You know what the letters stand for. That is not where notes go wrong.

They go wrong in the Assessment, on a visit with three problems, when all three get blended into one paragraph that says everything is stable. The format has one Assessment box. The visit had four things happening in it.

That mismatch is not your fault, and it is not a flaw in SOAP. It is what happens when a format built around one thing gets used for another.

Here is the blank template, the per-problem block that fixes the Assessment, and three worked examples across primary care, nursing and behavioral health.

Key Takeaway: A SOAP note records a clinical encounter in four sections: Subjective, what the patient reports; Objective, what you measured and observed; Assessment, your clinical judgment; and Plan, what happens next [1]. Larry Weed developed the structure as the progress note component of the problem-oriented medical record, where every active problem on a numbered list got its own assessment and plan [1][2]. Most notes today blend all of a visit's problems into a single Assessment block, which is where multi-problem visits go vague. Writing the Assessment and Plan per numbered problem fixes it, and it matches how the visit level is now judged, since medical decision making is scored partly on the number and complexity of problems addressed [3]. Sully.ai's AI Scribe captures the reasoning you say out loud during the visit and writes it into the note in the structure you have saved.

What a SOAP Note Template Covers

Four sections, in a fixed order, and the boundary between the first two is where most notes leak.

The Four Sections and What Belongs in Each

Subjective is what the patient tells you. The chief complaint, the history of present illness, the relevant review of systems, medications and allergies [1].

Objective is what you measured or observed yourself. Vital signs, exam findings for the systems you actually examined, results you reviewed [1].

Assessment is your clinical judgment. Not a restatement of the complaint, and not a bare diagnosis label [1].

Plan is what happens next, and StatPearls is specific that it goes "for each problem": the testing needed and why, the therapy, referrals, patient education [1].

Where the Subjective and Objective Line Sits

The test is who produced the information.

"Patient reports chest tightness walking uphill" is Subjective, because the patient said it. "Regular rate and rhythm, no murmur" is Objective, because you listened.

The review of systems is Subjective even though it is organized by body system. That is the single most common misfile, and it happens because the format looks anatomical. It is still a set of questions you asked and answers you were given.

What SOAP Was Originally Built Around

Here is the part almost nobody teaches, and it explains the Assessment problem completely.

SOAP was not designed as four boxes for a visit. Larry Weed built it as the progress note component of the problem-oriented medical record, and its unit was the problem, not the encounter [1][2]. Every active problem sat on a numbered, maintained problem list, and each one got its own assessment and plan, so the reasoning on problem three could be read without wading through problems one, two and four.

Notice that the instruction survived even where the reason did not. StatPearls still tells you to write the Plan for each problem [1]. It never explains the record that requirement came from.

The collapse is documented rather than cultural. A 2026 scoping review in Family Practice synthesized 103 studies and found problem lists "central to the problem-oriented medical record" but commonly "incomplete, outdated, or inconsistently maintained," with unclear ownership and real disagreement about what even counts as a problem [2].

Per-problem documentation did not fall out of fashion. Its foundation stopped being maintained.

And going back to it is not nostalgia. Since the evaluation and management revisions, the visit level rests on medical decision making or total time, and medical decision making is scored partly on the number and complexity of problems addressed [3][4]. Blending four problems into one paragraph hides the exact thing the level rests on.

The SOAP Note Template

Copy the first block. Use the second one inside it.

The Blank Template to Copy

This block is also available as a SOAP note template you can open directly.

Patient: | Date and time: | Encounter type:

SUBJECTIVE

• Chief complaint:

• History of present illness:

• Review of systems (pertinent positives and negatives):

• Relevant past history:

• Medications:

• Allergies:

OBJECTIVE

• Vital signs:

• Physical exam (systems examined only):

• Results reviewed:

ASSESSMENT AND PLAN (per numbered problem, see below)

Signature and credentials:

Delete the lines you did not perform. An empty field is honest. A default value is a claim.

The Per-Problem Assessment and Plan Block

This is the repeating unit. Use it once per active problem.

1. [Problem], [new / stable / worsening / resolved]

• Assessment: One or two sentences of reasoning. What supports it, and what you considered and ruled out.

• Plan: Specific actions. Tests with the reason for ordering them, medication changes, referrals, patient education, follow-up interval.

2. [Problem], [status]

• Assessment:

• Plan:

Two details in that block are doing real work.

The status word matters because "stable chronic illness" and "worsening" are the language the problems element of medical decision making is graded on [3]. Writing "stable" is not padding. It is the grading term.

The ruled-out clause matters because it is the only place in a note that shows a differential was considered at all. Without it, a correct diagnosis and a lucky guess look identical on paper.

How to Write the Assessment Most Notes Get Wrong

A Vague Assessment Rewritten Per Problem

Here is a real-shaped weak Assessment from a three-problem primary care visit:

Assessment: Patient with diabetes, hypertension and knee pain. Overall stable. Plan: Continue current medications, follow up in 3 months.

Nothing in that is false. It is just unreadable as clinical reasoning, and it is unscoreable as documentation.

The same visit, per problem:

1. Type 2 diabetes mellitus, stable

• Assessment: A1c 7.1, down from 7.6. Home glucose logs consistent with the A1c. No hypoglycemic episodes reported. Diet adherence has improved since the last visit.

• Plan: Continue metformin 1000 mg twice daily. Repeat A1c in 3 months. Annual retinal screening is overdue, referral placed today.

2. Essential hypertension, worsening

• Assessment: 152/94 today, 148/90 at last visit, up from a run in the 130s. Reports missing doses roughly twice a week. No chest pain, no headache, no visual change, so this reads as adherence rather than a new secondary cause.

• Plan: Continue lisinopril 20 mg daily and add a pill organizer discussion. Home blood pressure log for 2 weeks. If the average stays above 140/90, add amlodipine at the next visit.

3. Right knee pain, new

• Assessment: Three weeks of medial joint line pain, worse with stairs, no trauma, no effusion, no locking or giving way. Presentation is consistent with degenerative change rather than internal derangement.

• Plan: Weight-bearing knee radiograph to assess joint space, since it changes whether this is managed conservatively or referred. Acetaminophen as needed. Reassess in 4 weeks.

Read the two versions again. No clinical work was added. The clinician already knew the A1c had dropped, already suspected adherence, already had a reason for ordering the film. The rewrite only wrote down decisions that had already been made in the room.

Why the Problem List Stopped Working

If per-problem notes are better and the format was built for them, the obvious question is why almost nobody writes that way.

The scoping review answers it. Across 103 studies, problem lists were widely recognized as foundational, and just as widely incomplete, outdated and inconsistently maintained. Ownership was unclear, meaning nobody was quite responsible for curating them, and clinicians disagreed on what qualified as a problem in the first place [2].

You cannot write per problem against a list you do not trust. So the list decayed, and the Assessment absorbed the mess.

The same review found that newer tools helped most when they were built into clinician-led workflows rather than running on their own [2]. That distinction matters more than it sounds.

Where Sully.ai Fits in SOAP Documentation

The reasoning that belongs in the Assessment gets said out loud during the visit. You explain to the patient why the blood pressure reading concerns you, why you want the film, what you are watching for. By the time the note gets written after clinic, all of that has compressed into three diagnosis labels.

Sully.ai's AI Scribe captures what is said during the encounter and writes it into the note in the structure you have saved, so a SOAP note comes back as SOAP with the problems kept apart instead of merged. It runs on a single integration across Epic, Cerner, Meditech and Athenahealth.

Now the part worth saying on a page that also sells a scribe.

A 2026 paper in JMIR Medical Education argues the chart note was historically a clinician-centered cognitive tool for reasoning and teaching, and raises cognitive off-loading and de-skilling as underexplored risks of ambient AI documentation [5]. That concern is legitimate and we are not going to wave it away.

The line we hold is this: a scribe should capture the reasoning you did, not stand in for doing it. Ambient capture is good at the Subjective and the Objective, which are recall problems. The Assessment is a judgment problem, and it is the one section worth reading closely before you sign, every single time.

Sully operates across 5,000+ providers, has delivered 50M+ hours of AI work, and prices each AI role 80 to 90 percent below the human equivalent [6].

SOAP Note Examples Written Out

Three settings. The four sections never move. What counts as Objective changes in every one.

A Primary Care Visit With Three Problems

Established patient, 58, routine follow-up.

SUBJECTIVE: Here for diabetes and blood pressure follow-up. Reports feeling well overall. Home glucose readings mostly 110 to 140 fasting. Admits missing blood pressure medication about twice a week, usually on weekends. New complaint of right knee pain for 3 weeks, worse going down stairs, no injury. Denies chest pain, shortness of breath, headache or visual change. Medications: metformin 1000 mg BID, lisinopril 20 mg daily. NKDA.

OBJECTIVE: BP 152/94, HR 76, BMI 31.2, temp 36.8. General: well appearing, no distress. Cardiovascular: regular rate and rhythm, no murmur, no peripheral edema. Respiratory: clear to auscultation bilaterally. Right knee: medial joint line tenderness, no effusion, full range of motion, ligaments stable. Results reviewed: A1c 7.1 (was 7.6), basic metabolic panel within normal limits.

ASSESSMENT AND PLAN: The three numbered problems from the rewrite above.

A Nursing Shift Note

Post-operative day 1, total knee arthroplasty. Day shift.

SUBJECTIVE: Reports pain 6 out of 10 at rest in the operative knee, 8 out of 10 with movement. States the overnight dose "wore off early." Reports nausea after breakfast. Denies chest pain or shortness of breath.

OBJECTIVE: BP 128/76, HR 88, RR 18, temp 37.1, SpO2 96% on room air. Dressing dry and intact, no strikethrough. Operative leg warm, distal pulses palpable, capillary refill under 3 seconds, able to dorsiflex and plantarflex. Calf soft and non-tender. Intake 900 mL, output 750 mL. Ambulated 15 feet with a walker and one assist, tolerated with reported pain 8 out of 10. Antiemetic given at 0930 with relief by 1015.

ASSESSMENT AND PLAN:

1. Acute post-operative pain, inadequately controlled

• Assessment: Pain is limiting participation in mobility, which is the main risk to the recovery plan on day one. Current interval dosing is not covering the window before therapy.

• Plan: Notified surgical team at 1040 regarding pre-therapy dosing. Ice applied. Reassess pain 30 minutes after next dose and before the afternoon therapy session.

2. Post-operative nausea, resolving

• Assessment: Responded to antiemetic within 45 minutes. Tolerated clear liquids afterward.

• Plan: Advance diet as tolerated. Antiemetic available as needed.

One thing to notice, because nurses get corrected on it constantly. The Assessment in a nursing note is a judgment about the patient's response to the plan of care, not a medical diagnosis. "Inadequately controlled" is a nursing assessment. "Osteoarthritis" is not.

A Behavioral Health Session

Depression follow-up, fifth session, 45 minutes.

SUBJECTIVE: Reports mood "a bit better, maybe a 5 out of 10" compared with 3 out of 10 a month ago. Sleeping 6 to 7 hours, up from 4. Returned to a weekly running group two weeks ago and has gone three times. Still describes low motivation on weekday mornings. Denies suicidal ideation, plan or intent. Taking sertraline 100 mg daily, reports no side effects.

OBJECTIVE: Alert and oriented. Grooming improved from prior sessions. Mood reported as "better," affect mildly constricted but reactive, brighter when discussing the running group. Speech normal rate and volume. Thought process linear and goal directed. No psychomotor slowing observed. No evidence of psychosis. Insight and judgment intact. PHQ-9 score 11, down from 17 four weeks ago.

ASSESSMENT AND PLAN:

1. Major depressive disorder, recurrent, moderate, improving

• Assessment: PHQ-9 down 6 points with sleep and behavioral activation both improving. Response is partial rather than complete, since morning motivation and constricted affect persist. Improvement tracks the reintroduction of the running group as much as the medication, which is worth naming to the client.

• Plan: Continue sertraline 100 mg daily, no change this visit. Continue behavioral activation with the running group as the anchor. Repeat PHQ-9 in 4 weeks. If the score plateaus above 10, discuss a dose increase with the prescriber.

Objective here means mental status and observed behavior, plus a scored instrument. There are no vitals in it, and that is correct. This is the section behavioral health SOAP notes most often leave vague, and a psychiatry SOAP note template is the fastest way to stop that happening.

Book a demo and bring yesterday's busiest visit. The question worth asking is how many problems it had, and how many of them you can find separately in the note.

FAQ

Q: What does a SOAP note include? Four sections. Subjective is what the patient reports, including the history of present illness and the review of systems. Objective is what you measured or observed, such as vital signs, exam findings and results reviewed. Assessment is your clinical judgment on each active problem. Plan is what happens next for each of those problems, including tests with the reason for ordering them, treatment changes, referrals and follow-up [1].

Q: What are some common SOAP note mistakes? Four show up constantly: copy-forward history padding the Subjective, an Objective section listing systems nobody examined, an Assessment written as a bare diagnosis label instead of reasoning, and a Plan that says "continue current management" with no rationale. The structural mistake underneath all of them is blending every problem of a multi-problem visit into one Assessment, which hides the number and complexity of problems addressed, the element the visit level is partly scored on [3].

Q: Is there a standard SOAP note template? There is a standard structure but no single mandated form. The four sections are consistent across settings [1]. What fills the Objective section is not: vitals and physical exam in primary care, wound and intake and output on a nursing shift, mental status and a scored instrument in behavioral health. The blank template on this page works in all three, and Sully.ai's AI Scribe can return notes in whichever structure you save.

Q: Do nurses have to do SOAP notes? It depends on the employer and the setting, not on any national rule. Many nursing programs teach SOAP, and plenty of facilities use narrative, DAR or SOAPIE formats instead. When nurses do write SOAP, the Assessment is a judgment about the patient's response to the plan of care rather than a medical diagnosis. There is a full post-operative shift note above.

Q: Do you have a free SOAP note template? Yes. The blank template and the per-problem Assessment and Plan block are both on this page as copy-paste blocks, with three worked examples across primary care, nursing and behavioral health. Sully.ai's AI Scribe can also capture the encounter and write the SOAP note in the structure you have saved.

Sources

[1] StatPearls, National Center for Biotechnology Information — SOAP Notes [2] Family Practice — Current state of electronic problem lists in primary care, a rapid scoping review [3] American Academy of Family Physicians — Evaluation and Management Coding [4] American Medical Association — CPT Evaluation and Management [5] JMIR Medical Education — Re-Centering Clinical Documentation in the Age of AI Scribes, Four Aims of the Patient Chart Note [6] Sully.ai — The AI Workforce for Healthcare