In partnership with

Need a personal assistant? We do too, that’s why we use AI.

Ready to embrace a new era of task delegation?

HubSpot’s highly anticipated AI Task Delegation Playbook is your key to supercharging your productivity and saving precious time.

Learn how to integrate AI into your own processes, allowing you to optimize your time and resources, while maximizing your output with ease.

Get the free guide here.

Consistency Without Compromise

A few weeks ago, we talked about the power of ontologies. This week, I want to continue down that path and explore how well-defined ontologies can be used to handle messy datasets—especially in industries like AEC, where consistent data is often elusive.

“Clients all want different things” is a refrain I hear over and over again—and it’s true. I can’t refute that. But letting that get in the way of consistent data is a choice—and one that harms any firm hoping for a future where intelligent machines can help them handle the boring, time-consuming aspects of their work. The fact is, inconsistent data is kryptonite for AI systems and data-driven tools writ large.

So what can we do about it?

We can leverage ontological classifications to clean our data downstream from production and make it understandable to data-driven systems. Let’s say we’ve worked hard to develop an ontology for rooms in hotel floor plans. Maybe something like this:

With this ontology in hand, we can train classification models that take information about hotel rooms and classify each into one of these categories. For example:

• “The Washington Room” → Suite

• “The Grand Ballroom” → Dining Area

• “4th Floor Elevators” → Elevator Lobby

This enables teams to maintain naming flexibility for their clients while enabling downstream labeling consistency.

What can you do with standardized data?

1. Automate Reporting: With consistent classifications, generating automated reports for clients becomes much easier. Instead of manually aggregating inconsistent labels, standardized data can feed directly into reporting pipelines.

2. Train More Accurate Predictive Models: Clean data allows for better training of predictive models, whether for occupancy forecasting, energy consumption optimization, or maintenance scheduling.

3. Improve Collaboration Across Teams: Standardized ontologies provide a common language for different teams—designers, engineers, and data scientists—making cross-disciplinary collaboration more effective.

4. Enable Interoperability: In a world where firms increasingly rely on multiple software tools, standardized data makes it easier to integrate those tools. Consistent classifications reduce friction when transferring data between BIM, CAFM, and other systems.

5. Support Advanced Analytics: With well-structured data, firms can perform advanced analyses such as clustering similar rooms to find cost-saving design patterns or identifying underutilized spaces.

Consistency without Compromise

Achieving consistency in data does not mean sacrificing flexibility or creativity—it means creating a framework that supports both. By leveraging ontologies, firms can keep their client-specific naming conventions while ensuring data is standardized downstream. This balance allows businesses to meet unique client demands while still enabling the development of scalable, data-driven solutions.

In an industry like AEC, where projects are complex and variability is high, the ability to maintain both flexibility and consistency is key. Firms that master this balance will be better positioned to leverage AI, automate routine tasks, and unlock the full potential of their data—without compromising on what makes each project unique.

Join the Community!

Thanks for making it all the way through this weeks Not Magic, Just Math. It means a lot to me that you made it all the way here. I’d bet that you’d be a great addition to our little community here and might get value out of subscribing! Please think about doing so - and stay current on all the exciting ML and AI applications we discuss in and around the AEC Industry! Thanks!