Five minutes of walking per month of age, twice a day. A four-month-old gets twenty minutes; a six-month-old gets thirty. It is the most repeated number in puppy ownership, and unlike most of what circulates about dogs, it is not internet folklore. A real organisation publishes it.
That turned out to make the story more interesting, not less.
The rule is real, and nobody who publishes it cites anything
The Royal Kennel Club publishes it today, almost word for word as it gets repeated:
a good rule of thumb is a ratio of five minutes of formal exercise per month of age (up to twice a day)
Blue Cross publishes a version of it too. So does the American Kennel Club, in a phrasing worth reading twice: “Veterinary researchers recommend that puppies can go for walks for about five minutes multiplied by every month of their age.”
Which veterinary researchers? The page does not say. No study, no name, no reference. That is authority by assertion. It is the shape of a citation without one inside it.
None of the three publishers we checked cites a source. The Royal Kennel Club’s own framing is “a good rule of thumb”, which is an honest description of a heuristic and not a claim to evidence.
And one major veterinary charity says so outright. The PDSA:
there’s no scientific evidence behind this rule, and although it might work for some, it’s not appropriate for most puppies.
So the position is not that some websites got it wrong. Serious organisations publish this number, another serious organisation says there is nothing behind it, and none of them shows their working.
The popular version is stricter than the original
Two qualifiers get dropped in transmission, and both matter.
The Royal Kennel Club says “formal” exercise: structured lead walks. Not pottering in the garden, not play, not sniffing about. The version most people meet applies the limit to everything the puppy does.
And it says “up to twice a day”, which is a ceiling, not a prescription. Follow the rule as its own publisher writes it and a four-month-old may have forty minutes of formal walking, and a six-month-old an hour. Most repetitions present a single session as the whole allowance.
One consequence is worth noticing. At five minutes per month twice daily, a twelve-month-old is already on an hour twice a day. That is ordinary adult exercise. The rule only really restricts anything between about two and six months.
What has actually been measured
Start with what the rule is about: minutes of walking, and joints.
Exercise duration has been measured in growing dogs. Twice, that we found. Neither study looked at the skeleton.
Kiviranta and colleagues, 1988 put six young beagles on a treadmill against eight sedentary controls: an hour a day, five days a week, 4 km/h, on a 15° uphill incline, for fifteen weeks, ending at forty weeks of age. The cartilage in their knees came out 19 to 23% thicker than the controls’. Fourteen laboratory beagles, and the outcome was cartilage structure, not hip scores, not lameness.
Hyytiäinen and colleagues, 2026 went further. Fifteen Labrador puppies, eight trained and seven controls, starting at about sixteen weeks. Over eight weeks the trained group built from walks of 1 hour 20 minutes to 2 hours 20 minutes, plus sprints and underwater treadmill work. That is roughly six times what the five-minute rule allows.
They found no adverse short-term effects. They also noted, in their own words, a “lack of assessment of long-term impact to the musculoskeletal system”.
So the honest position is narrower and stranger than “nobody has looked”. People have prescribed real doses to real puppies and measured what happened. What nobody has done is relate how far a puppy walks to whether its hips end up dysplastic.
A systematic review of 62 papers on risk factors for canine joint disease found only four that examined exercise at all. One of those four, Slater 1992, did collect “the usual amounts and types of exercise”. It asked owners of 31 dogs with osteochondritis dissecans and 60 matched controls. It found no link with exercise at all. What it did find was dietary calcium, well water, and playing with other dogs.
The study the rule is usually hung on is Krontveit and colleagues, 2012. It followed 501 dogs across four large breeds, from birth to hip screening at twelve or eighteen months. It is a good study. It measured no minutes, and the paper says why: “Preliminary analyses revealed bias in the owners’ recording regarding exercise. The exercise variables were therefore reclassified from categorical to dichotomous variables.”
That sentence is worth sitting with. The researchers tried to capture graded exercise data from owners and found it too unreliable to use. Not “we didn’t think of it”, but “we tried, and owners cannot report this accurately.”
| What was measured, birth to 3 months | Odds of radiographic hip dysplasia |
|---|---|
| Daily use of stairs | 1.96 (95% CI 1.14–3.35) |
| Daily off-lead exercise in park terrain | 0.31 (95% CI 0.15–0.65) |
Odds are relative to puppies not doing that activity daily. The paper also found being born on a farm protective (OR 0.34), and significant clustering within litters. That is a reminder that much of this is settled before the puppy comes home.
Its conclusion is both a restriction and a permission, and it usually gets quoted by halves. Puppies under three months “should not be allowed access to stairs”. The same sentence says they “should be allowed outdoor exercise on soft ground in moderately rough terrain”.
The stairs half is the one clear caution here. The other half is what the five-minute rule sits awkwardly against.
The stated reason has no evidence in dogs
The rule is nearly always justified the same way: exercise damages growth plates, the soft zones near the ends of young bones where new length is added.
We could not find a single canine study showing that ordinary exercise damages them. Not one.
What we found instead is absence in places you would expect presence. The AAHA’s 2019 canine life stage guidelines run to twenty-four pages and do not contain the words “growth plate” or “physis” anywhere. A 2024 clinical review of puppy exercise written for veterinary professionals cites fourteen studies and never mentions growth plates either. The WSAVA publishes twelve sets of global guidelines and none of them is about exercise.
Growth-plate injuries in dogs are real and well documented, but as fractures, from trauma. That is a different claim from walking too far.
The two studies above that did prescribe a measured dose to growing dogs both found no harm. One found benefit. Neither examined growth plates directly, so that is an absence of evidence rather than a clearance. A companion paper to the beagle study reports in its title that long-distance running reduced cartilage content. So there is probably a dose above which loading stops helping. We could not read its methods, so we are not guessing where that sits.
The cost of the restriction, which nobody counts
Here is the part that makes this more than a debunking.
The rule bites hardest on the youngest puppies: at eight weeks, five minutes twice a day. That window collides with something that is well evidenced. The AVMA’s literature review on socialisation puts it plainly:
Puppies are most responsive to learning from exposure to unfamiliar dogs between the ages of 3 to 14 weeks.
And on what happens if they miss it: puppies kept from such exposure until after fourteen weeks “lose such flexibility and may be forever fearful in these situations”.
Three to fourteen weeks is exactly when the five-minute rule permits five to seventeen minutes a day. A rule with no evidence behind it is restricting a puppy during the one developmental window the evidence is clear about.
Two large Finnish surveys point the same direction in adult dogs. Less activity and less participation in things go with more fearfulness. Read those carefully, though: they are owner questionnaires, cross-sectional, and a fearful dog plausibly gets walked less rather than the reverse. They are a reason for concern, not proof of it.
So what should you actually do
Nothing here licenses marching a fourteen-week-old Great Dane round the block for an hour. “No evidence for the limit” is not “no limit”. Nobody has measured the harm side of restriction in puppies either, and an absence of evidence points in both directions at once.
What the evidence does support is a different kind of answer. It is about the shape of the exercise, not the minutes:
- Avoid stairs in the first three months. This is the one clear, measured finding, and the effect was close to a doubling of risk.
- Favour soft, varied, uneven ground over hard flat surfaces. Also measured, and protective.
- Let the puppy set the pace and stop. A puppy that has had enough lies down. That signal is better than a timer, because it accounts for the dog in front of you, the terrain and the day.
- Do not trade the socialisation window for the walk limit. Exposure to the world between three and fourteen weeks is the best-evidenced thing in this whole article, and it can be had in short, calm, carried outings if the dog is small or the ground is hard.
- Big dogs finish later. The major long-bone growth plates close at roughly eight to twelve months in a medium dog and ten to fourteen in a large one, earlier in small breeds. Early neutering can push that later still. If you want to know which end of that your puppy sits at, adult size is what drives it. Our calculator estimates that.
If you want numbers rather than principles, one named veterinary source gives them. Writing in the Veterinary Ireland Journal, Menzies suggests sessions of 15 to 30 minutes at two to four months, rising to 45 to 60 minutes at four to six months. Endurance training waits until the growth plates close.
Be clear what that is. It is one clinician’s judgement in a peer-facing journal, and it is much more generous than the five-minute rule. It is not a measured finding either. But someone put their name to it.
If your puppy has a diagnosed joint problem or you are managing a breed with a high dysplasia burden, that is a conversation with your vet and not a rule of thumb from anyone, including us.
What would change this answer
A cohort study that recorded exercise in minutes rather than as yes-or-no activities, and followed those dogs to skeletal maturity, would settle in one paper a question that has been answered by assertion for decades. The Krontveit cohort shows the design is achievable; it simply measured a different variable.
Failing that, any of the three organisations publishing the rule could name the research behind it. We would update this page the day one did.
What we could not verify
We could not date the rule. The Wayback Machine and Google Books full-text search were both unreachable from our tooling, so we cannot say when the Royal Kennel Club first published it or whether the wording has changed. The earliest publication date we could actually evidence anywhere is Blue Cross’s, 18 April 2019. That is certainly not the origin, only the oldest dated page we could open.
We could not find its origin at all. Every organisation publishing it does so without a source, and every veterinary professional we found writing critically about it also failed to trace one. That is a search that stopped, not a search that came back empty.
We could not retrieve the high-dose running studies. Two companion papers to the beagle study above appear to describe the other end of the dose curve. We have their titles and nothing else, so no figures from them appear here.
We could not read the one cohort that recorded puppy walk duration at scale. A 2022 paper from the Generation Pup study looked at walk frequency and duration in dogs up to fifteen months. The journal is closed to our tooling, so we cannot say whether it merely describes what owners do or relates it to any outcome. If it does the latter, part of this page needs revising, and we would rather flag that than leave it unsaid.
We could not check BSAVA. Their manuals are paywalled, so we cannot say whether British small-animal practice guidance contains a numeric recommendation.
And nobody has measured the harm side of restriction. The socialisation evidence is about exposure, not exercise volume, and the fearfulness studies are in adult dogs. We think the collision between the rule and the socialisation window is the most important thing on this page. We also think it has not been tested directly, and we would rather say so.