024, which will increase in 2025 as more companies are designated as gatekeepers and more enforcement action becomes available to regulators. Marketers will need to build strategies for when regulation enforcement drastically changes the landscape, like when the EU deemed Meta’s ‘pay or okay’ consent model to be illegal. This isn’t just a European problem: The USA is laying groundwork for similar regulations across the Atlantic.

Plus, AI regulation is just kicking off. The EU is the vanguard in this respect with its AI Act, and although the act won’t be fully in effect until 2026, there are components becoming valid in 2025 that require immediate action. AI has become an everyday part of marketing; every single marketer needs to know what they can and can’t do to get the job done within legal – and ethical – boundaries.  

AI decisioning arrives in the Martech stack

AI decisioning has been around for a long time, with companies like SAS and Pega pioneering the idea that you can use statistical modeling to determine the best decision to take to drive customer experience outcomes. In the past, these technologies have had an inbound focus or have been applied more towards customer service, but that is about to change as decisioning becomes the clearest application of applied AI in the Martech stack.

We’ve already seen companies like Hightouch, OfferFit and Movable Ink launch products in the space, and I expect that we’ll see more marketing-focused AI decisioning products released throughout the year. Hopefully some big brands take the plunge, resulting in high-quality case studies showing the power of AI as a centralized decisioning engine.

Most brands use some level of applied AI within their marketing, but this will explode this year, assuming marketers can achieve a level of trust in the technology to make the right decisions when bringing more of marketing into its scope. Brands – particularly those in commodified industries – will look to AI decisioning to drive better, scaled customer experience.

AI decisioning used to be more about experience, but in 2025, it will become a lever for growth.

LLMs and chatbots will become a genuine consumer marketing channel

No, they probably won’t knock Google Search off its perch, but the conditions are in place for AI “answer” engines to become a serious consumer channel.

The first condition is that Google Search has become terrible. Yes, everyone still uses it, but the experience is not anywhere near what it used to be. Google also has a terrible recent track record in rolling out GenAI to Search, which opens the doors for competitors to roll out a better way to search for information and products, with a bigger focus on UX.

The second condition is incentives. Perplexity, OpenAI, and even Google’s AI Overviews have or are planning to introduce ads, which is the only way these tools can achieve the scale to make them genuine consumer channels.

The third condition is that GenAI chatbots present a new and different way of searching for information and products. The conversational interface, the multi-step capabilities, and consumer AI agents that can act on the consumer’s behalf all ladder up to a different and simpler user experience that is a significant step change from the incumbent.

Google Search is so ingrained in how we search for information that it is not going to be toppled in a year, but there will be enough traffic for certain types of companies to invest in AI chatbots as a consumer marketing channel. 

So, those are our predictions for the Martech industry in 2025. Will we look back at these with horror at the end of the year? Probably. Will I try to twist the events of the year to prove my predictions were correct? Only time will tell. Want to join me? Reply to this email with your number #1 prediction for Martech in 2025 and I’ll repost the best to LinkedIn.

All I know is that we are in for a hell of a year with plenty of change, but also plenty of the same opportunities and challenges.

Plus ça change, plus c'est la même chose.

Stay Curious,

Keanu Taylor

Make sense of marketing technology.

Sign up now to get TMW delivered to your inbox every Sunday evening plus an invite to the slack community.

Want to share something interesting or be featured in The Martech Weekly? Drop me a line at juan@themartechweekly.com.