Data Modelling
Data & TrackingAlso: Data Model · Marketing Data Model
Quick definition
Data modelling is the process of deciding how your marketing data is structured, named and related before it's stored. It defines what a session is, what counts as a conversion, and how a user in your CRM connects to a session in your analytics tool.
How it varies across Australia
Most Australian businesses build their data model by accident, one tool integration at a time, rather than by design. The businesses with a genuine competitive edge in data and tracking tend to be the ones who modelled their data before they built their dashboards, not after.
See data and tracking maturity across Australian industries →What it actually means
Data modelling is the blueprint stage most marketing teams skip. Before you build a dashboard or connect a tool, someone has to decide what a 'customer' actually is. Is it a person, an email address, a household, a company. Someone has to decide what a 'conversion' means, and whether it's the same definition your finance team uses.
Without that blueprint, every tool you plug in invents its own answer. Your CRM counts customers one way. Your analytics platform counts sessions another way. Your ad platform reports conversions on a third definition. None of them are wrong exactly, they're just answering different questions, and nobody agreed on the question first.
A good data model sits underneath attribution, segmentation and lifetime-value calculations. It's the layer that makes those things consistent instead of each report contradicting the last one. It's unglamorous work, closer to plumbing than strategy, but every serious data and tracking capability is built on top of one.
A dashboard is only as trustworthy as the model underneath it. Most businesses build the dashboard first and discover the model was wrong later.
How it shows up
Data modelling shows up as the schema behind your data layer, the entity relationships in your customer relationship management (CRM) system, and the joins your analyst writes to connect ad spend to revenue. It also shows up in disagreements. When the marketing report says 400 new customers and finance says 310, that gap is usually a modelling problem, not a data quality problem. Two teams built two different definitions and nobody reconciled them before the numbers hit a slide.
The Australian context
Australian businesses running on a mix of local and global platforms (a Shopify store, an Australian-hosted CRM, a US-based ad account) often inherit three separate data models by default. Each platform ships with its own assumptions about what a session or a customer is. Reconciling those into one shared model is usually the first real step toward genuine attribution and reliable segmentation, and it's a step most Australian mid-market businesses haven't taken yet.
Where people get this wrong
Related terms
Common questions
What's the difference between data modelling and a data layer?
A data layer is where events and values are captured and passed to tools. A data model is the underlying blueprint that decides what those events and values mean and how they relate to each other. The data layer is the plumbing, the model is the plan the plumbing follows.
Who should own data modelling in a marketing team?
Ideally a shared decision between marketing, finance and whoever owns analytics or engineering. The definitions inside a data model, like what counts as a customer, affect reporting across the whole business, so it shouldn't sit with one team in isolation.
Do small businesses need formal data modelling?
Not a formal schema, but they do need agreement. Even a shared one-page document defining what counts as a lead, a customer and a conversion prevents most of the reporting disagreements that come from tools using their own default definitions.
How does data modelling affect attribution?
Attribution depends entirely on how sessions, users and conversions are defined and connected. A weak or inconsistent data model produces attribution reports that look precise but rest on definitions that don't match across the tools feeding them.
Debrief
Get the next one
No spam. No fluff. Just the next article, straight to your inbox.
Keep exploring
About New Rebellion
New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.
How we think →