5 Things We Learned Building AI Inside an Inc. 5000 DSO

Over the last year, we have been building and deploying AI inside Shared Practices Group, an Inc. 5000 DSO.

That experience has taught us a lot about what actually matters when bringing AI into a fast-moving dental organization.

The DSO world is competitive. Speed matters. Efficiency matters. Small operational improvements can compound quickly across dozens of practices.

It is easy to talk about AI at a high level. It is much harder to make it useful inside a real organization with real employees, real workflows, and real data.

Here are five of the biggest lessons we have learned.

1. Great UX Drives Adoption

One of the most consistent complaints we heard while researching the dental technology market was that many existing tools felt clunky.

There was plenty of data available, but accessing and understanding it often required too much work.

That became one of the principles behind Root Data.

Quality and usability had to come first.

If an operations team needs to click through multiple reports, dig through complicated dashboards, or spend too much time figuring out where to look, the software becomes another task rather than something that makes their job easier.

We focused on making Root simple to navigate, fast to use, and easy to understand.

The result was significant.

Inside SPG, teams began using Root roughly 10 times more than other analytics tools they had access to.

That reinforced an important lesson. Adoption is not always a training problem.

Sometimes the product simply needs to be better.

If software makes someone's job easier, they are much more likely to use it.

2. Faster Data Creates a Faster Organization

Data is only useful if you can access it quickly enough to act on it.

DSOs are constantly moving.

Production changes. Appointments change. Leads come in. Patients cancel. Teams outperform or underperform. Revenue leaks can appear quickly.

If you have to wait until the next day to understand what happened, you are effectively managing the organization while looking in the rearview mirror.

We wanted Root Data to operate differently.

Instead of relying on next-day updates, Root syncs data regularly throughout the day.

That gives operators a more current view of what is happening across the organization.

The faster you can spot an issue, the faster you can investigate it.

The faster you investigate it, the faster you can fix it.

Across a large DSO, that time advantage compounds.

It also improves accountability because teams can see what is happening while there is still time to respond.

3. Edge Cases Are the Product

We initially set out to rebuild dental analytics from the ground up using AI.

Once Root was deployed inside SPG, something interesting happened.

Different teams started asking for more.

Can you pull this data?

Can you show this differently?

Can you add this feature?

Can you help us understand what is happening here?

Suddenly, product development became much more organic.

Operations had different needs than sales. Insurance teams had different needs than leadership. Different practices had different workflows and different problems.

These edge cases were not distractions from the product.

They became part of the product.

Many of the most valuable opportunities inside a DSO are the annoying operational problems that require too much human coordination, too many spreadsheets, or too much manual work.

Solving one may seem small.

Solving dozens of them across an organization can create meaningful operating leverage.

4. AI Should Reduce Decisions, Not Create More Dashboards

Dental practices already produce a massive amount of data.

The problem is not that operators need more information.

The problem is figuring out what matters.

Historically, understanding practice performance often meant opening multiple reports, comparing metrics, matching data between systems, and trying to identify patterns manually.

AI changes that.

Instead of presenting another dashboard full of numbers, AI can help surface the signal inside the noise.

That is where we believe the real opportunity exists.

The goal should be to take thousands of data points and turn them into something simple enough for an operator to act on.

Which practice needs attention?

What changed?

Where is revenue leaking?

What should the team focus on next?

The best AI tools will not give operators more things to analyze.

They will reduce the amount of analysis required before taking action.

5. Get the Data Right Before You Build the AI

This may be the most important lesson of all.

It is relatively easy today to open ChatGPT or Claude and start experimenting with AI.

That does not mean you have built an enterprise AI system.

Before AI becomes useful inside a DSO, the underlying data needs to be pulled correctly, structured correctly, synced reliably, and understood in context.

Dental practices produce enormous amounts of data.

There are likely opportunities hidden inside that data that operators are not even aware of yet.

But AI can only surface those opportunities if the foundation is solid.

Once the data infrastructure is reliable, the possibilities expand dramatically.

You can build analytics tools, AI insights, operational alerts, workflow automation, forecasting, and products that solve problems you may not have originally anticipated.

That is ultimately the biggest takeaway from our experience building Root Data inside SPG.

Adding AI is not the hardest part.

The hard part is getting the data right, understanding how the organization actually operates, uncovering the edge cases, and building useful products around them.

Once you solve that foundation, AI becomes much more powerful.

If you're ready to explore how AI can transform your DSO, reach out to hello@rootdata.ai.