For years, Marsch was divided into Registration systems and Commitment systems.
Registration systems stored “teachers” data versions. CRMS and CDP for customer data. PIM for product data. Drive for brand assets. ERP for inventory data. (I know, there are a lot of acronyms in this piece, almost to the point of parody. Click the image above to obtain a large -scale version to read the Ring of Acronym Decoders to the right).
Participation systems interacted with customers, either directly with tools such as email delivered to maps or web experiences delivered in DXP or indirectly with tools such as SFA that support sales equipment that were dedicated to customers.
However, in reality, this division was Nart so clean. For example, CRMS and CDP often combine data management and customer participation functionality. But as a way of talking about different categories of Marsch and its roles in a technological pile, it is enough.
But I don’t think it’s the right mental model.
With the revolution of the sweeping everything in marketing and technology, I think that a better frame is Context systems and Real systems.
Now, narrowing my eyes on my diagram, I might ask: “Are not just these participation systems and registration systems with different labels?” They are similar, yes. Both delineate approximately a layer of data and a layer of services. But here is why they are different.
First, in the data layer, the registration systems of historically combined responsibilities for the storage and arbitration of data. Each registration system had its own database that was closely combined with business logic to determine what could be written or read in that database and in what format.
In the model that I have drawn here, the Real systems The pile section recognizes that the storage and arbitration thesis has now separated from the legs, thanks to the increase in the cloud/Lakehouse data warehouse.
Data warehouses in the cloud/lake -casas store and distribute data throughout the organization. But the very force of this universal data layer, which can house each and every one of the data that flow throughout the organization, is also its weakness. The data that are flooded are not necessarily standardized or rationalized in all its varied sources and destinations. Conflict and content can work unbridled.
You still need software to arbitrate what is correct and canonical data, which governs how it is written or read, in which format, validated with standard definitions and relevant commercial logic. With something as important and complex as the records of master clients, which is traditionally stored in CRM arbitration and governance and governance is not trivial and is critical of the mission.
This is the reason why many “classic” Marsch registration systems: CRM, MDM, PIM, dam, etc. – They still play an important role in the era of the cloud/lakehouse data warehouse: they are still the truth arbitrators for the data within their domain. Even if your data is increasingly stored in an independent layer below.
Will there be a new generation of such data arbitrators platforms? Or will today’s main platforms evolve to adapt to this new environment? Probably both.
These Martech classic data platforms also provide useful Contextualization Or data. For example, combining a list of customers with a set of advertising audience segments creates a context For those client data within a specific marketing campaign. This is exemplified with compound CDPs. They work directly with the data stored in a data warehouse in the cloud, but organize and manage that data for a wide variety or different contexts in which marketing specialists want to use it.
In fact, a compound CDP is possibly more Context system than the real system.
(If the word “truth” bothers you for philosophical reasons, and you would prefer that we change it to something else, I’m sorry, we put ourselves aside).
We never achieved a real single Real system (SSOT). It turns out that there are too many specific domain data. But with a universal data layer on the lower and specific data governing platforms in the upper part, we now have many systems. plural – Or truth.
So what does it do? Context systems Different from commitment systems?
The commitment systems have been relatively “fixed” in the context they provide. For example, with the MAP and CEP platforms, he learned to use them, often adapting the way he works to their structures and processes. When he builds a website on a DXP, the experience that clients received was the context he had in mind when he designed it. It is not really its context. It is the context your Think they have.
With the emergence of AI agents, with the employee -oriented context and with the client, it is created more dynamically. Many AI agents can be turned, each adapted to a specific work task or flow for an employee or hyperpersonalized for the experience of a specific client.
Context systems differ from participation systems quantitatively – Increasingly, there are more AI agents that proliferate through the battery than traditional SAAS platforms, and qualitatively Because they are a purpose for much more specialized contexts.
The extreme incarnation of this is the individual AI agents who create software experiences on the march for each employee or client, for any work they want to complete at that time. This is what we describe as the new “hypertail” or Marshal Marsch for 2025 Report a couple of months ago.
Dynamically generated customer -oriented agents, what we will call CONSERJE OF AI AGENTSIt will offer many more relevant experiences with contextation to these users. They will hear exactly what the client wants and will report all the existing data of the clients and the company in our real systems, will deliver exactly the content and services that the client wants on that album.
Instead of a monologue, where a brand serves a contextual experience based on its own definition of the buyer’s trip and what is Thought The buyer wants, the agents of the conserve will involve buyers in a real dialogue To understand and serve your real context.
A little further on the horizon, but maybe not so far, buyers will use their own AI agents to interact with our context systems. These arural agents that control the vendors. These are customersproprietary Ia agents They Control. Inherently they will shape the experience of the buyer’s context.
Two more points:
First, in the stack scheme in the upper part of this publication, I included a complete group or a differential of product products. I do not say that each pile needs every thesis. Unless it is in a large company, it is likely that you only have, need or because it is a subset. I only included a lot to show where I thought the different categories were adjusted in this architecture. Nor is it in any way.
Second, to illustrate this as a pile, perfectly packaged components, stacked one on top of the other, such as perfectly adjusted LEGO blocks, it is not a precise representation of reality.
The stack view is easier to understand: at least I think so, since this is how we have thought about Mars Pilas for so long.
But a more precise representation would be a graphic view. All these different products and platforms, and all these different applications, agents and automatities are all nodes in the cloud that connect with any of the others. I still think about truth systems at the center of this cloud, surrounded by many context systems.
In short, yes, we still have records, and we still have commitment.
But the defining characteristic of Mars the “pile” in a world of AI will be context and the TRUE It is wrapped.
PD that speaking of Marsch batteries, we hope you consider sending a slide illustration for our Awards 2025 Stackie. Seriously, it is one of the most fun prize programs throughout marketing, certainly in my highly biased opinion. But there is no rate to enter, and in fact, we donate $ 100 to UNICEF for each legitimate entry, up to $ 10,000 in total. The deadline to enter is April 4.
