Self-Watering Pots vs Ollas vs Spikes: A Straight Comparison
Self-watering pots, ollas and terracotta spikes get sold to the same person for the same reason — you would like to stop killing plants with an unreliable watering schedule — and they get talked about as if they were three price points of one product. They are not. They are three different physical mechanisms, and the differences are not cosmetic: one moves water by capillary rise into the mix regardless of whether the plant wants it, one releases water through a porous clay wall in direct response to how dry the surrounding soil is, and one runs on a cord whose flow rate has nothing to do with the plant at all[12][13]. That single distinction decides which plants each one suits, how long it lasts, how it fails, and whether it will quietly poison your rootball with salt. This comparison takes all three apart and tells you which to buy for which situation.
Unglazed terracotta — ollas and spikes — is the only one of the three that is genuinely demand-driven. Water inside the clay sits at roughly zero water potential; the soil outside gets more negative as it dries, and water crosses the clay wall down that gradient. Dry soil pulls harder, wet soil pulls almost nothing. A self-watering pot works by capillary rise from a reservoir, which keeps the mix above the reservoir continuously moist and suits thirsty tropicals but not anything that needs a real dry-down. A wick system is open-loop: flow depends on the cord and the height drop, not the plant. Buy spikes for indoor pots, an olla for large containers and beds, and a self-watering pot only for plants that never want to dry out.
01 · THE SHORT ANSWER
Self-watering pots vs ollas vs spikes: which should you actually buy?
For an ordinary indoor pot holding a plant that likes even moisture, a terracotta spike is the most forgiving option, because its output tracks soil dryness instead of running at a fixed rate. For a large container, a raised bed or a garden row, a buried olla does the same job at a scale that actually matters, and does it with water-use efficiency that is well documented in arid-land agriculture[12]. For a fern, a peace lily or a nursery-style pot of thirsty tropical foliage that genuinely never wants to dry out, a self-watering pot is a reasonable purchase. For a cheap holiday fix on a plant you do not mind gambling with, a wick will do.
The mistake almost everyone makes is treating a self-watering pot and a terracotta spike as competing versions of the same idea — a cheap one and an expensive one, a low-tech one and a high-tech one. They behave differently enough that they suit different plants. A self-watering pot maintains a wet zone; a terracotta emitter maintains a moisture level and then largely stops. Put a snake plant in a self-watering pot and you have built a slow rot chamber. Put the same plant next to a spike and, most of the time, very little water leaves the spike at all, because the soil is not pulling.
None of the three is a substitute for knowing when your plant is dry. What they buy you is tolerance: a wider margin between the day you should have watered and the day the plant notices you did not. If you are still working out what that margin looks like for your plants, start by learning how long each one can genuinely go between waterings before any of this hardware is worth buying.
02 · MECHANISM ONE
How a self-watering pot actually works
A self-watering pot is a reservoir in the base of the container connected to the growing mix either by a fabric wick or by a column of mix that dips down into the water. Water climbs from the reservoir into the mix by capillary action, filling the smallest pores first and moving upward against gravity because the dry mix above holds water at a more negative matric potential than the free water below[13][14]. A fill tube lets you top the reservoir up without disturbing the plant, and an overflow hole stops you flooding it.
It is fair to say the flow responds somewhat to soil dryness, because the gradient across the wick steepens as the mix above dries. But the response is weak and, crucially, it is bounded at the wrong end: the mix immediately above the reservoir stays continuously moist no matter what the plant is doing. That zone never dries. For a plant whose roots want continuous access to water — ferns, peace lilies, calatheas, most nursery-grown tropical foliage — that is exactly right. For anything that evolved to dry out between rains, it is a permanent low-oxygen zone at the base of the rootball, which is where root rot organisms do their work on roots already weakened by anoxia[8].
The failure modes are specific and worth knowing before you buy. Filling the reservoir before the plant has rooted down is the most common: a freshly potted plant has no roots near the reservoir zone, so nothing draws water up, and the mix simply sits wet for weeks. Water from the top for the first few weeks and only start using the reservoir once roots have reached it. Salt accumulation is the structural one: because water moves upward and evaporates or transpires away, the dissolved minerals it carries stay behind and concentrate near the surface, and nothing ever flows downward to carry them out[11]. There is no way to flush a self-watering pot except by top-watering it, which is worth doing thoroughly every month or two — see bottom watering vs top watering for why only downward flow leaches. Reservoir stagnation is the third: standing water that is never emptied goes anaerobic and smells. Let the reservoir run dry for a day or two between refills.
03 · MECHANISM TWO
How ollas and terracotta spikes work — and why it is different
An olla is a buried unglazed clay vessel. A terracotta spike is the same physics in a smaller body, pushed into potting mix. Both hold water inside a porous fired-clay wall, and both work because of a pressure difference the plant itself creates.
Water inside the clay is free water, held at approximately zero water potential. The soil outside, as roots and evaporation remove water from it, develops an increasingly negative matric potential — it holds its remaining water under tension. That difference in potential across the clay wall is the driving force. Water moves through the pore network of the clay from the high-potential side to the low-potential side, which is to say from the reservoir into the drying soil. As the soil around the clay rewets, its potential rises back toward zero, the gradient collapses, and outflow slows to almost nothing[12][13].
This is the single most important technical distinction in the whole comparison, so it is worth stating plainly: as the soil dries, a terracotta emitter delivers more water; as the soil wets, it delivers less. It is a self-regulating, demand-driven system with no moving parts, no timer and no electricity. Bainbridge's review of buried clay pot irrigation documents very high water-use efficiency for exactly this reason — water is delivered into the root zone at the rate the soil can accept it, with negligible surface evaporation and negligible deep percolation[12]. The same principle underlies the low-cost pitcher irrigation methods promoted for smallholder agriculture in water-scarce regions[9].
The honest limits matter as much as the mechanism. Capacity bounds duration: a spike holding around 17.5 fluid ounces will run a small-to-medium indoor pot for several days to a couple of weeks depending on pot size, light and season, and no amount of clever physics extends it beyond what the reservoir holds. Soil contact is essential: if the mix shrinks away from the clay, or you have set the spike in an air pocket, there is no continuous water path and nothing moves. Firm the mix around it and rewet a hydrophobic mix before installing. Mineral scale reduces porosity over months in hard-water areas, as carbonates precipitate in the clay pores; soaking in a dilute white vinegar solution dissolves the deposit and restores flow. Clay cracks if water freezes inside it, so ollas come indoors or get emptied before a hard frost. And a terracotta emitter does nothing useful in soil that is already saturated — if you have a drainage problem, this does not fix it.
04 · MECHANISM THREE
Wick systems, and how all four compare on regulation
A wick system is a cord running from an external reservoir — a jar, a bucket, a bottle — into the pot. Water travels the cord by capillary action, and the rate at which it travels is set by three things: the cord material, its cross-sectional area, and the height difference between the water surface and the point where the cord enters the mix. Note what is absent from that list. The plant. A wick delivers at a rate determined entirely by its own geometry, which makes it the definition of an open-loop system: it will happily keep supplying a pot that is already sodden, and it will just as happily under-supply a plant in a heatwave.
That is not a reason to dismiss it. A wick is nearly free, scales to any number of pots from one bucket, and works fine if you have tested it in advance and know roughly how much it moves per day for your particular cord and drop height. It is the cheapest way to keep a shelf of pots alive for a fortnight. It is simply not self-regulating, and no amount of marketing copy changes that.
01 · Terracotta spike
Output rises as soil dries
Flow tracks the potential gradient across the clay wall. Dry mix pulls hard; rewetted mix pulls almost nothing. Genuinely demand-driven.
02 · Olla
Same response, larger scale
Identical physics in a buried vessel holding litres rather than millilitres. Wets a radius of soil in a bed or big container.
03 · Self-watering pot
Partly responsive, always moist
Capillary rise steepens a little as the upper mix dries, but the zone above the reservoir never dries at all. No true dry-down.
04 · Wick system
Ignores the soil entirely
Rate is fixed by cord, thickness and head height. Open-loop, so it can over-supply a wet pot or under-supply a thirsty one.
05 · HEAD TO HEAD
Duration, capacity, cost and maintenance
Duration and capacity. Every one of these systems is bounded by how much water it holds. A pair of terracotta spikes at roughly 17.5 fluid ounces each will typically carry an indoor pot of six to ten inches for something between five days and two weeks — the range is wide because a bright warm room in July drains a pot far faster than a dim one in January. A self-watering pot's reservoir is usually the largest indoor option and can run two to four weeks on a thirsty plant. An olla buried in a raised bed holds litres and can go a week or two in summer between refills, depending on bed size and crop. A wick runs until the jar empties, at whatever rate the cord dictates.
Cost. Wicks are near-free. Terracotta spikes are inexpensive and last for years with occasional descaling. Self-watering pots cost more than plain pots and lock you into that pot for that plant. Ollas cost the most per unit but cover a large area, which makes them cheap per square foot of bed.
Maintenance. Terracotta needs periodic descaling in hard water and protection from freezing. Self-watering pots need periodic top-watering to leach salts, and the reservoir wants to run dry occasionally so it does not go stagnant. Wicks foul with algae and need replacing. On raw upkeep, spikes are the least demanding of the four.
Salt management is where the three systems separate most sharply, and it is the point buying guides almost always miss. Any method that only ever moves water into a pot — capillary rise from a reservoir, a wick, a clay emitter — leaves behind every dissolved mineral it carries. Fertiliser residue and tap-water carbonates concentrate wherever water evaporates, and high salinity lowers the osmotic potential of the soil solution so roots have to work harder to extract water even from moist mix[11][3]. The symptoms — crisp brown leaf margins, stalled new growth — get misread as underwatering, and people respond by adding more water and therefore more salt. Whichever system you choose, top-water thoroughly every four to six weeks and let the runoff go down the drain.
06 · FAILURE MODES
How each system fails, stated honestly
Nothing here is maintenance-free, and any guide that says otherwise is selling something. These are the four failures that account for most disappointed reviews.
- Self-watering pot filled too early. A newly potted plant has no roots down near the reservoir, so nothing draws water upward and the lower mix stays saturated for weeks. Top-water for the first month, then start using the reservoir once roots have reached that zone.
- Terracotta installed without soil contact. An air gap between clay and mix breaks the continuous water path, and the spike appears to “not work” while staying full. Rewet a hydrophobic mix first, push the spike in firmly, and press the surrounding mix against it.
- Hard-water scale closing the clay pores. Carbonates gradually precipitate inside the pore network and output drops over a season. A soak in dilute white vinegar, then a rinse, dissolves the deposit and restores the flow rate.
- Any system used on a plant that wants to dry out. Continuous moisture around the roots of a succulent, cactus, snake plant or ZZ plant produces the same low-oxygen conditions that let root-rot pathogens take hold. Match the mechanism to the plant, not to the gadget you already own.
A fifth, less obvious failure is trusting a cheap moisture meter to tell you whether the system is working. Inexpensive analogue probes infer moisture from electrical conductivity, which is heavily influenced by dissolved salts and by how well the probe contacts the mix, and two probes in the same pot routinely disagree. If you want to verify that a spike or reservoir is keeping up, lift the pot and judge by weight — see whether moisture meters actually work before spending money on one.
07 · PLANT SUITABILITY
Which plants suit which system
Sort your plants into two groups and most of the decision makes itself. The first group wants consistently moist mix and sulks visibly when it dries: ferns, peace lilies, calatheas and marantas, fittonia, most nursery-grown tropical foliage. The second group evolved with real dry periods and needs the mix to actually dry through before the next watering: snake plants, ZZ plants, cacti, most succulents, and to a lesser degree rubber plants and many philodendrons.
Group one is the only group a self-watering pot genuinely suits. The permanently moist zone above the reservoir is a feature for a fern and a slow disaster for a ZZ plant. Group two should be on terracotta or on nothing — and terracotta works for group two precisely because outflow collapses once the surrounding soil is wet, so the plant gets a slug of water and then a long stretch of gradually falling moisture that looks a lot like a natural dry-down.
Pot material shifts all of this too. An unglazed terracotta pot loses water through its own walls, so a plant in terracotta dries considerably faster than the same plant in plastic or glazed ceramic, and any reservoir system will need refilling more often. That interaction is covered in how pot material changes your watering schedule, and it is worth reading before you assume a system has failed when it has simply been outpaced by evaporation.
08 · THE RECOMMENDATION
What to buy, by scenario
A small indoor pot, four to eight inches. A terracotta spike. The volume is right, the demand-driven release suits almost any foliage plant, and there is no pot to replace. This is the AcquaTerra spike's home ground.
A large indoor floor plant in a ten- to fourteen-inch pot. Two spikes placed on opposite sides of the rootball, refilled together, or a small olla sunk into the mix if the pot is deep enough to bury one. One spike in a big pot wets a limited radius and leaves the far side dry.
An outdoor container or patio pot. An olla, buried with its neck above the surface and capped to keep out debris and mosquitoes. Outdoor evaporation and wind make small reservoirs pointless; you want litres, not millilitres.
A raised bed or vegetable row. Ollas, spaced so their wetting radii overlap slightly — roughly two to three feet apart in typical loam, closer in sand, wider in clay. This is the application the technique was developed for and where its water savings are largest[12][9].
Travelling for two weeks. Water everything thoroughly the day before you leave, move plants out of direct sun to slow transpiration, and add spikes to the pots that will not make it. For the specific arithmetic of matching reservoir volume to trip length, how to water plants while away works through it properly.
A plant that needs to dry out. No reservoir pot, ever. A spike is acceptable because it stops delivering once the mix is wet, but honestly the best answer for a snake plant or a ZZ is to water it properly once a month and leave it alone.
MATCH THE SYSTEM TO THE POT
AcquaTerra spikes hold 17.5 oz each and suit indoor pots. Acqua Olla is the buried version for large containers, raised beds and outdoor gardens.
Shop AcquaTerra →Shop Acqua Olla →Self-watering pots, ollas and spikes are not three grades of the same product. A reservoir pot keeps a zone of mix permanently moist and is the right answer for plants that never want to dry out and the wrong answer for everything else. Unglazed terracotta releases water down a potential gradient the plant itself creates, which makes it the only genuinely self-regulating option of the three, in a small body for indoor pots and a large buried one for beds and outdoor containers. A wick is cheap and open-loop and should be treated as such. Choose by mechanism and plant, top-water periodically to leach whichever you pick, and none of them will surprise you.
Frequently asked questions
Are terracotta spikes better than self-watering pots?
They are better for most plants because their output is demand-driven: water crosses the clay wall faster as the soil dries and almost stops once it is wet. A self-watering pot keeps the mix above its reservoir permanently moist, which suits ferns and peace lilies but harms plants that need a real dry-down.
What is the difference between an olla and a terracotta spike?
They use identical physics at different scales. An olla is a larger unglazed clay vessel buried in a bed or big container and holding litres of water, while a spike is a small version pushed into potting mix and holding ounces. Choose an olla outdoors and a spike for indoor pots.
Why does a terracotta spike release more water when the soil is dry?
Water inside the clay sits at roughly zero water potential while drying soil holds its water at increasingly negative potential. Water moves through the porous clay wall down that gradient, so a steeper gradient means faster release. When the soil rewets the gradient collapses and flow nearly stops.
Can succulents and snake plants go in a self-watering pot?
No. Succulents, cacti, snake plants and ZZ plants need the mix to dry through between waterings, and a reservoir pot maintains a continuously moist zone at the base of the rootball. That low-oxygen zone is where root rot begins. Water these plants by hand instead.
Do self-watering pots cause salt buildup in the soil?
Yes. Water moves upward from the reservoir and evaporates or transpires away, leaving fertiliser residue and tap-water minerals concentrated near the surface with no downward flow to remove them. Top-water thoroughly every four to six weeks and discard the runoff to leach them out.
How long does a terracotta watering spike last before refilling?
A spike holding around 17.5 fluid ounces typically runs a six- to ten-inch indoor pot for five days to two weeks. The range is wide because bright warm rooms and porous terracotta pots empty a reservoir far faster than dim cool rooms and plastic pots do.
Why has my terracotta spike stopped releasing water?
Usually either the soil around it is already wet, in which case it is working correctly, or there is an air gap breaking contact with the mix, or mineral scale has closed the clay pores. Firm the mix against it, and soak the spike in dilute white vinegar to dissolve hard-water deposits.
References
- [1] University of Minnesota Extension. Watering houseplants. extension.umn.edu.
- [2] Penn State Extension. Houseplant Care: Watering, Fertilizing, and Repotting. extension.psu.edu.
- [3] Clemson Cooperative Extension. Indoor Plants — Watering. Home & Garden Information Center. hgic.clemson.edu.
- [4] North Carolina State Extension. Extension Gardener Plant Toolbox — houseplant profiles. plants.ces.ncsu.edu.
- [5] Michigan State University Extension. Watering indoor plants. canr.msu.edu.
- [6] University of Wisconsin–Madison Division of Extension. Houseplant care. hort.extension.wisc.edu.
- [7] University of Florida IFAS Gardening Solutions. Container gardening and irrigation. gardeningsolutions.ifas.ufl.edu.
- [8] University of Illinois Extension. Houseplants — common problems and root diseases. extension.illinois.edu.
- [9] Food and Agriculture Organization of the United Nations. Low-cost micro-irrigation and pitcher irrigation methods. fao.org.
- [10] Oregon State University Extension Service. Watering container plants efficiently. extension.oregonstate.edu.
- [11] Colorado State University Extension. Soluble salts in potting media and leaching. extension.colostate.edu.
- [12] Bainbridge, D.A. (2001). Buried clay pot irrigation: a little known but very efficient traditional method of irrigation. Agricultural Water Management, 48(2), 79–88.
- [13] Hillel, D. (2004). Introduction to Environmental Soil Physics. Elsevier Academic Press.
- [14] Taiz, L., & Zeiger, E. (2010). Plant Physiology, 5th ed. Sinauer Associates.
- [15] Brady, N.C., & Weil, R.R. The Nature and Properties of Soils. Pearson.