Over the past few years, HighLevel has undergone a significant transformation. What began primarily as a marketing automation platform has steadily evolved into a comprehensive customer relationship management (CRM) system.

That evolution changes how we should design our databases.

Many experienced HighLevel users still build systems around tags because that was historically the recommended approach. However, as HighLevel has introduced more advanced CRM capabilities—including opportunities, custom reports, relationship tracking, custom objects, AI, workflows, dashboards and increasingly sophisticated custom fields—the architectural role of tags has become much smaller.

In many modern HighLevel implementations, fields should now be considered the primary source of truth, while tags have become secondary tools used for categorisation and automation.

The Difference Between a Marketing Database and a Relationship Database

Traditional email marketing platforms such as Mailchimp, ActiveCampaign and early marketing automation systems primarily focused on audiences.

Contacts were organised using:

• Tags

• Lists

• Segments

• Campaign membership

The objective was simple:

Who should receive this message?

A CRM has a very different objective.

Instead of asking who should receive an email, it asks:

• Where is this customer in their journey?

• What is their current relationship with the business?

• What happened previously?

• What should happen next?

This requires structured data rather than collections of labels.

Understanding Database Architecture

Every CRM needs a single source of truth.

For example, consider a quotation process.

A customer may move through these stages:

• New Lead

• Quote Calculated

• Quote Sent

• Awaiting Follow-up

• Address Received

• Booked

• Completed

At any given moment, the customer occupies only one of these states.

This makes the information ideal for a custom field.

Quote Stage = Quote Sent

Later...

Quote Stage = Booked

The previous state disappears because it is no longer true.

The database always reflects reality.

Why Tags Become Increasingly Difficult to Manage

Many older HighLevel systems represent every stage using a separate tag.

For example:

• Quote Sent

• Awaiting Follow-up

• Booked

• Completed

Initially this appears simple.

However, over time the database begins accumulating historical labels.

A contact may eventually contain:

• Quote Sent

• Awaiting Follow-up

• Booked

• Completed

Which one represents the customer's current status?

The answer depends entirely on whether every workflow remembered to remove every previous tag.

As systems become larger, this becomes increasingly difficult to maintain.

Eventually the tags stop representing reality and instead become a historical record of everything that has ever happened.

Fields Represent Current Truth

Fields are fundamentally different.

A field contains one value.

That value represents the current state.

Examples include:

• Quote Stage

• Appointment Status

• Payment Status

• Customer Type

• Property Type

• Cleaning Frequency

• Last Appointment Date

• Account Manager

Each field answers a simple question:

What is true right now?

This makes reporting, filtering and automation dramatically simpler.

Tags Represent History and Classification

Tags still have an important role.

They simply serve a different purpose.

Rather than representing current status, they are better suited for recording historical events or classifications.

Examples include:

• Facebook Lead

• Google Ads

• VIP Customer

• Newsletter Subscriber

• Commercial Property

• Existing Customer

• Requested Callback

Unlike stages, these characteristics can all exist simultaneously.

A customer can legitimately be:

• Facebook Lead

• Newsletter Subscriber

• VIP Customer

There is no conflict because none of these attempt to describe the customer's current position in a workflow.

HighLevel's Direction Supports Field-Based Architecture

Recent improvements throughout HighLevel strongly suggest that the platform itself is moving towards a more traditional CRM architecture.

Examples include:

• More powerful custom fields

• Better opportunity management

• Enhanced workflow triggers based on field changes

• Improved reporting

• AI actions that read and write structured data

• Relationship-based automations

• More advanced pipelines

• Better dashboard capabilities

These features all rely on structured information.

AI, reporting and automation perform far better when reading a single field than attempting to interpret multiple tags.

Fields Improve Automation Design

Consider a quotation workflow.

Instead of adding and removing tags throughout dozens of workflows:

Add Tag
Quote Sent

A more robust design is:

Update Field Quote Stage = Quote Sent

Every downstream workflow simply watches for one event:

Quote Stage changes

This creates a single, authoritative workflow architecture.

Instead of dozens of tags driving the system, one field becomes the central controller.

Reporting Becomes Much More Reliable

Imagine asking:

How many quotations are currently awaiting follow-up?

With fields:

Quote Stage = Awaiting Follow-up

The answer is immediate.

With tags, you may discover contacts containing:

• Quote Sent

• Follow-up

• Booked

• Completed

without older tags ever being removed.

The report becomes unreliable.

Structured fields eliminate this ambiguity.

AI Benefits from Structured Data

Artificial Intelligence performs best when information is structured.

An AI assistant can easily understand:

Quote Stage = Address Received

or

Appointment Status = Confirmed

It is much harder for AI to infer meaning from a collection of historical tags.

As HighLevel continues investing heavily in AI, structured fields will become increasingly valuable.

A Modern HighLevel Architecture

For most businesses, an effective design follows a simple principle.

Use Fields For

• Pipeline stages

• Sales stages

• Appointment status

• Payment status

• Customer type

• Last activity

• Dates

• Numbers

• Structured business information

Use Tags For

• Lead source

• Marketing segmentation

• Interests

• Historical milestones

• Campaign membership

• Temporary workflow flags

• Automation markers

The Future of HighLevel

Tags are not disappearing.

They remain an excellent solution for segmentation, marketing and behavioural classification.

However, they are no longer the backbone of a well-designed HighLevel CRM.

As HighLevel continues evolving into a fully featured relationship database, fields increasingly become the system's single source of truth.

From a systems architecture perspective, this leads to cleaner workflows, more reliable reporting, better AI integrations and databases that are considerably easier to maintain.

The most scalable HighLevel implementations are no longer built around collections of tags.

They are built around structured data.