TL;DR: The ROI of referral intake automation for a multi-site oncology group comes from two sources: recovered new-patient revenue from fewer leaked referrals and faster consults, and labor savings from removing manual intake work. Because each converted oncology patient typically carries far more downstream revenue than a standard specialty visit, even a small drop in referral leakage often covers the cost of an oncology referral intake automation tool within months, before counting coordinator time saved.
If you're a CFO or COO at a multi-site oncology group, you've probably heard the pitch for referral intake automation already. The harder part is building a business case your board, partners, or PE sponsor will believe.
This article lays out a practical ROI model: the inputs you need, a worked example with clearly labeled assumptions, the costs vendors tend to leave out, and how to measure payback after go-live. Swap in your own numbers; the structure is what matters.
Where does the ROI come from?
Referral intake automation creates value in two places, and it's worth modeling them separately because they behave differently.
1. Revenue recovery. This is usually the larger number. It comes from converting more referrals into first consults, which happens when referrals are processed faster, missing records are chased immediately, and patients are contacted before they book elsewhere. In oncology, a new patient often leads to a treatment episode that can include infusion, radiation, imaging, labs, and follow-up visits over months.
2. Labor efficiency. Intake coordinators spend a large share of their day opening faxes, keying data into the EHR, sorting documents, and chasing records. Automation removes much of that work. The savings show up either as reduced overtime and agency staff, or more often as the ability to absorb volume growth and new sites without adding headcount.
There's also a third category that's harder to put a dollar figure on: fewer rescheduled consults, better referring-provider relationships, and less coordinator burnout. Mention these in your business case, but don't lean on them for the numbers.
What inputs does the ROI model need?
Pull these from your own data before you build the model. Estimates are fine to start, but label them clearly.
- Monthly new-patient referral volume across all sites
- Current referral-to-consult conversion rate (or its inverse, leakage rate)
- Average referral-to-consult time, broken into administrative time and physician availability time
- Coordinator minutes per referral for intake work: opening, reading, keying, filing, chasing records
- Loaded labor cost per coordinator hour (wages plus benefits and overhead)
- Average downstream contribution margin per new oncology patient over the first year, from your finance team
- Current rate of consults rescheduled due to missing records
The downstream margin number is the one that drives the model, so get it from finance rather than estimating. Use contribution margin, not gross revenue, so the model holds up under scrutiny.
For context on the broader opportunity, the 2025 CAQH Index estimates roughly $21 billion in remaining industry savings from automating manual and partially manual administrative transactions. Referral intake is one of the most manual workflows left in most specialty practices.
A worked ROI example (with labeled assumptions)
The numbers below are illustrative assumptions, not benchmarks. They're meant to show how the model works. Replace every one with your own data.
Assumptions for a hypothetical multi-site oncology group:
- 400 new-patient referrals per month across all sites
- 12% of referrals currently leak (never result in a consult)
- 20 minutes of coordinator time per referral for intake work
- $35 loaded cost per coordinator hour
- $15,000 average first-year contribution margin per converted new patient (this varies enormously by cancer type and payer mix; get your own figure)
Labor savings:
- Current intake time: 400 referrals × 20 minutes = 8,000 minutes, or about 133 hours per month
- If automation removes 70% of that work: about 93 hours per month saved
- At $35 per hour: roughly $3,300 per month, or about $39,000 per year
Revenue recovery:
- Current leaked referrals: 400 × 12% = 48 per month
- If faster intake and record chasing recover one quarter of those: 12 additional consults per month
- Assume half of those become treatment patients with the average margin above: 6 patients × $15,000 = $90,000 per month
Even if you cut the revenue recovery assumption in half, the revenue side still dwarfs the labor side. That's the pattern in most oncology ROI models, and it's why you should spend the most effort validating your leakage rate and margin figures.
Run a sensitivity analysis before you present it
A single-point ROI estimate invites skepticism. A simple sensitivity table earns trust. Build three scenarios:
- Conservative: recover 10% of leaked referrals, 40% conversion to treatment, 50% of intake labor removed
- Expected: recover 25% of leaked referrals, 50% conversion to treatment, 70% of intake labor removed
- Optimistic: recover 40% of leaked referrals, 60% conversion to treatment, 80% of intake labor removed
Using the illustrative inputs above, the conservative case recovers about 5 consults and 2 treatment patients a month, still roughly $29,000 in monthly contribution margin before labor savings. Show your board that the investment clears its hurdle even in the conservative case. If it only works in the optimistic case, it's not ready.
Multi-site groups have an extra lever
Groups with several locations usually have uneven intake performance. One site might process referrals same-day while another runs a three-day backlog. When you model ROI, look at leakage and turnaround by site. Bringing your slowest sites up to your best site's performance is often a more credible story than an across-the-board percentage improvement, and it's easier to verify after go-live.
How treatment delays change the math
There's a clinical dimension that belongs in an oncology business case, even though it isn't a line item.
A 2020 cohort study in JAMA Network Open of more than two million patients found that longer time to treatment initiation was associated with higher all-cause mortality, with absolute risk increases of roughly 1.2% to 3.2% per week in some cancers, including lung, kidney, and pancreas. A meta-analysis in The BMJ found that a four-week treatment delay was associated with a 6% to 13% increase in the risk of death depending on the treatment type.
For a board or physician partners, that evidence reframes the investment. Faster intake isn't only a revenue project; it's a quality-of-care project. It also supports value-based oncology arrangements, where timely care and avoided downstream complications affect shared savings.
Administrative delay is also one of the few parts of time-to-treatment that operations leaders can directly control. Physician capacity, payer approvals, and patient readiness all matter, but the days a referral spends in a fax queue are purely a process problem.
What costs should you include?
A credible business case includes the full cost of ownership, not just the subscription price. Vendors often underplay these:
- Software cost. Platform fees, per-referral or per-page pricing, and any per-site charges. Model cost at your projected volume in year two, not just today.
- Implementation. EHR integration work, configuration of required-document checklists and urgency rules, and testing. Ask vendors for typical implementation hours required from your team.
- Change management. Training coordinators on exception review, updating workflows, and adjusting roles. Budget time for a supervisor or project lead during the first few months.
- EHR-side costs. Some EHR vendors charge for API access or interface setup. Clarify this early.
- Ongoing oversight. Someone needs to review exceptions, monitor accuracy, and tune rules. This is usually a fraction of current intake effort, but it's not zero.
Put these costs against the benefits by month, not just annually. Most models show a ramp period of one to three months before benefits reach steady state, so plot cumulative cost and cumulative benefit to show the payback month clearly.
One more cost to name honestly: the opportunity cost of doing nothing. If your group plans to add sites or physicians in the next two years, the status quo means hiring more intake coordinators in a tight labor market. Include that avoided hiring in the model, using your actual recruiting timelines and turnover rates, not just the salary line.
How to measure payback after go-live
A business case is a prediction. Once you're live, you need to show the prediction came true. Define your measurement plan before implementation so you have a clean baseline.
Track these metrics monthly, by site:
- Referral volume received and referral-to-consult conversion rate
- Time from referral received to ready-to-schedule
- Time from referral to first consult
- Coordinator hours spent on intake, from time studies or workload tracking
- Consults rescheduled due to missing records
- Percentage of referrals processed without human touch versus routed as exceptions
Compare each against the baseline you captured before go-live. Report results at 30, 90, and 180 days. If conversion or timing isn't improving by day 90, dig into why. Common culprits are outreach bottlenecks after intake, or EHR filing that still requires manual steps.
Honey Health's Referral Intake agent is the kind of investment this model evaluates: it captures and processes referrals across sites, files them into the group's existing EHR, and works with Data Fetching and Eligibility agents so fewer referrals stall on missing records or coverage questions. For multi-site groups, standardizing intake across every location is often where the fastest gains appear.
Frequently Asked Questions
How long does it take to see ROI from referral intake automation?
Most oncology groups see labor savings within the first one to three months as intake volume shifts to automation. Revenue recovery takes longer to confirm because you need enough referral volume to measure conversion changes. Plan to report early results at 90 days and a fuller picture at six months.
What's the biggest driver of ROI in oncology referral automation?
Recovered new-patient revenue is usually the largest driver, because converted oncology patients carry significant downstream treatment revenue. Labor savings are real but smaller. Validating your current leakage rate and downstream margin per patient is the most important step in building a credible model.
Should we model ROI per site or across the whole group?
Both. Group-level ROI supports the investment decision, but site-level metrics show where intake problems are worst and where gains are coming from. Sites with the slowest turnaround or highest leakage often deliver the fastest payback.
Does automation reduce intake headcount?
Sometimes, but more commonly it lets a group absorb referral growth and new sites without adding coordinators. Many practices redeploy intake staff to patient outreach and navigation, which further improves conversion.
How do PE sponsors typically evaluate this kind of investment?
Sponsors generally look for a clear payback period, measurable impact on new-patient volume, and scalability across sites and future acquisitions. A model that separates labor savings from revenue recovery, uses the group's own data, and includes full implementation costs tends to hold up best in diligence.
