How to Track Migraines Accurately: A Step-by-Step Guide

Learning how to track migraines accurately is one of the most practical things you can do for your own care. It sounds simple: write down your headaches. But the details matter more than people expect, and small logging errors accumulate into a distorted picture that misleads both you and your doctor. This guide covers the full process from what counts as a migraine day, to what to capture during an attack, to the common mistakes that quietly ruin months of data.

What Counts as a Migraine Day

This is the definition that headache specialists and clinical trials use: a migraine day is any calendar day on which you have a migraine-type headache lasting at least four hours without treatment, or a headache of any duration that you treat with a migraine-specific acute medication.

A few practical clarifications:

• Duration under four hours, treated: If you take a triptan or CGRP acute medication and the attack resolves in two hours, that still counts as a migraine day.
• Duration under four hours, untreated: If the attack resolves on its own in two hours without medication, most specialists do not count it unless it fits the full International Headache Society criteria (two of four pain characteristics plus nausea/vomiting or light/sound sensitivity). When in doubt, log it and note the duration. Your doctor can make the call.
• Attacks spanning midnight: If a single attack runs from 10 PM to 3 AM, you have one migraine that crossed midnight. Count the calendar day with four or more hours of pain as the migraine day. If both calendar days have at least four hours of pain, count both.
• Headache days vs. migraine days: Not every headache is a migraine. Log all headaches, but note whether they met migraine criteria. Tension-type headaches and mild headaches are still clinically relevant, just recorded separately.

The reason precision here matters: your monthly migraine days count is the primary metric your neurologist uses to assess severity and track improvement over time.

Building a Reliable Baseline

A baseline is the pre-treatment or pre-change period against which everything else is measured. It is the "before" in the before-and-after story of your treatment.

To establish one:

1. Log every day, not just attack days. Headache-free days need to be recorded explicitly. Without them, you cannot compute monthly migraine days or measure reduction accurately.
2. Run the baseline for at least three months. Two months can be distorted by an unusually good or bad month. Three months gives you a stable average.
3. Do not change treatments during the baseline period if you can help it. If you start a new preventive while still collecting baseline data, you contaminate the pre-treatment measurement.
4. Record the date your baseline period starts. When you compare later, you need to know exactly which months you are averaging.

If you are starting a CGRP preventive, your prescriber needs this baseline to determine whether the medication is working at your follow-up. The number of monthly migraine days before treatment compared to the number during treatment is the core measure. For more on what that comparison looks like in practice, see the guide on measuring CGRP progress.

What to Log During an Attack

Consistent data fields make your log comparable across attacks. During or immediately after each attack, record:

• Date and time of onset (the exact moment the headache began, not when you noticed it had been going for a while)
• Severity at peak (a simple 1 to 10 scale is fine; be consistent about how you use it)
• Pain location (one side, both sides, frontal, occipital, behind the eye)
• Duration (time from onset to pain-free, not just until you felt better)
• Aura (visual disturbances, sensory changes, speech difficulties, and their duration)
• Accompanying symptoms (nausea, vomiting, light sensitivity, sound sensitivity, smell sensitivity, cognitive fog)
• Possible triggers noted at the time, not reconstructed later
• Medications taken (name, dose, and time taken after onset)
• Medication response (did it help? How quickly? Did you need a second dose?)

The complete field list and why each entry matters for your doctor is covered in detail in what to log in a migraine diary.

Common Mistakes That Corrupt Your Data

These are the patterns that quietly undermine months of logging:

Logging only severe attacks. This is the most common error. When a moderate attack resolves quickly, it feels not worth recording. Over months, this creates a dataset that understates your frequency and overstates average severity. Every attack goes in the log.

Logging from memory at the end of the week. Memory for pain intensity and timing degrades fast, often within 24 hours. Prospective logging (recording during or immediately after) is significantly more accurate than retrospective reconstruction. Set a daily reminder if you need one.

Inconsistent severity scales. Calling an attack a 7 one month and a 4 the next because your reference point drifted makes severity comparisons meaningless. Anchor your scale: a 10 is the worst you have ever had, a 1 is barely noticeable but clearly there, a 5 is significant but you can still function with effort.

Not logging headache-free days. This one bears repeating. If your log has entries only on attack days, it tells you nothing about frequency. A log that says "January: entries on the 3rd, 7th, 14th, 22nd, and 28th" is only meaningful if you can confirm the other 26 days were pain-free.

Stopping logging when things improve. The whole value of tracking over time is the comparison. If you stop logging once a treatment starts helping, you lose the data that would prove it is helping.

Reclassifying attacks in hindsight. After a month goes by, you may be tempted to decide a logged attack was "just tension" and not count it. Do not revise historical entries. Log what happened when it happened, and let your doctor interpret it.

Capturing Context That Explains Patterns

Beyond attack data, context entries on non-attack days are what let you identify patterns. Useful daily or near-daily context fields:

• Sleep hours and quality (rough rating)
• Stress level (1 to 5 scale)
• Menstrual cycle phase for people who menstruate
• Notable dietary or caffeine changes
• Physical activity
• Weather or barometric pressure notes

You do not need to log all of these every day forever. The goal is enough context that when you see a cluster of attacks, you can look backward and check what else was going on. Context without attack data is useless; attack data without context misses the "why."

Turning a Log Into a Document Your Doctor Can Use

A raw list of attack dates is a starting point. What makes a log clinically useful is structure:

• Monthly migraine day counts for each month
• Average severity per month
• Acute medication use days per month (this matters for detecting medication overuse headache)
• Notable context patterns (attacks cluster around cycle phase, stress periods, or specific conditions)

If you are on a CGRP preventive, your neurologist or prescriber needs this summary formatted for insurance documentation at renewal. The CGRP prior authorization renewal process specifically requires before-and-after monthly migraine day comparisons.

Migraine Tracker: CGRP Log is built exactly for this workflow. One-thumb daily logging captures attack data and context as it happens. The app automatically calculates monthly migraine days, tracks acute medication use, and generates a one-page renewal report with your baseline versus current numbers, so the data you collected actually reaches your doctor in a format they can act on.