Training Treats: How Many Is Too Many?

There are two kinds of dog owners – the ones who count training treats like calories at a wedding, and the ones who hand them out like confetti and wonder why the labrador is 3kg over by winter. Both miss the point. Training treats aren’t the problem. The volume, the size and the order in which you stop using them – that’s where most owners trip up. We’ve worked with hundreds of dogs in group classes around the country, and the same question lands on the mat every week: how many is too many?

Training treats should make up no more than 10% of your dog’s daily calories. For a 10kg dog that’s roughly 40 to 50 calories a day, or about 20 to 30 pea-sized pieces of soft, low-calorie reward. Trim a little kibble from the dinner bowl on heavy training days, and you’ll keep the rewards working without padding the waistline.

Why this matters more than owners think

A treat is just a paycheck. Useful, motivating, expected – and easy to overdo. The dog who gets 40 treats by 9am isn’t learning faster; in most cases, they’re learning that food appears whether they offer behaviour or not. Worse, they’re creeping past their daily calorie budget five days a week, and that’s how a vet visit turns into a body condition score conversation. The 10% rule isn’t a guideline pulled from thin air. It’s the figure most vet bodies, including the AVA, point to when discussing reward-based methods. Stay under it and treats stay valuable. Go past it and the rewards lose their bite, the dog gains theirs.

If you’re still pinning down which method suits your dog, our broader training methods guide is the place to start – this article assumes you’re already on the positive-reinforcement bus and just want to know how many tickets to print.

What actually counts as a training treat

A training treat is small, soft and quick to eat. That last bit matters more than the brand on the packet. If your border collie spends 12 seconds chewing, the moment’s gone and the reward stops marking the behaviour. We’re after pea-sized pieces – around 5 to 8mm – that disappear in under two seconds.

Aussie shelves carry plenty of options that fit the brief. Prime100 single-protein rolls cut into tiny cubes; ZIWI freeze-dried lamb or venison snapped into thirds; Scratch jerky torn into slivers; freeze-dried liver from Petbarn or PETstock for the high-value moments. None of these are mandatory – a fingertip of cold roast chicken from last night’s dinner is often the best treat in the bag.

Two distinctions worth keeping clear:

• Low-value rewards: plain kibble, a sliver of carrot, dried fish skin. Use for known behaviours – sit on the back step, hand target in the kitchen.
• High-value rewards: anything smelly, fatty or unusual. Use for new behaviours, distractions, or recall when there’s a kookaburra in the gum tree.
• Daily food, not extra: on busy training days, weigh out half the dinner kibble in the morning and use it as the day’s rewards. The maths takes care of itself.
• Non-food rewards: a 15-second tug game, a release to sniff a fence post, a thrown ball. These don’t count toward the 10% rule at all – and most owners under-use them.

The number, in plain English

Most adult dogs need somewhere between 25 and 30 calories per kilogram of body weight per day, give or take, depending on age and activity. That’s a rough rule – your vet will give you a tighter figure for your dog. The 10% ceiling sits on top of that.

Here’s the maths most owners avoid:

Dog weight	Approx. daily calories	10% treat budget	Pea-sized pieces (approx.)
5kg	220 cal	22 cal	10 to 15
10kg	400 cal	40 cal	20 to 30
20kg	700 cal	70 cal	35 to 50
30kg	900 cal	90 cal	45 to 65

Figures are general averages for a healthy adult dog – puppies, working dogs and seniors all sit outside this. Check with your vet.

Most commercial training treats run between 1 and 5 calories per piece. The very small, soft ones – the ones built for clicker work – sit at the lower end. So a 10kg cavoodle can absorb 20-odd treats in a 10-minute session and still have headroom for the rest of the day. A 30kg kelpie can take 50. The catch: most owners forget that the dental chew after dinner, the half-Schmacko at the park and the chunk of cheese in the morning all eat into the same budget.

How to run a session without blowing the budget

Tomorrow morning, before the kettle’s boiled. Five steps, 10 minutes.

1. Weigh out 30 grams of your dog’s normal kibble into a small bowl. That’s your free training currency – calories you’d have fed anyway. Subtract the same amount from the dinner serve.
2. Cut 8 to 10 high-value pieces – a strip of cold chicken, a square of liver treat – into pea-sized chunks. These are your jackpot rewards for new behaviours and hard distractions.
3. Pick one behaviour. Just one. Sit, hand touch, name recognition, the door wait – whichever is closest to wobbly. Don’t try to teach three things in one session; in most cases dogs learn faster from depth than spread.
4. Run 5-minute rounds with a 30-second sniff break in between. Mark the moment you want with a marker word (‘yes’) or a clicker, then deliver the treat within two seconds. Late timing is the single most common reason positive reinforcement appears ‘not to work’.
5. Finish before the dog’s bored. End on a behaviour they nailed, a quick scratch, no big farewell. The treats that didn’t get used go back in the dinner bowl, not the bin.

Bribery vs reinforcement – the bit most blogs miss

This is where the wheels come off for a lot of owners, and it’s the reason the 10% conversation matters at all. Bribery is showing the dog the treat to get the behaviour. Reinforcement is delivering the treat after the behaviour, as a consequence.

If your dog sits because they can see the cheese in your fingers, that’s bribery. If they sit because sit has earned cheese 200 times before, that’s reinforcement. The first one collapses the minute the cheese isn’t visible. The second one holds up in the rain at Bondi at 6pm.

The practical fix: hide the treats. Off your body, in a tin on the bench, in your pocket but in a wrapped foil pouch. The dog should not be able to predict whether you have food. That single change – making the reward unpredictable – is what shifts treats from a payment scheme to an actual training tool. And it’s the reason most dogs do not, in fact, need treats forever.

Five misunderstandings owners keep walking into

Common mistakes – we’ve all made at least one:

• Treats forever. After a behaviour is fluent in three environments, you start fading. Variable schedules first – reward every third or fifth correct response – then real-life rewards (a door opening, a ball thrown, freedom to sniff). Most dogs are off food rewards for known behaviours within six to twelve weeks.
• Treat size creep. Owners start with pea-sized pieces and end up with thumb-sized chunks because they’re easier to cut. By week three the labrador’s on 90 calories of cheese a day. Stay small.
• Reward timing past three seconds. If you fish around in your pocket while the dog drifts off, you’re reinforcing drift, not sit. A marker word buys you the time – it tells the dog ‘yes, that, the treat is coming’.
• Confusing soft with safe. Soft semi-moist treats are often higher in calories per gram than dry biscuits. Read the back of the packet at least once.
• Skipping non-food rewards entirely. For most working breeds – kelpies, border collies, mals – a thrown tug toy after recall is a higher-value reward than chicken. Use it. Saves the calorie budget for hard moments.
• Counting only the training treats. The bone after dinner, the bit of toast at breakfast, the dental chew – all of it eats into the 10%. Most owners under-count by half.
• Switching brands too often. Soft, semi-moist treats from a different protein each fortnight can upset sensitive stomachs. Pick two or three trusted options and stick with them.

What if your dog isn’t food motivated

First, hedge – most dogs flagged as ‘not food motivated’ are food motivated for the wrong food, in the wrong environment, at the wrong intensity. A border collie who turns her nose up at kibble in the kitchen will hoover diced kabana at the dog park. Try the upgrade before the diagnosis.

Second, check the context. A dog who’s been fed within the last hour, who’s anxious, or who’s in pain will refuse food they’d normally take. None of those are training problems – they’re feeding-schedule, behaviour or vet problems. Sort those first.

Third – some dogs really aren’t treat-led. For them, the reward menu is play (a tug, a flirt pole, a fetched ball), access (the back door opening, the lead going on, the harness coming off), or contact (a chest scratch, a quiet ‘good dog’). The 10% rule doesn’t apply to any of these. The timing rules still do – deliver within two to three seconds, or it’s not the reward you think it is.

Training treats in the Australian context

Two things shape how we train here that the US-centric blogs skip. The first is the climate. From November through March, training outdoors past 9am is rough on dogs – not just heat stroke risk on bitumen above 35°C, but soft treats melting in your hand inside two minutes. We move sessions to before 8am or after 7pm in summer, and switch to drier, crumblier reward formats in the heat.

The second is who trains the trainers. Look for a certified handler who’s done the work – Pet Professional Guild Australia members and Delta-accredited trainers commit in writing to reward-based methods only. The AVA position backs that approach for a simple reason: aversive tools have measurable welfare costs and no measurable training advantage.

If you want a deeper dive into the science behind reward-based work, the AVSAB statement is the most-cited summary in English. It’s American, but the research it leans on is international.

Treats aren’t the opposite of rules

The biggest myth about reward-based training is that it means ‘no boundaries’. It doesn’t. Reinforcement-based training is precise about which behaviours earn what. The rules are still there – sit before the door opens, four paws on the floor before the lead clips on, settle on the mat before dinner lands. The treats just make the rules clearer to the dog than yelling at them ever could.

That’s why training treats work as a method, not a bribe. They’re a way of telling the dog, in the only language they reliably parse, ‘this thing you just did – do that again’. The number stays under 10%. The rules stay firm. The dog gets clearer, not softer.

When to call in help

If you’re working a new puppy through foundation behaviours, a group class within the first 16 weeks is worth more than any blog. For reactivity, resource guarding, or anxiety – things that don’t respond to more treats – see a vet behaviourist before a trainer. Both Sydney University and the University of Queensland run referral behaviour clinics.

Frequently Asked Questions

Count the pieces, not the packets. Most owners who think they’re overfeeding training treats are actually under-counting everything else – the dental chew, the toast crust, the half-sausage at the BBQ. Get the 10% number right across the whole day, deliver the treats inside two seconds, hide them off your body, and the dog learns faster on fewer rewards than the dog drowning in cheese.