2025 Technology Predictions
‘AI is infused throughout martech at this point. Nearly every major martech vendor has added generative AI features into their products. And hundreds of new martech startups have been born natively on the back of LLMs. It’s ballooned the martech landscape to more than 14,000 products’ (Chiefmartec, The State of Martech 2024)1
Bear in mind, when Chiemartech first released what would become their annual research into the marketing technology landscape in 2011, there were only 150 products listed. They also emphasise that there are thousands of marketing related chat agents available that are built on existing AI frameworks (such as OpenAI) and are not included in the main numbers.
Despite this rapid and continued growth of new tools and the potential AI is unleashing for streamlining how existing products fit into and improve our own marketing stacks; the necessity for collecting and maintaining quality data has never been as important…or exciting. Looking forward to 2025, it is clear that there will be an increased emphasis on the flows of data around data warehouses, as well as individual tracking and reporting platforms. Additionally, the influence of new methods for activating business intelligence will significantly impact how businesses connect with customers.
THE REPORTING LAYER BECOMES THE MAJOR PLAYER
In recent years, data centralisation has been the talk of the town; or maybe I just go to some strange pubs. Consolidating data from various sources – such as analytics, advertising channels, shopping platforms, and CRM systems – usually focused on a single dashboard has become essential. The main advantage lies in the significant time saving, as it eliminates the need to compile reports from disparate sources. While beneficial, this approach has predominantly focused on retrospective reporting and been siloed to individual channels, with any actionable insights being implemented manually elsewhere.
The current phase of data and marketing technology involves the widespread use of centralised data sets in cloud data warehouses for many businesses. The focus over the past year has shifted towards enabling seamless interrogation across data sources and creating combined cross-platform measurements. This development is now contributing to both business performance and enhancements in customer engagement.
Beyond just dashboards as a reporting hub, 2025 will see reporting layers connected to an increasing number of other tools and applications using AI, machine learning and new ways of engaging with and activating data. Why now? There are a lot of factors. Other tools and software developing their APIs to improve not only the push into data warehouses, but also work with the data coming out – making it easier to connect data and automate what you do with it. Increased use and capability of AI to not only digest and simply analytics, but also take some automated actions based on insights is playing a role.
Analysing customer journeys for all but the simplest of business cases is extremely difficult – not only in having to think about the vast array of marketing touchpoints and the impact of increasing engagement and traffic from AI chatbots, macroeconomic conditions, user generated content, influencers. Using techniques and tools to ensure you can use the right messaging and creative to suit the customer at each stage, rather than worrying about the stages themselves should increase the focus of everything on what your ultimate goals are. Being able to speed up the process of monitoring lead quality, customer satisfaction and optimising to those goals will see incredible boosts this year.
GA 4 MORE THAN JUST REPORTING
GA4 is a great example of a shift to data being activated in a platform, mostly aided by machine learning capabilities. A good bulk of the features to drive that have been adopted by some businesses relatively recently and will certainly become standard practice for all users in the coming months.
Universal Analytics finally left us in the summer, cast under the bed like Woody as we all run off to play with Buzz. In the four years (four years!) since GA4 was officially launched, a lot of its potential for embracing machine learning and being able to track more detailed customer engagement with websites and apps was being outweighed by frustrations on the new interface and reports (limited if you didn’t embrace the flexibility for customising everything). The past 18 months had also seen a focus on Consent Mode and ensuring businesses were actually collecting user consent for tracking in Analytics and able to make use of modelling to uplift missing data.
We are getting beyond that now and 2025 is the major departure from just passively looking at reports simply telling us how many sessions and goals we had last month. Custom Audiences are being increasingly set up and activated – using detailed understanding of user engagement and interactions to then be used for personalisation and targeting in ad platforms such as Google Ads. Built-in predictive audiences (hello AI again), mainly used for ecommerce purchase and revenue estimations, are starting to be implemented or merged with other behaviours in detailed targeting strategies. First-party user data submitted through forms on sites and apps is increasingly being captured for use with Enhanced Conversions. Where implemented this will continually be used to improve attribution and reporting; data that may have potentially been lost if cookies had been blocked.
GET CRACKING WITH SERVER-SIDE TRACKING
How you collect data in the first place is still fundamental to all data initiatives. Managing data accuracy and maintaining volumes whilst ad blockers and other restrictions come into play, whilst still always respecting user consent choices and complying with prevailing privacy regulations is a continual and important balancing act.
Server-side tracking routes user data from your website to analytics and other platforms via a server, rather than relying on the user’s web browser. Many ad platforms and third-party tools now recommend server-side tracking to ensure reliable data connections and reduce data loss. For example, GA4 event data can be sent to a cloud-hosted Google Tag Manager, which then relays the information to platforms like Meta using APIs. This approach bypasses issues caused by browser-based tracking, such as cookie blocking by privacy settings or ad blockers. However, it is still essential to obtain user consent and ensure transparency through your CMP.
Capturing data and choosing what is sent to various platforms in a more secure environment is the evolution in the next year for businesses that are on board with the importance of what their tag management systems can do for marketing best practice and effectiveness. You can read more on that here.
CDP-EASY
Businesses have really started to embrace data warehousing and advanced to the stage of integrating analytics, ad data and transactional CRM into their pipelines. The next phase of growth in business intelligence will involve more capabilities to automate customer profiling, create specific audience segments, and create and deliver personalised content – all of which will be made easier to manage by AI and ML features in Customer Data Platforms (CDPs).
The case for Customer Data Platforms is compelling in terms of attempting to get a centralised view of customer data and how they’ve interacted with you, understand their profiles and, crucially, fulfilling the idea of being able to serve personalised content and experiences through your website, ad platforms and other touchpoints. Attitudes amongst consumers towards personalised targeting and offers from brands are largely positive2, but achieving it at scale has typically required a considerable commitment of time and money with rigid CDP set up requirements.
On top of that, making use of all of the features available perhaps doesn’t always fit use cases that are unique to a business and there may be a duplication of efforts in preparing data sources to integrate with the platform, whilst also using a data warehouse for other functionality that the CDP doesn’t cover. As solutions to these potential barriers, we’ll be hearing an increasing amount about ‘CDP components’3, or even more specifically, ‘warehouse activation’ products, that will serve to close some of the loops between the customer and product data you have in your warehouse and where it can be put to effective use.
IT’S GETTING BOT IN HERE
Chatbots or conversational agents being more embedded in existing ad platforms, analytics and reporting tools doesn’t seem to be the most intrepid of predictions, but usage of those and the development of general business intelligence bots will definitely ramp up quickly into 2025.
It is the cross-channel or even ‘cross-insights’ view that is the most exciting. Being able to ask holistic questions on customer health or product performance, whilst also getting responses on specific campaign or ad results if desired, is better suited to empowering multiple stakeholders beyond always looking at things from the same angle in dashboards. We often find that 85% of the work for a dashboard with multiple sources is in the data collection, cleansing and transformation and often only a fraction of the available data points and ways to create custom calculations are realised in traditional BI reports – more will be unlocked by conversational analytics and actively support business decision making in the very near future.
The growing appreciation for the power of a consolidated reporting layer has also led to more industry and business-specific data (such as product pricing and inventory) being connected up to ad delivery, call tracking and web engagement data. Prime conditions for predictive analytics features to come to life and start being deployed in some of the sectors we work in, which will no doubt be connected back into chat interfaces to both understand the forecasts being shown and be used in deeper assessments of business performance and potential.
- https://chiefmartec.com/wp-content/uploads/2024/05/state-of-martech-2024-report.pdf
- https://customerthink.com/composable-cdp-is-dead/
- https://www.marketingweek.com/dynamic-pricing-personalisation-brand-safety-5-interesting-stats-to-start-your-week/
Ref: https://spaceandtime.co.uk/blog/the-importance-of-gtm-for-marketing-best-practice/