Signature Dental Partners automated clean claims first, reports denials down 50%+
In late 2024, the VP of revenue cycle at Signature Dental Partners, a DSO with about 100 practices, said she talked to about five AI vendors and then automated one workflow: clean claim submission, with Thoughtful AI. She reported denials down by more than half and claim turnaround cut from 45 days to 17. These are early, self-reported figures. They come from a session co-presented by the vendor and written up 'in collaboration with' it.
Published 20 October 2026 · Last checked 20 October 2026 · Fact-checked against 9 sources · How we verify
- Practice
- Multi-site DSO
- Size
- ~100 practices (2024)
- Location
- Phoenix, AZ (HQ)
- Tool
- Thoughtful AI claims automation
- Cost
- Not disclosed
- Source
- Mixed
Did it work? Results
- Perry said denial rates fell by more than 50% 'so far' after the group started working with Thoughtful AI. She presented this in late 2024 as progress, not a final result.
Source - Perry said claim turnaround dropped from 45 days to 17. The article does not say how turnaround was measured (for example, from submission to payment).
Source - The first automated workflow was clean claim submission. Perry said she chose it because it was the group's biggest obstacle, and she got her executive team to approve that one process first.
Source - In May 2024, the show notes for a podcast episode with Thoughtful AI's CEO described Signature as a success story of fewer days sales outstanding and growth without extra hiring. This is a promotional summary and gives no numbers.
Source
Who tried it, and why
Dental support organization (DSO) with a centralized revenue cycle management (RCM) team. About 100 practices in late 2024, aiming for 200. A 2026 case study from a different vendor (Overjet) puts it at 100+ offices and 150+ dentists.. Headquartered in Phoenix, AZ, with affiliated dentists in the Mountain and Southwest US (per a private-equity directory listing).
Cara Perry, Vice President of Revenue Cycle Management at Signature Dental Partners, has 15 years of experience in DSOs and RCM. She had centralized the group's billing functions into one RCM team that she built from the ground up. The business goal was to grow from about 100 practices to about 200 without her team growing at the same rate. Her biggest problem was claim quality: claims were leaving with errors and getting paid late. The panel also named industry-wide pressures, including more claims, harder payer relationships and high staff turnover on RCM teams.
What did they set up?
- Centralized the revenue cycle first. Perry had built one RCM team for the group before bringing in AI.
- Tied the project to a business goal: double the number of practices without adding billing staff at the same pace.
- Interviewed about five AI companies before going with Thoughtful AI, which sells AI agents for claims processing, eligibility, prior authorizations and payment posting.
- Asked the executive team to approve one process only: clean claim submission, which she called the group's biggest obstacle. She did not try to automate the whole revenue cycle at once.
- The vendor's method, as its co-founder Dan Parsons described it, starts with the denial data. You find which denial reasons and which payers cause most of the denials and fix that slice first. His example was a practice where 30% of denials come from eligibility or prior authorization, mostly from five payers. This describes the vendor's general approach, not Signature's own data.
- The collaboration-labelled recap calls the claims project a foundation for adding more AI workflows later. It reports no results from later phases.
- Timeline: show notes for a podcast episode with Thoughtful AI's CEO, published May 31, 2024, already called Signature a success story. Perry and Parsons presented at Becker's Fall Future of Dentistry Roundtable on Oct 31, 2024. Becker's published its recap on Nov 19, 2024.
Tools used
- Thoughtful AI: AI agents and automation for revenue cycle tasks (claims, eligibility, prior authorizations, payment posting). Combined into Smarter Technologies in a deal announced in May 2025.
What did it cost?
Not disclosed. No source we fetched gives the price, contract structure or setup fees. Two numbers in the same article come from the vendor's co-founder, not from Signature: working one denial can cost up to $118 per claim, and 9 out of 10 denials are preventable. A $118-per-claim appeal cost was also reported in 2017, from a Change Healthcare report based on 2016 hospital claims. It is an old benchmark and not specific to dental.
“Claims were going out ‘dirty,’ and they weren’t getting paid on time.”
What didn't go well?
- No problems were disclosed. A Thoughtful AI co-founder co-presented the session, and the write-up is labelled 'In collaboration with Thoughtful AI'. Setbacks rarely come up in that setting, which does not mean there were none.
- There is no baseline. The recap doesn't give Signature's starting denial rate or say how the 45-day and 17-day turnaround figures were measured, so you can't compare them directly with your own numbers.
- These are early results from 2024. We found no later follow-up from the practice that we could open and verify.
- The vendor has changed since then. In May 2025, New Mountain Capital announced it was combining Thoughtful AI with SmarterDx and Access Healthcare to form Smarter Technologies. Today thoughtful.ai redirects to Access Healthcare, an RCM services company with a dental RCM page. A practice evaluating this tool now would be dealing with a different, larger company than the one Signature hired.
- The two headline statistics in the article, the $118 cost per denial and 9 out of 10 denials being preventable, came from the vendor's co-founder, not from Signature's own data.
What should your office take from this?
- Start with your single biggest leak. For Signature that was claims going out with errors. Perry automated that one workflow first and treated it as a base to build on.
- Talk to several vendors. Perry interviewed about five AI companies before choosing one.
- Before you start, write down your baseline: denial rate, top denial reasons by payer, and days from submission to payment. Without it you can't check a result like 45 days to 17.
- Treat vendor statistics as sales claims. The $118 cost per denial and the 9-in-10-preventable figure came from the vendor, and the $118 matches a 2017 hospital figure, not dental data.
- Signature had a central billing team and about 100 offices, which gave it a lot of volume to work with. A single office can still use the same method: pull 90 days of denials, sort them by reason and payer, and fix the top one.
- Ask any AI claims vendor how much of the work is software and how much is people, and which company you are actually signing with. Thoughtful AI is now part of Smarter Technologies, and its website redirects to an RCM services firm.
Try it yourself
I run billing for a dental practice with [number] locations. Below is a de-identified export of our denied and rejected claims from the last 90 days. It has no patient names, dates of birth or member IDs. The columns are: payer, CDT code, denial or rejection reason, billed amount, and days to resolution. 1. Group the denials by reason and by payer. Show the 3 reason + payer combinations that cost us the most in dollars and in rework. 2. For each of those 3, tell me whether it looks preventable before submission (missing attachment or narrative, eligibility, frequency limit, coding or bundling) or only fixable after the denial. 3. Recommend ONE workflow to fix or automate first, and explain why in plain language. 4. Write a short pre-submission checklist my team can start using this week for that workflow. 5. List 5 questions I should ask any AI claims vendor. Include how they will measure our baseline denial rate and days to payment before go-live, and how much of the work is software versus people. If the data is too thin to support a conclusion, say so instead of guessing. [paste de-identified data here]
Prompt written by us, not by the practice. Never paste names, dates of birth or other patient details into a general AI tool.
Sources
- How AI drives quick wins in RCM: 4 takeaways for DSOs
- How AI drives quick wins in RCM: 4 takeaways for DSOs (archived copy used for reading)
- Session: DSOs Winning with AI: Real-World Success Stories in the Revenue Cycle (Fall Future of Dentistry Roundtable)
- New Mountain combines portfolio companies to launch AI revenue cycle firm
- Revolutionizing Healthcare Management with Advanced AI Technologies (podcast episode with Thoughtful AI CEO Alex Zikoff; show notes)
- Revenue cycle management built for dental practices (thoughtful.ai now redirects to this company's site)
- How Signature Dental Partners Achieved 13X ROI with Overjet AI at Costa Verde Dental (used only for DSO size)
- Signature Dental Partners: portfolio company profile (used only for location)
- News: Claims appeals cost hospitals up to $8.6 billion annually (Change Healthcare report, ~$118 per claim)
How we checked this
Checked on 20 October 2026. We read all 9 sources listed above, and a second, independent check re-opened every one of them before publication. Anything we could not confirm was removed.
Figures published by a vendor are reported as that vendor's claims, not as verified facts.
Spotted an error? Tell us and we will re-check it. How we verify.
Get one real case study a week
One email a week. We measure opens and clicks. Unsubscribe anytime. Privacy notice.
More case studies
AI insurance verification, fifth tool tried: Overjet reports 15+ hrs/week saved
A three-dentist family practice in South Carolina had tried four verification tools, most of which returned only about half the benefit details it needed. Overjet reports that its AI now returns 85% or more of those details and saves the practice 15+ hours a week.
Dentist drops DentalXChange Eligibility AI within 3 months over crown error
A general dentist cancelled DentalXChange's Eligibility AI within about three months. It was accurate on easy checks and saved maybe a minute or two of front-desk time, but it got harder checks wrong, including a crown it showed as covered at 50% when coverage was 0%. The whole report is one short Reddit post and one reply, and the vendor's own claims come with no independent figures.
AI receptionist for overflow and after-hours calls: Wolfe Dental's results
A two-office practice in the Portland area with one receptionist per office let Arini's AI take overflow and after-hours calls. Arini reports missed calls fell from as many as 500 to about 50 a month and $134k in booked production over eight months, but it gives no price.