How is plumbing related to data architecture?

When I was in Verona, I asked a few Italians: What’s the most important part of pizza? Which is like asking someone who’s God. Some said cheese, others said sauce, but the answer that rocked the room was dough. They battled over this like wolves over sheep, because it’s the base of everything. Every bite is affected by this.

So, in pizza, the dough is the key. But in data-marketing, data is the key.
To gather valuable insights, build smart & personalized customer journeys, we need our data to be organized in a way that we can work with it.
Data architecture is the way your data is managed – from gathering information to its flow into different systems. The goal is to ensure our data is managed properly and meets business needs.
In an ideal state, the data architecture is planned before you get data. However, most businesses do not do it, resulting in a chaotic database. In addition, the data sources & the data itself change with time, and we have to update the architecture.

So, how do you create the right data architecture?

Mapping and collecting – The first step is to map all the places where we store information and collect it from the various sources where it is stored.

Cleaning – Make sure there are no duplicates or errors in the data
(Like: @gamil instead of @gmail)
Consolidate – Find the various identifiers that will allow us to combine the information into one customer card (postcode, email address, phone number, etc.).

Several teams must work together to build the data architecture.

1. Marketing – They need to define what data is relevant

2. Business growth\ management

3. IT/Developers – They have to build the infrastructure and make sure the data flows right 

4. Data analysts

5. Cyber security

Tech-wise, there are several topics to consider when planning the data architecture: Do you plan on creating a data lake? How will the data flow? Will you use external tools or develop them yourself? Etc. But before diving into technology, we must first understand what information we hold.

Where does it go and what are our marketing and business objectives? Only then can we choose the appropriate technological solution.

Sources of information

The site's management and referral system
The order system (on an eCommerce site).
The CRM system
POS system
The customer service/chat system
The survey/opinion system
The analytics system
Marketing automation system
Different databases
Excel files

And more

How is plumbing related to data architecture?

Think of the data architecture in an organization as a pipe system – when everything is connected in an optimal way, there is a good flow of information from all the different sources (website, CRM system, POS system, etc.). There is no leakage of information (information that gets "lost" and is not used to give better service or as part of customer journeys).
When building the data architecture, it's a must to connect the various sources. When information flows, we need to ensure it doesn't leak.

What are the risks of bad plumbing?

A data architecture that is built in an overly complex way is also known as "spaghetti architecture" (no relation to the flying spaghetti monster). The term refers to an overly complex information architecture with many connections. If we take the example of plumbing we gave earlier – think of a system of pipes that is supposed to transfer information from point A to point B. Instead of transferring it smoothly between the points, the information flows from point A to point C then point D then point F and finally reaches point B.
The result is a high chance that the information will not pass properly or stop passing optimally soon.


A well-designed data architecture allows a business to use all the information it has collected. This allows it to provide better service, analyze the information and gain more business insights which can lead to better business performance and a competitive advantage. There are additional benefits to a properly planned data architecture – cost savings (consolidated information that saves on storage costs, less information that needs to be cleaned, arranged and consolidated before analysis, etc.), increased security capabilities, etc.

Like this article?

Hippobyte Newsletter

A bite from Data, Marketing and everything in between.

We got a new lead!


Check out your e-mail inbox