Cost of Manual Prior Authorization: Build a Request-Level Cost Ledger

The useful question is not which national cost-per-request figure is most persuasive. It is what your process costs, which activities create that cost and which of them a proposed service or software change would actually alter.

Define what one request includes

Choose the unit being costed. For this worksheet, one request means a single identified administrative authorization request followed through the defined observation period. A resubmission or information response stays attached to that request unless your documented counting rule says otherwise.

Record the boundary. Are you including initial preparation, repeated status checks, additional information responses and final administrative documentation? Are appeals outside the analysis? If so, show them separately. A narrow baseline can still be useful when its limits are clear.

Distinguish a request that is resolved from one that is still open. The recorded cost of an open request is cost to date, not its final cost. Do not compare that figure directly with fully observed completed requests without explaining the difference.

Record work by role and activity

Use one row for each observed activity. Ask staff to record the task and time without including patient-identifying details in a shared costing worksheet.

A task can be necessary without being preventable. Recording its cost does not prove that automation would remove it. Keep clinical judgments and other work requiring qualified people visible when comparing future arrangements.

For the hourly rate, ask finance which compensation and overhead elements are included. The BLS National Compensation Survey concepts distinguish wages and salaries from benefit costs. That distinction explains why an hourly wage alone is not a fully loaded labor rate. It does not tell you the correct rate for your practice.

Include rework without double counting it

Suppose a coordinator spends time checking a submission, later responds to a missing-information request and then records the outcome. Those are separate activities. Count each once against the original request.

Be careful with multitasking and waiting. A portal that remains open for an hour does not establish an hour of labor. If a person is on hold and cannot do other work, record the observed activity according to the team's costing method. If they work on another request, do not charge both requests for the same undivided hour.

Shared work needs an allocation rule too. A team meeting or a system-wide reconciliation can be included as an overhead pool, but it should not also be recorded in full against every request. Write down the rule and keep it stable during comparisons.

The cycle-time measurement guide helps distinguish elapsed intervals from hands-on effort. The two measures can sit next to each other, but they answer different questions.

Want to compare your ledger with an automated workflow? Book a Linear Health demo and identify which recorded activities you want to examine.

A hypothetical request-level calculation

All rates and times below are invented for arithmetic illustration. They are not market wages, an industry benchmark or a Linear Health customer result.

Coordinator labor is 30 minutes multiplied by $30 per hour, divided by 60, or $15. The separately recorded five minutes at $72 per hour adds $6. Total observed labor is $21.

If the worksheet includes a documented $2 allocation for a tool that is not already included in the hourly rates, the modeled total becomes $23. Label that addition. If the tool cost is already in the loaded rate or another overhead pool, adding it again would overstate the result.

This example does not assign a revenue loss to the request. It also does not assume every request uses the same amount of work. Apply the calculation to your observed records, then describe the distribution and the scope of the sample.

Distinguish capacity from cash savings

Removing an activity can release staff time without reducing payroll. That capacity may be useful for other work, but the financial claim should describe what actually changes.

In another hypothetical scenario, eliminating six minutes of work across 100 comparable requests releases 600 minutes, or 10 hours. At an assumed loaded rate of $30, that is $300 of modeled labor capacity. It becomes cash savings only if an actual expense changes, such as a documented reduction in overtime or a service invoice.

Avoid converting every released hour into booked savings while also claiming the full value of the work performed with that hour. Those are different financial scenarios. Finance should decide which one the budget uses and what evidence would support it.

Keep potential revenue effects outside the labor ledger until they have their own justified assumptions. A delayed request, an unbooked appointment and a lost payment are not interchangeable events.

Use the baseline in a purchase decision

Compare each proposal against the same activity list. For every row, mark whether the work is removed, transferred, changed or retained. Include new tasks such as exception review, quality checks and administration of the service.

Then combine the quoted fee with the retained work and any documented transition costs. The prior authorization software cost guide addresses proposal costs. The outsourcing versus automation comparison addresses the operating-model decision. This ledger supplies the practice baseline for both.

Linear Health's prior authorization offering can be evaluated against those specific administrative tasks. A demonstration should clarify the remaining responsibilities and the evidence available to measure the changed process, rather than requiring you to accept a universal cost reduction.

Bring the work, not an assumed average

A short, traceable ledger gives a vendor discussion a concrete starting point. Book a demo to compare your administrative activities with Linear Health's workflow.

Frequently asked questions

Sources

Sources accessed October 21, 2026.

• BLS National Compensation Survey: Concepts.