Memory is a poor instrument for understanding behaviour. By the time you are worried enough about a behaviour to mention it to a vet or a trainer, you will have been living with it for weeks, and your recollection will have flattened it. You will remember the worst episode and forget the ordinary days. You will think something is getting worse when it is actually a long pattern that has been fluctuating all along (Canine Companion Consulting, "The Simple Habit That Helped Me Understand My Dog").
A behaviour diary is the fix. It is not a journal of feelings, and it is not a chore with paragraphs to write every night. It is a few short fields written in the moment, kept over a few weeks, so that patterns become visible instead of being reconstructed from memory. This guide walks through what to record, what to leave out, and how to turn the result into something a professional can actually use.
What a diary is for
The diary does three jobs at once, and it helps to know them before you start.
First, it finds triggers. A behaviour that feels random is very often linked to something specific, being left alone, visitors arriving, a particular time of day, a day without a walk, and the only way to see the link is to record both the behaviour and what was happening around it. Dogs Trust, in its advice on aggressive behaviour, tells owners to keep a record of where the behaviour occurs, when it occurs, and other potential influences, because a diary can help you spot the pattern behind something that feels unpredictable (Dogs Trust, "Signs of Aggression in Dogs and What To Do").
Second, it makes you useful in a consultation. A vet or behaviourist has limited time, and "he spins quite a lot" earns sympathy where "episodes went from twice a week to daily over six weeks and almost all follow being left alone" earns a treatment plan. The diary turns a vague worry into numbers a professional can act on.
Third, it shows you change. Behaviour change is slow and uneven, gains followed by setbacks, and without a record it is easy to feel like nothing is improving. The diary is how you notice that the dog is recovering faster, or reacting less intensely, when the day to day difference is too small to feel.
What to record for each episode
Keep the fields short and the same every time. A five-part entry is enough, and short fields are what makes the habit survive.
Note the time and date. Note what happened immediately before, in plain words, "doorbell", "put in the crate", "the lead came out", or "nothing obvious". Note what the dog did and how long it lasted, in seconds or minutes rather than a rough guess. And note how it ended, stopped on their own, stopped when called, needed physical separation, could not be interrupted. "Nothing obvious" is a real and useful answer too, because a behaviour that no longer needs a trigger is further along than one that still does.
The most diagnostic two of these are duration and interruptibility. Whether a behaviour could be interrupted, and how much effort it took, tells you far more about its severity than how "bad" it looked, because your sense of what counts as bad will drift as you get used to the behaviour, where a measured duration does not. These are exactly the measures a professional will want, so recording them from the start means the data is already in the right shape.
Record the context too, not just the incident
A list of episodes explains nothing on its own. Next to it you need a lighter daily note of the things that set the stage, because those are usually where the pattern hides.
Worth capturing briefly each day: roughly how much exercise and what kind; how much mental work, training, scent games, puzzle feeders, or none; anything unusual in the household like visitors, absences, noise or schedule changes; sleep, if it is disturbed; and any medication, including doses missed. The exercise and mental-stimulation column is the one owners most often regret leaving out, because the link between a low-enrichment day and a bad episode is usually the single most actionable finding in the whole diary.
The point is not to write a life history. It is a line or two, enough that when you look back you can say whether a bad day followed a quiet day or a busy one.
Log the quiet days too
This is the part nearly everyone skips, and it is the part that makes the rest useful. If you only write down the bad days, you cannot tell the difference between a genuinely good week and a week you simply forgot to record.
One short daily line, even just "none today", is what makes the frequency countable. Without recorded quiet days, frequency is impossible to measure, and a real improvement can look like a failure, because after a treatment starts people tend to keep noticing the episodes that still happen while the absent days silently disappear from view. Logging the word "none" on an ordinary day is what lets you actually see that the bad days are thinning out.
Write in the moment, not at bedtime
The single biggest reason a diary collapses is that it is left to the end of the day, when the detail is already gone. Duration and trigger both degrade within hours, and the entry you write at bedtime will be shorter, vaguer and slightly wrong.
The fix is to make recording near the moment as easy as possible. A note on your phone, a sticky note by the door, a notebook in the kitchen, wherever the behaviour tends to happen, and the message is the same. If you cannot record as it happens, record at the next natural pause, rather than promising yourself that you will remember the cluster of small behaviours that have already blurred together.
Keep entries short or the diary will die
The most common way a behaviour diary fails is that the entries become paragraphs. Long entries collapse within a fortnight, because nobody sustains paying writing essays every day. Short structured fields survive.
The discipline is to treat the diary as data, not as a story. Seconds, not sentences. A one-line trigger, a duration, an interruptibility, a one-line outcome. The goal is a page you can fill out in a minute, because a diary that takes a minute is a diary that keeps going, and a small amount of consistent data over three weeks beats a rich, detailed diary abandoned after three days.
How long before patterns appear
Give it two to four weeks. That is long enough for the triggers to start showing up reliably, and it is the window a behaviourist will usually ask you to have recorded before your appointment. A week feels long when you are living through it, but it is the minimum for the pattern to separate itself from the noise of a few random days.
What typically emerges from two weeks of consistent logging is the structure behind the behaviour: episodes that cluster around confinement or being left alone, episodes that surge before walks or when visitors arrive, more frequent episodes on low-exercise days, or a clear time-of-day pattern tied to the household winding down. Each of those is a different conversation, and each becomes visible only because you wrote down both the behaviour and the day around it over and over rather than once. The diary does the detective work your memory refuses to do, because it never forgets the ordinary days.
Keep going past that first fortnight, because the same diary is also what tells you whether a change, a medication adjustment, or a training plan is actually working, and it is how you catch a relapse early instead of late.
Why a severity score drifts
It is tempting to add a score out of five of how severe each episode felt, and you can if you find it useful, but be aware of its weakness. Your sense of what counts as bad will shift as you get used to the behaviour, so a score of four in week one and a score of four in week four might describe very different actual events.
Duration and interruptibility do not drift in the same way. Thirty seconds is thirty seconds, and "could not be interrupted" means the same thing every week. That is why those two carry far more weight than a subjective score, and why it is worth building the diary around them even if you also keep a score for your own sense of the picture. The diary that tells the truth is the one built on measures that stay put.
Turn the diary into a summary before you go
Do not hand a professional a raw pile of entries and expect them to read it in the appointment. Summarise first. The diary exists to let you produce four numbers in thirty seconds.
How many episodes in the last four weeks, and how that compares with the four weeks before. The average and the longest duration. What proportion could not be interrupted. And the single most common preceding event. Those four numbers are the difference between a consultation that goes somewhere and one that starts from scratch.
If you have video of a typical episode, bring that too. It conveys what a written log cannot, and it helps a professional distinguish a behaviour from something else that looks like it. And in that video, the dog should simply be doing what the dog does, not prompted into a particular behaviour for the camera, which matters for reasons covered below.
Do not provoke behaviour in order to record it
There is a specific line the diary must never cross, and it is worth saying outright: you observe and record what the dog already does, you never set out to make the behaviour happen just so you can write it down.
This is for the dog's welfare and for the science of it. Deliberately exposing a dog to the thing they fear, or taking an item a guarding dog is protecting, or shutting a separation-stressed dog in a room, changes the behaviour you are trying to measure and can make it worse. A diary built on provoked episodes is not a baseline, and it is not kind. The record is only useful if it captures the dog's ordinary day, and that means letting the day happen, not staging it.
There is a wider version of the same rule for behaviours that involve real risk. If the dog is showing aggression, guarding, or distress that you cannot safely live around, the first move is not to logging it for weeks while testing the edges. It is to keep everyone physically safe by avoiding the situations that trigger it, and to get professional help. The vet-authored guidance for aggression is explicit that the first part of any treatment is preventing further episodes, because each episode reinforces the behaviour, and that you are not showing the dog any triggers until a professional has built the plan (VCA Animal Hospitals, "Aggression: Getting Started with Safety and Management"). For a sudden behaviour change, the first stop is the vet before the diary, because new guarding, hiding or snapping can be pain or illness rather than a behaviour pattern at all.
Decide what counts as an episode before you start
For the diary to produce clean counts, you need a definition of what you are recording, decided in advance and held steady. Otherwise your counts drift with your mood as much as the dog's behaviour.
If the concern is barking at the window, decide whether a single woof counts or only a sustained bout, and where the line is. If the concern is spinning, decide whether a couple of circles during play counts or only the repetitive spinning you are worried about. Write the definition at the top of the diary so a week from now you are still counting the same thing the same way.
This sounds fussy, but it is the thing that makes the "how many episodes this week versus last week" number mean anything at all. A count is only as good as the rule that produced it, and a rule you wrote down beats a rule you are improvising from memory.
Do not let the diary become rumination
One caution, because for a worried owner the diary can quietly tip from useful into obsessive. The diary should lower your anxiety by replacing vague dread with facts, not raise it by turning every glance at the dog into data collection.
Keep the recording to the moments that matter, and let the rest of the day be ordinary. The point is to notice the pattern, not to narrate the dog's every move into a document. If you find the diary making you watch the dog more anxiously rather than less, loosen the format, record fewer fields, or hand the record to a professional sooner. A diary that makes you calmer is working; a diary that makes you watch the dog like a hawk is not.
The diary is not the same as a training log
It is worth separating two things that are easy to confuse, because they answer different questions.
A behaviour diary records incidents and context, and it is for understanding a problem. A training log records skills and practice, and it is for noticing what the dog has learned. The behaviour diary is the tool for the "suddenly he snapped at the postman" question; the training log is the tool for the "can he sit when I ask outside now" question.
Both you find useful, and both feed a professional, but they should not be mashed into one document, or you end up with a form that captures neither well. For the skill of noticing progress, which is the training-log side of this, see tracking training progress without counting perfect days.
A minimal format you can start today
The way to start a diary is not to design a perfect system and never finish it. It is to write the five fields somewhere and start.
At its simplest, write these on a note and photo them each evening: time, what came before, what the dog did, how long, and how it ended, plus a line on exercise and anything unusual in the day, and the word "none" on quiet days. Keep it that small for a week and you will already see the shape of it, because the pattern you are looking for usually shows up as it accumulates over a couple of weeks of honest repetition.
The real reason people keep diaries going, though, is what it gives back. It takes a worry that feels huge and shapeless and turns it into something with structure that you can think about, share, and act on. For a dog and owner living with a hard behaviour, that is not a small thing. It is the difference between feeling behind your dog's behaviour and standing alongside it, and between guessing and knowing.
When the diary points at something that needs more than logging, the next article is when a dog needs professional behaviour help rather than another tutorial, and when you want to read the warning signs that go into the log properly, start from reading your dog's body language without guessing from one signal.
Our behaviour guides are gathered under our behaviour guides.