Dog Scent Work Training: Nosework and Tracking

Start by paying the dog for finding food in boxes without any help from you, then move the reward from the food itself to a target odour, because the skill being trained is independent searching - and the fastest way to break a scent dog is to keep telling it where to look.

Key facts

Difficulty
Moderate — needs consistency, not skill
Time to results
A dog searching a room for a hidden target in 2 to 4 weeks of short sessions; a tracking dog working an aged trail takes months
What you need
A dozen cardboard boxes or containers, and food the dog will search for; A long line and a harness for anything done outdoors; Somewhere the dog can work without being called or corrected; Water, and a plan for heat, because both change what the nose does
When to get professional help
A dog that stops working a scent it used to find, or that suddenly cannot follow a trail, needs a veterinary examination - nasal disease, systemic illness and some medications all reduce olfactory performance, and a dog that has gone off the game is often a dog that is unwell.
Approach
Reward-based only, per the AVSAB position statement

What nosework is, and what it is not

A searching game with a defined target. The dog is taught to hunt for one odour — food first, then usually birch, anise or clove in competition — and to tell you when it has found it, while you stay out of the way.

It is not obedience with a nose attached. The handler's job is to hold the line and read the dog; every instruction you give is information the dog did not have to work out, and the point of the exercise is the working out.

It is also not tracking. Nosework searches an area for a static target; tracking follows a person's trail across ground. Different problem, different training, and the two are covered separately below.

The one measured welfare result

Duranton and Horowitz put pet dogs through a cognitive bias test, then had one group practise nosework daily for two weeks and a control group practise heelwork for the same time, then tested them again.

The nosework group approached the ambiguous stimulus significantly faster after the two weeks. The heelwork group did not change. The authors' conclusion is that letting dogs spend more time using olfaction makes them more optimistic, and they frame it as a welfare gain from allowing foraging behaviour.

That is a small study and a single measure, and it should be read as one good result rather than a law. But it is the only controlled evidence in this hobby, and it points at the sniffing rather than the ribbons.

Starting nosework at home

Four stages, no equipment worth buying yet, ten minutes a session.

Put food in boxes and say nothing

A dozen open cardboard boxes, food in one, dog released into the room. Do not point, do not help, do not name it. The dog finds the food and eats it, and that is the entire first session.

Make the search harder before you make it different

More boxes, food deeper, boxes in corners and on chairs. You are building persistence — the willingness to keep working a room after the easy places are exhausted.

Move the reward off the target

Put a scented cotton bud in a tin with holes in it, and pay from your hand the instant the dog's nose is at the tin. The food is no longer in the box; finding the odour is what produces it.

Teach a trained final response

A sit, a stare or a nose freeze at source. Mark it and pay at the source, never behind you, because a dog paid away from the find learns to leave the find. The mechanics are the same marker and shaping skills covered in dog trick training.

Tracking is a different sport with a different cue

Tracking follows a specific person's trail on the ground, in order, at a walking pace, usually on a harness and a long line. The dog is not searching an area; it is committing to a line and staying on it.

The training runs backwards from a certainty: short, straight, freshly laid tracks with food in every footstep, so the dog learns that the trail itself pays. Length, age, turns and surface changes come later, one at a time.

The handler's discipline is harder here than in nosework. You cannot see the track, so the temptation is to steer — and steering a tracking dog teaches it to check with you instead of the ground, which is the one habit that ruins a tracking dog.

The conditions that change what your dog's nose can do

Jenkins and colleagues reviewed the health and management factors that affect olfactory performance in detection dogs, and most of them are things an owner controls on the day.

Heat and exertion matter: a panting dog is moving air for cooling rather than sampling, and high ambient temperature and lack of conditioning both reduce performance. Hydration and blood flow matter. Inflammation and systemic disease matter, which is why a dog that suddenly cannot do a search it used to do is a veterinary question.

Food matters too — the review reports that dietary fat content, meal size and the timing of meals have been shown to affect olfaction in dogs and mice. Practically: search before the big meal, not after it, and stop when the dog starts panting rather than when the clock says so.

Why searching tires a dog that running does not

Because searching is a decision problem and running is not. The dog is sampling, discarding, re-sampling and holding a criterion in mind, which is work of a kind a walk does not produce.

The MSD manual's observation about arousal is the practical version: a highly aroused dog responds reflexively and its ability to learn suffers. Scent work runs at low arousal by design, which is why it settles dogs that ball-throwing winds up.

For a dog that already hunts with its nose independently — a beagle, most obviously — this is also the honest outlet for a behaviour you are otherwise fighting. That page is beagle training, and the air-scenting version for a ranging gundog is in German shorthaired pointer training.

Do I need a special breed for scent work?

No. Detection work concentrates in a few breeds because of stamina, biddability and structure, not because other dogs cannot smell. Flat-faced dogs and very old dogs are the usual practical exceptions, and even they can do short indoor searches.

What predicts a good nosework dog is willingness to work away from the handler without checking in constantly, and that is trainable in most dogs by simply not helping them.

Can an old or deaf dog do nosework?

It is one of the few activities that suits both. There is no jumping, no impact and no need to hear a cue — the release can be a hand signal or a lead unclipped, and the dog sets its own pace.

For a dog with reduced mobility, shrink the search area rather than the difficulty. Three boxes on the floor of one room is a real session.

How often should we train?

Short and frequent beats long and occasional: five to fifteen minutes, three or four times a week, is enough to see change. Duranton and Horowitz's effect came from a daily activity over two weeks, not from long sessions.

Stop while the dog still wants another search. A dog that has been worked until it gives up has just practised giving up, and that is the one habit in this sport that is genuinely hard to undo.

Every guide we have written is on the dog training guide index.

Sources

  1. Let me sniff! Nosework induces positive judgment bias in pet dogs Applied Animal Behaviour Science 211:61-66 (Duranton & Horowitz), 2019.
  2. When the nose doesn't know: canine olfactory function associated with health, management, and potential links to microbiota Frontiers in Veterinary Science 5:56 (Jenkins, DeChant & Perry), 2018.
  3. Behavioral Problems of Dogs MSD Veterinary Manual (Merck), professional edition, 2026.
  4. Position Statement on Humane Dog Training American Veterinary Society of Animal Behavior (AVSAB), 2021.
  5. How to Choose a Trainer American Veterinary Society of Animal Behavior (AVSAB), 2026.