Medsender is an AI automation platform for healthcare front offices, built around four published products: referral management, fax and document intake, patient recapture for lapsed patients, and call automation through a voice agent called MAIRA. It reads inbound faxes and referrals, extracts and structures the data, matches or creates the patient chart, and then runs outreach by text, email and phone until the patient books. It integrates with several EHRs including athenaOne, ModMed EMA, Elation and AdvancedMD, and says it is trusted by more than 1,000 independent practices ranging from single locations to groups of 250 or more.

Is Medsender just a fax tool?

No, and treating it as one will lose you the evaluation. Medsender started in documents and its marketplace listing still sits in the records category, but the current product line covers referral management, patient recapture and a 24/7 voice agent that books appointments and routes refill requests. If you are comparing vendors on the basis that this one only moves faxes around, you are comparing against a product that no longer exists.

We are a specialty practice buried in referral faxes. How do we choose?

Both products do this job, so run the comparison on the referral that goes wrong rather than the one that goes right. Ask each vendor what happens to a referral that arrives missing the imaging or the authorization, who chases it and on which channel, how many attempts it makes before it is given up on, and which objects change in athenaOne when the visit is finally booked. Then ask what happens to the patient who calls your main line that afternoon about something unrelated. That second question is where the two products separate, because one of us is scoped to the intake lane and the other runs the whole front office on athenaOne.

Medsender works with several EHRs. Is athenaOne-only a limitation?

It is the whole point. Building across EHRs means building to what those systems have in common, because a workflow that only works in one of them is a support burden everywhere else. We hold production access to 820+ athenaOne APIs and build against the parts of athenaOne that have no equivalent elsewhere: your appointment types and scheduling templates, provider groups and departments, the order and document queues, eligibility checks and the patient portal. If you are running more than one EHR across your group, that is a fair reason to look at a multi-EHR vendor. If you are on athenaOne and staying, depth in that one system is what decides how much of the work actually finishes.

Yes, and some groups will want to while they are mid-contract. Settle two things in writing before go-live. Which system owns patient outreach for a referral that needs scheduling, because two agents texting the same patient about the same appointment is the failure mode that makes staff turn both off. And which system is the source of truth for referral status, so the queue a manager opens on Monday morning has one answer rather than two.

How do we compare the results each vendor publishes?

Read them as shapes rather than as numbers you will inherit. Medsender publishes 85% faster referral turnaround, 2 to 3 times more referrals converted to visits and 40 or more admin hours saved per location per month. We publish that the AI handles about 70% of more than 100,000 calls a month start to finish at our largest deployment. Neither figure transfers to your practice unchanged. Customer-reported results. Your numbers will vary by workflow, staffing, seasonality and call mix. The number worth asking both vendors for is the same one either way: of the referrals that arrived last month, how many reached a completed first visit, and how long did they wait.

← All comparisons

Pretty Good AI vs Medsender

Pretty Good AI vs Medsender for athenaOne Fax and Referrals

Medsender automates fax intake and referrals across several EHRs. Pretty Good AI builds only for athenaOne. What each one finishes inside the athena chart.

Last reviewed September 23, 2026 8 min read

The short answer

Medsender is an AI automation platform for front-office intake, published around four products: referral management, fax and document intake, patient recapture, and call automation through a voice agent it calls MAIRA. It integrates with several EHRs including athenaOne, ModMed EMA, Elation and AdvancedMD, says it is trusted by more than 1,000 independent practices from 1 location to 250 or more, and publishes 85% faster referral turnaround, 2 to 3 times more referrals converted to visits, and 40 or more admin hours saved per location per month. Its athenahealth Marketplace listing went live in April 2026 and shows a 0 to 10 practice connection range as of the 14 September 2026 sync. Pretty Good AI is the AI operations platform built exclusively for athenaOne practices, on production access to 820+ athenaOne APIs. Both automate inbound referrals and answer the phone. The difference is scope: Medsender is built to solve intake across whichever EHR a practice runs, and we are built to run the whole front office inside one of them.

Pretty Good AI vs Medsender, side by side

Feature-by-feature comparison of Pretty Good AI and Medsender
Dimension	Pretty Good AI	Medsender
What the product is built around	One system, all the way down. Referrals, inbound and outbound calls, secure two-way text, web scheduling, self-check-in, forms, eligibility, prior-auth preparation, balances and recall all run on the same athenaOne integration, so a request that starts in one place finishes in the chart.	Intake, across systems. Four published products: referral management, fax and document intake, patient recapture, and call automation. Medsender describes itself as handling faxes, referrals, calls and lapsed patients so nothing waits. ^[1]
EHR scope	athenaOne only, deliberately. We build to the full capability of one system rather than the features every EHR has in common.	Several EHRs published, including athenaOne, ModMed EMA, Elation and AdvancedMD, positioned as working with the EHR a practice already uses. ^[1]
Integration depth	Production read and write access to 820+ athenaOne APIs, no middleware, and our app runs inside athenaOne with no extra login. Most marketplace vendors pick a narrower wedge and build deep inside it; we carry significant depth across the whole of athenaOne.	Medsender describes athenaOne as its customer's system of record and itself as the layer in front of it, capturing faxes and documents, structuring referral data and writing it back so records stay current. ^[2]
Inbound referral intake	Fax, provider web form, patient self-referral form and campaign landing pages land in one intake and review workflow. Referral faxes are charted with a confidence score, patient matching and deduplication, and a hold-for-review step before anything is written. After staff review, the agent calls the patient to schedule.	The core product. Referrals are read, labeled, sorted and routed automatically, checked for duplicates, matched or created as a chart, and then worked by outreach that goes text first and email second, with exceptions flagged for a referral coordinator. ^[3]
Reaching the referred patient	Voice and secure two-way text on the same athenaOne integration. The referral that has sat nine days gets a phone call, and the person who answers can book the visit on that call against live slots and the practice's own appointment types.	Text first with a self-scheduling link, then email, then retry, run by agents that keep going until the patient books or a person needs to step in. Medsender also places outbound calls as part of referral management and recapture. ^{[3] , [4]}
Answering the main line	Answered in under five seconds, day and night, and resolved on the call: booking, rescheduling, refill requests routed to the right queue, eligibility, balances and payment by secure link, written into athenaOne while the caller is still on the phone.	MAIRA picks up on the first ring, 24/7, answers common questions such as hours, directions and pricing, captures refill requests and routes them with a summary, and transfers complex calls to the team with full context. Medsender publishes 26 languages. ^[4]
Languages	We run the full administrative call in 10 languages, English, Spanish, French, German, Hindi, Russian, Portuguese, Japanese, Italian and Dutch, with the same workflows in each.	Published at 26 languages, day and night. ^[4]
Everything that is not intake	The same platform works balances and takes payment, prepares prior authorizations by payer and pulls status back into the chart, fills canceled slots, runs confirmation and recall calls, and runs post-visit surveys and review requests.	Outside the four published products. Medsender markets itself as automating the work a growing practice has outgrown at the front of the workflow, rather than as a general front-office or revenue platform. ^[1]
athenahealth Marketplace record	45 practice connections, 22 ratings, all five-star. athenahealth Marketplace partner since 2025. As of October 2026.	Listed since April 2026, with a published connection range of 0 to 10 and no ratings, as of the 14 September 2026 sync. Listed under Medical Records and Documentation, Clinical Workflow Acceleration, and Marketing and Patient Communications. ^{[5] , [6]}
Published results	Commonwealth Pain & Spine, a 35-location specialty group with 130 providers, runs on Pretty Good AI. We handle more than 100,000 patient calls a month for them. The AI handles about 70% from start to finish. One in eight bookings now comes in after hours, when the office used to be closed. Customer-reported results. Your numbers will vary by workflow, staffing, seasonality and call mix.	85% faster referral turnaround, 2 to 3 times more referrals converted to visits, and 40 or more admin hours saved per location per month, with named customer quotes including one region moving referral conversion from 20% to 62%. ^{[2] , [1]}
Published pricing	No setup fees. First 30 days live free. Month to month after. No annual lock-in. Priced by usage, providers and locations, and new workflows are priced when you turn them on.	No published price. Medsender routes pricing through a 30-minute demo request. ^[1]

Two products that overlap on the loudest problem

A referral arrives by fax. Nobody sees it for two days. By the time a coordinator calls, the patient has been seen somewhere else. Every specialty practice on athenaOne knows this story, and it is the problem both Medsender and Pretty Good AI are pointed at.

Where they stop being the same product is everything either side of that fax. Medsender is built to solve intake wherever a practice keeps its chart, across athenaOne, ModMed EMA, Elation and AdvancedMD. We are built to run the front office of one system, and we only do athenaOne.

That single design choice explains almost every difference below, so it is worth deciding which shape you want before you look at a feature grid.

What Medsender does well, stated plainly

Its referral workflow is genuinely agentic and it is specific about the steps. The referral is read and labeled as it arrives, checked for duplicates, matched to an existing chart or given a new one, and then worked by outreach that starts with a text carrying a self-scheduling link, follows with an email if nothing comes back, and retries after that. Anything unusual is flagged for a human coordinator rather than guessed at.

The voice agent is real too, not a roadmap slide. MAIRA answers on the first ring around the clock, handles hours, directions and pricing questions, captures refill requests with a summary, transfers the complicated calls with context, and is published in 26 languages.

If your entire problem is inbound documents and the outreach that should follow them, this is a serious product and you should look at it.

Where the comparison actually turns

Ask what happens to the work that is not intake.

A referral coordinator is rarely just a referral coordinator. The same person is answering the main line, telling a patient what they owe, rebooking the Thursday cancellation, finding out why an eligibility check failed last week, and re-sending a form nobody completed. Automating the referral lane moves one queue. It does not change what that person’s afternoon looks like.

Depth in one EHR is what lets the rest of it move. Because we hold production access to 820+ athenaOne APIs, a call can finish rather than become a task: the slot is checked against your real appointment types and scheduling templates, the refill goes to the right order queue, the eligibility check runs, the balance and payment link are handled, and the result is written into the chart before the caller hangs up. That list is athenaOne-shaped on purpose. It is the part a platform built to work across four EHRs cannot build to, because three of them have nothing to build against.

We adapt to your practice. Your practice does not adapt to us.

Production access to 820+ athenaOne APIs is what makes an unusual workflow buildable at all. Most marketplace vendors pick a narrower wedge and build deep inside it; we carry significant depth across the whole of athenaOne.

We map the workflow your staff already run and automate that. You do not re-cut your scheduling, intake or queue conventions to match a product’s opinion of how the work should go. New workflow scope is measured in days and weeks, not quarters, inside an onboarding window that already runs weeks.

Tell us how your front office runs today and we will tell you what we would build against it.

An example worth taking to both demos

A multi-site orthopedic group takes referrals at five locations, each with its own fax number and its own front desk running its own queue. Two of those locations share a coordinator who covers one of them only on Tuesdays and Thursdays.

The interesting referral is not the clean one. It is the fax that arrives at 4:50pm on a Friday for a workers’ compensation case, with the adjuster’s authorization missing and a referring provider whose name is misspelled on the cover sheet. Automation can do most of it. It can read the fax, identify the document type, match the patient on date of birth and insurance, create the chart, and start texting the patient a self-scheduling link.

What it should not do is guess. The missing authorization means the visit is not billable yet, and the misspelled referring provider means the loop will close back to the wrong practice if nobody checks. In our version that referral is written with a confidence score and held for review rather than pushed through, the coordinator sees it in one queue with its age on the face of it, and the agent starts calling the patient only once a person has confirmed the case is clean. On the Tuesday the coordinator is not in, the same queue is visible to whoever is covering, and the referrals that have aged past the practice’s own threshold are at the top of it.

Ask both vendors to walk that referral, not the clean one. The demo script never includes the Friday fax.

Five questions that settle it

What happens to a referral that arrives missing the authorization, and who decides it is safe to proceed?
Which athenaOne objects change when your software finishes, and can you show them in the chart rather than in your own dashboard?
When the referred patient calls back two days later about something else entirely, who answers?
How many attempts does outreach make, on which channels, before a human is told it failed?
What does a manager open on Monday morning to see every referral that has stalled, and how old is the oldest one on it?

Take the answers from both vendors side by side. The first two separate the products on depth, the third separates them on scope, and the last two are how you find out whether anything is actually being measured.

Where we are not a fit

We build on athenaOne and nothing else. If your group runs more than one EHR, or you are mid-migration, or a second system is staying for a specialty division, part of your referral problem lives somewhere we do not go. A multi-EHR intake vendor covers that ground and we do not pretend to.

Clinical decisions stay with your care team. We run the administrative work around them, and we say so on every page.

The shortest version

Medsender solves intake in whichever EHR you run. We run the front office of the one you already chose. If athenaOne is where your practice is staying, ask both vendors the same five questions above, and weigh the answers on how much of the work finishes without landing back on someone’s desk.

The Pretty Good AI platform has attained HITRUST i1 certification. Pretty Good AI signs a HIPAA Business Associate Agreement (BAA) before any patient data is touched. SOC 2 Type II and ISO 27001 audit reports are available upon request.

Frequently asked questions

What is Medsender?: Medsender is an AI automation platform for healthcare front offices, built around four published products: referral management, fax and document intake, patient recapture for lapsed patients, and call automation through a voice agent called MAIRA. It reads inbound faxes and referrals, extracts and structures the data, matches or creates the patient chart, and then runs outreach by text, email and phone until the patient books. It integrates with several EHRs including athenaOne, ModMed EMA, Elation and AdvancedMD, and says it is trusted by more than 1,000 independent practices ranging from single locations to groups of 250 or more. ^{[1] , [3]}
Is Medsender just a fax tool?: No, and treating it as one will lose you the evaluation. Medsender started in documents and its marketplace listing still sits in the records category, but the current product line covers referral management, patient recapture and a 24/7 voice agent that books appointments and routes refill requests. If you are comparing vendors on the basis that this one only moves faxes around, you are comparing against a product that no longer exists. ^{[1] , [4]}
We are a specialty practice buried in referral faxes. How do we choose?: Both products do this job, so run the comparison on the referral that goes wrong rather than the one that goes right. Ask each vendor what happens to a referral that arrives missing the imaging or the authorization, who chases it and on which channel, how many attempts it makes before it is given up on, and which objects change in athenaOne when the visit is finally booked. Then ask what happens to the patient who calls your main line that afternoon about something unrelated. That second question is where the two products separate, because one of us is scoped to the intake lane and the other runs the whole front office on athenaOne. ^[3]
Medsender works with several EHRs. Is athenaOne-only a limitation?: It is the whole point. Building across EHRs means building to what those systems have in common, because a workflow that only works in one of them is a support burden everywhere else. We hold production access to 820+ athenaOne APIs and build against the parts of athenaOne that have no equivalent elsewhere: your appointment types and scheduling templates, provider groups and departments, the order and document queues, eligibility checks and the patient portal. If you are running more than one EHR across your group, that is a fair reason to look at a multi-EHR vendor. If you are on athenaOne and staying, depth in that one system is what decides how much of the work actually finishes. ^[1]
Can we run both?: Yes, and some groups will want to while they are mid-contract. Settle two things in writing before go-live. Which system owns patient outreach for a referral that needs scheduling, because two agents texting the same patient about the same appointment is the failure mode that makes staff turn both off. And which system is the source of truth for referral status, so the queue a manager opens on Monday morning has one answer rather than two.
How do we compare the results each vendor publishes?: Read them as shapes rather than as numbers you will inherit. Medsender publishes 85% faster referral turnaround, 2 to 3 times more referrals converted to visits and 40 or more admin hours saved per location per month. We publish that the AI handles about 70% of more than 100,000 calls a month start to finish at our largest deployment. Neither figure transfers to your practice unchanged. Customer-reported results. Your numbers will vary by workflow, staffing, seasonality and call mix. The number worth asking both vendors for is the same one either way: of the referrals that arrived last month, how many reached a completed first visit, and how long did they wait. ^[2]

Sources

Everything stated here about another vendor comes from that vendor's own public material or from the athenahealth Marketplace listing, on the date shown. Vendors change their products and their pricing; if something below is out of date, email contact@prettygoodai.com and we will correct it.

Medsender, AI Fax, Referral and Call Automation (accessed 2026-09-20)
Medsender, athenaOne Fax and Referral Automation (accessed 2026-09-20)
Medsender, Referral Management Software for Practices (accessed 2026-09-20)
Medsender, Medical Call Automation (accessed 2026-09-20)
Medsender, athenahealth Marketplace listing (accessed 2026-09-14)
Pretty Good AI, athenahealth Marketplace listing (accessed 2026-09-14)

Why athenaOne practices evaluate Pretty Good AI

We are your AI team for athenaOne.

Your practice has its own providers, processes, and rules. Pretty Good AI works with your team to fit the first workflow to the work you want to improve, agree on how to measure it, and keep improving it as your practice changes.

Live and measured

100,000+: patient calls a month, at more than one customer
About 60%: of those calls handled start to finish by the AI at the largest deployments
Hundreds: of providers inside a single group
20: specialties on our athenahealth Marketplace listing

Customer-reported results. Your numbers will vary by workflow, staffing, seasonality and call mix.

How the platform works Pricing Customer results Security and compliance

Map your first workflow.

See it against your own athenaOne data

The honest way to compare is on your own call volume, your own schedule and your own payer mix. Book a working session and we will walk your numbers.

Book a demo How our pricing works