How to Keep Deer Out of Your Garden
What actually works, in the order worth trying it. Most gardens are lost at the ground line and the gate rather than to the jump everyone plans against.
Ceaser Sinclair · July 31, 2026 · 8 min read
There is no single trick. There is a short list of things that work, a shorter list of things that work permanently, and a long list of things that people try first because they are cheap. This page is the map: what to do, in what order, and what to stop wasting money on.
First, make sure it is deer
Deer have no upper incisors. They grip and tear, so browsed stems end in a ragged, stripped stub rather than a clean cut. That single detail settles most cases.
A clean cut at roughly 45 degrees is rabbits or groundhogs. Ragged holes in the middle of leaves are slugs. Whole seedlings vanishing overnight with no stem left is usually rabbits or voles. Bark stripped in vertical scrapes on a young trunk is a buck rubbing his antlers, which is deer but is a different problem from browsing and needs a different fix.
Other confirmations: tracks are two-toed and heart-shaped, roughly two to three inches long. Droppings are clustered pellets. And damage tends to stop at about five feet, because that is where the animal stops reaching.
Identifying deer damage covers this properly. It takes two minutes and it regularly redirects the entire response, because a repellent aimed at rabbits is money spent on nothing.
Where deer actually get in
Almost everyone plans against the jump, and the jump is the least of it.
A deer will go under a barrier far more readily than over one, because going under costs nothing. Any gap over about six inches will be used, per Oregon Department of Fish and Wildlife fencing guidance. The gate is the next failure, either because it is shorter than the fence or because it was left open. The jump comes third, and only matters once the first two are closed.
If you take one thing from this page: close the ground line and the gate before spending anything on height.
Why they are in your garden at all
Worth understanding, because it explains why this is a management problem rather than a problem you solve once.
Deer are edge animals. They feed in open ground and shelter in cover, so the most valuable habitat available to them is the boundary between the two, and a suburban neighbourhood is essentially nothing but edge: lawn against woodland, garden against hedge, over and over. Add irrigated, fertilised, year-round planting and no hunting pressure, and the result is deer densities well above what the surrounding woodland alone would carry.
Two consequences follow, and both are practical.
They are not passing through. A garden that gets browsed once is on a route. Deer are creatures of strong habit, moving on established trails between bedding and feeding areas, and they will return nightly for as long as the reward holds. This is why a single good year proves nothing and why a plant untouched for three seasons can vanish in the fourth.
Nothing you do removes them. Repellents move the decision, fences move the boundary, and better planting removes the reward. None of it reduces the local population, and a garden made unattractive simply moves that pressure next door. That is not an argument against doing it, but it is the reason "keep deer out" is the honest framing and "get rid of deer" is not.
It also explains the seasonal pattern below. Deer pressure tracks the availability of everything else, so your garden matters most to them in the weeks when the countryside offers least. Nothing about your garden changes in February. What changes is the alternative.
One thing genuinely does reduce a garden's appeal at the structural level: removing cover immediately next to it. Deer are reluctant to feed far from an escape route, so an open lawn between the woodland edge and the beds does measurable work, where a hedge running right up to the border does the opposite. That is a landscaping decision rather than a purchase, and it is free if you are already planning the layout.
The four approaches, honestly compared
| Approach | Works | Effort | Cost | Fails when |
|---|---|---|---|---|
| Fencing | Always, if built right | High once | High once | The gate is left open |
| Repellents | Reduces damage substantially | Ongoing | Low, forever | You skip a reapplication |
| Plant selection | Permanently, for what you plant | Low | Low | Pressure is severe |
| Scare devices | Briefly | Low | Low | The deer works out it is harmless |
Most gardens end up using the first three together, and that is the right answer rather than a compromise. Fence the vegetable garden, choose better plants for the ornamental borders, and spray the handful of things you are not willing to give up.
Fencing, the only method that always works
A fence is the only approach that does not depend on how hungry the deer are, which is why it is the expensive answer that never disappoints.
Eight feet is the working height for a single upright fence. Six feet does not work: deer clear it routinely, and it is the most common expensive mistake in this subject because it looks substantial and costs most of what eight feet costs.
If an ordinance or a neighbour caps you lower, two workarounds exploit the fact that deer have poor depth perception. Two parallel fences of four to five feet, set three to four feet apart, work about as reliably as one eight-foot fence because the animal cannot judge the landing. A single fence angled outward at 45 degrees does the same job with less footprint.
The full picture is in the fencing guide, and how tall a deer fence needs to be covers the height question on its own, including what to do on sloping ground, where deer jump from the higher side.
You very likely do not need to fence the property. Perimeter grows much faster than area: a square quarter-acre is about 420 feet of fence, a square acre about 835, so you buy double the fence for four times the land. Fencing the vegetable garden is a small perimeter around high value and is almost always the right first purchase. What a deer fence costs has a per-100-foot shopping list you can price at any supplier.
Repellents, the practical middle
Repellents reduce browsing substantially. They do not stop it, and anything claiming otherwise is selling something.
Two mechanisms, and one is much better. Odour repellents are detected on approach, so the plant is never bitten. Taste repellents work only after a mouthful has been taken, which on a young tree's leading shoot is already the damage you were trying to prevent. Prefer odour.
The consistent finding across university and extension testing is that putrescent egg solids outperform nearly everything else, which is useful because it is also the thing you can make. The homemade egg-based recipe reproduces the active ingredient in the sprays that win those trials.
The part people get wrong is not the product, it is the schedule.
Efficacy falls off over two to four weeks, heavy rain removes most of it at once, and new growth is unprotected the moment it appears. Reapplication is the programme, not an optional extra. Rotate between two or three products with genuinely different active ingredients, because deer habituate to a signal that never changes.
The deterrents guide covers the whole subject, including honest assessments of Irish Spring soap and predator urine, both of which have a real but small effect and are consistently oversold.
Plant selection, the fix nobody starts with
This is the cheapest permanent answer and it is almost always tried last, after several seasons of replacing the same casualties.
Deer avoid plants by smell before they ever taste them, which is why the mint family carries most deer-resistant planting: catmint, salvia, russian sage, lavender, bee balm. Texture does the rest, with felted leaves like lamb's ear and leathery ones like boxwood.
Where to start depends on what you are replacing:
- Shrubs, the most expensive mistake available, because a browsed shrub often never recovers its shape
- Perennials, the most forgiving, since a browsed one regrows from the crown
- Trees, where the real risk is antler rub rather than browsing
- Annuals, the cheap place to experiment
- Shade-tolerant plants, which is the hardest case and the most searched
The honest version: if you have deer and you keep replanting hosta, daylily, rhododendron or arborvitae, no repellent programme will save you. Those are among the worst-rated plants on this site and the pressure will find them.
Scare devices, and why they stop working
Motion-activated sprinklers are the only category with a real claim, because water is a genuine startle and the timing is unpredictable. Even those fade.
Everything else in this category fails for one reason: deer habituate to any signal that never carries a consequence. Lights, noise, reflective tape, radios and ultrasonic devices all produce a few good days followed by nothing. Ultrasonic repellers in particular have no credible evidence behind them, and deer hearing does not work the way the marketing implies.
If you use scare devices at all, move them, and treat them as a bridge while something permanent gets built.
A plan, by budget
Nothing to spend. Confirm it is deer. Stop replacing the plants that keep being eaten. Close any gap at the base of existing fencing and hang the gate so it shuts itself.
A little. Cage the three or four plants actually being destroyed, using welded wire cylinders about five feet tall. Start an egg-based repellent programme on everything else and put the reapplication dates in a calendar.
More. Fence the vegetable garden or the one bed that matters. Small perimeter, high value, and it removes the problem permanently rather than managing it.
Most. Fence the boundary, and only after the above. Budget for a real gate and braced corners rather than a bigger roll of mesh.
A plan, by season
Late winter is the hardest window, when browse is gone and evergreens get stripped. Have protection in place before February rather than after.
Spring is the second. New growth is soft, sweet and low in the defensive compounds a mature plant has. A perennial ignored in July can be eaten to the ground in April, and nursery-fresh plants are at their most palatable the week they go in.
Summer is the easy season. Wild forage is everywhere and your garden is one option among many. This is when to plant and when to build.
Autumn brings the rut, and with it antler rubbing. Every young tree needs a trunk guard on before it starts, because a buck can girdle a two-inch stem in one evening and the tree dies the following summer for no visible reason.
What to do next
If you are starting from nothing, in order: confirm the damage is deer, close the ground line and the gate on any fencing you already have, stop replanting the casualties, and put a repellent schedule on the things you are keeping. Then decide whether the one bed that matters is worth fencing properly.
Everything above has a full guide behind it: fencing, deterrents, and the plant database, which rates 540 plants on the Rutgers scale by how badly deer actually damage them.
If you are weighing which questions this data can and cannot settle, the plant FAQ covers what the ratings mean and why published lists contradict each other.