How to Make Ice Cream in a Blender

A blender cannot churn, so it does the job in two steps instead of one. A chef explains both methods, the ratio that keeps the base scoopable, and why yours came out icy.

Updated

A white bowl of thick pink raspberry ice cream topped with whole raspberries, on a wooden counter

A blender cannot make ice cream the way an ice cream machine makes ice cream, and every article that skips past this is setting you up to be disappointed by a result that is actually quite good. The gap is mechanical, not a matter of effort or a better recipe. Once you understand exactly where it sits, you can work around it, and what comes out of the jar is genuinely worth making on a weeknight.

I ran a restaurant group’s test kitchen for four years. We had a proper compressor machine, and we also had a lot of evenings where somebody wanted frozen dessert in ninety minutes and the machine was full of something else. The two workarounds below are what we used, and they are the same two methods that every credible source on the internet has independently arrived at. What almost nobody explains is why there are exactly two of them, and why each one produces a different kind of not-quite-ice-cream.

What You Are Actually Making

Three things get filed under “ice cream in a blender” and they behave completely differently.

Real dairy ice cream, frozen without a churn. Cream, sugar, flavor, frozen in a home freezer with the blender doing the aerating or the crystal-breaking. This is what the rest of this article is about.

Nice cream. Frozen fruit, usually banana, blended into a soft mass. No dairy fat, no added sugar, nothing to control crystal growth except the fruit’s own fiber. Good, fast, and not ice cream.

Ninja Creami output. A different machine entirely. It shaves a pre-frozen pint with a downward-driven paddle rather than spinning a blade in a jar, and it gets closer to churned texture than a blender does. It also takes 24 hours and a dedicated appliance.

Why a Blender Cannot Churn

An ice cream machine does three things at the same time, and the simultaneity is the whole trick. It freezes, it agitates, and it aerates, continuously, for twenty to forty minutes. A dasher scrapes newly formed ice off the frozen cylinder wall the instant it appears, so no single crystal ever gets the chance to grow, and the same motion beats air into the mix while it thickens.

That matters because of one number. Your tongue starts detecting individual ice crystals as grit at somewhere around fifty microns. Below that, the mass reads as smooth. A churn keeps crystals well under that threshold by never letting any of them mature, and it holds them apart in a fat-and-air matrix.

A blender can only do those three jobs one after another, and that gives you exactly two options.

You can aerate first and freeze second, which is the whip-and-freeze method. You whip air into cream, then freeze it statically. Crystals grow uninterrupted during that freeze, but they grow inside a base that is already full of insulating air bubbles and fat, which slows and disrupts them.

Or you can freeze first and fragment second, which is the reblend method. You let the base freeze solid, crystals and all, then physically shatter the large crystals into smaller pieces with the blade. This is fragmentation, not the fine nucleation a churn produces, which is why the texture lands close to real ice cream without ever being identical to it.

That distinction is the honest answer to “why doesn’t mine taste like the good stuff,” and it is also why both methods work about as well as each other despite looking nothing alike.

Method 1: Whip and Freeze

This is the one to use if your blender is ordinary, and it is the method America’s Test Kitchen landed on after testing alternatives, including a stand mixer, which incorporates so much air the result goes past ice cream and into frozen whipped topping.

Blend cold heavy cream on medium until it thickens to soft peaks. In a jar blender this takes forty to sixty seconds and you want to stop while it still moves, not when it has seized. Blend in sweetened condensed milk, vanilla, a pinch of salt, and whatever else you are flavoring with, on the lowest speed, just until combined. Every extra second here knocks air back out.

Pour into a shallow metal pan rather than a deep container. Metal pulls heat out faster than glass or plastic, and a shallow layer freezes faster than a deep one, and both of those buy you smaller crystals. Cover the surface with plastic wrap pressed down onto the ice cream, not stretched across the top of the pan, and freeze at least six hours.

No second blending step. That is the point of this method.

Method 2: Freeze Then Reblend

This one gives you real control over flavor, because you can build any base you like rather than working around condensed milk. It also asks considerably more of the machine.

Make your base, chill it thoroughly, then freeze it in an ice cube tray overnight. Cubes rather than a single block, deliberately: a solid brick is close to unblendable and it is where most people burn out a motor. Drop the cubes into the jar with two or three tablespoons of cold cream or milk, pulse, scrape down, pulse again, and keep going until it comes together at soft-serve consistency. Then either eat it immediately, which is my preference, or spread it in a shallow pan and firm it up for two hours.

The scrape-and-pulse rhythm is not optional. A blender vortex will happily carve a cavity around the blade and spin in it while the outside of the load never moves, and pulsing is what collapses that cavity back down onto the blade.

A food processor does this job as well as a blender and sometimes better, because a wide flat blade sweeping the floor of a shallow bowl meets frozen chunks more evenly than a narrow vortex does. If you already own one, use it. If you are choosing between the two machines, our comparison of a food processor versus a blender covers where each one wins, and a combination machine covers both jobs in one footprint.

The Ratio, and Why the Numbers Matter

Nearly every recipe you will find gives you fixed quantities. Almost none explains the two variables you are actually managing, which means you cannot adapt anything.

Fat. Heavy cream is at least 36 percent milkfat. That fat physically obstructs the path between growing ice crystals, which is why a cream base freezes smoother than a milk base at identical sugar. Federal labelling rules require 10 percent milkfat before something can be sold as ice cream at all, and that floor exists for exactly this reason. Below roughly a quarter of your liquid volume in heavy cream, texture falls off fast.

Sugar. Sugar dissolved in water lowers the temperature at which that water freezes. In a properly sugared base, a portion of the water is still liquid at freezer temperature, and that unfrozen fraction is what makes the mass scoopable rather than solid. Cut sugar for health reasons and you will get a brick, every time, and no amount of blending fixes it. Corn syrup is useful here because it depresses the freezing point and adds body without tasting as sweet as the equivalent sugar.

A working starting point for a quart: three cups heavy cream, one cup whole milk, three quarters of a cup of sugar or a half cup of sugar plus a quarter cup of corn syrup, and a half teaspoon of salt. Adjust flavor freely. Adjust those two variables carefully, and in opposite directions if you must, because dropping one while dropping the other compounds.

The Pantry Stabilizer Nobody Mentions

Commercial ice cream uses guar gum and locust bean gum to bind free water so that less of it is available to migrate into existing crystals during storage. You can approximate this with things you already have.

Whisk two teaspoons of cornstarch into a quarter cup of the milk, bring the rest of the dairy to a bare simmer, stir the slurry in, cook for one minute until it thickens slightly, then chill completely before you blend. An extra egg yolk does a similar job through lecithin and protein, and two tablespoons of softened cream cheese is the laziest version and works better than it has any right to.

The difference shows up on day three, not day one. Skip it for something you are eating tonight.

Which Blender Features Actually Change the Result

Wattage matters only for the reblend method, where a thousand watts is the practical floor and twelve to fifteen hundred is comfortable. A tamper matters for the same method and is close to essential, because it lets you push frozen material into the blade without stopping. Pulse control matters for both. Jar shape matters more than people expect: a narrow tapered jar keeps pulling material down toward the blade, while a wide flat-bottomed one lets it sit at the edges.

Two smaller notes from experience. Repeated frozen loads plus high-speed friction heat is the hardest thermal cycle you can put a jar through, so if you are shopping for a glass-jar blender this is the one job where the material genuinely matters. And a full-power blender working on frozen cubes is loud in a way a smoothie never is, which is worth thinking about in an apartment or with a sleeping child in the house. Our roundup of the quietest blenders covers which machines keep the noise down under load, and if you already own a Ninja, our operating guide explains why its stacked blade assembly wants ingredients loaded in the opposite order from every other blender.

How Long It Keeps

Two to three days at its best. A week before it is obviously worse. This is not a preservative issue, it is a crystal-growth issue: every freezer defrost cycle warms the surface a fraction, melts a little ice, and refreezes it onto crystals that are already there. They only ever get bigger.

Shallow container, plastic wrap pressed onto the surface, back of the freezer rather than the door. Small batches, eaten quickly, are the real answer.

Where I Would Start

Make the whip-and-freeze version first, tonight, with whatever blender is on your counter. It is the lower-risk method, it tells you immediately whether you like the texture this approach produces, and it will not stall your motor. If you like it and want control over the flavor, graduate to the reblend method and make sure the machine is up to it before you fill the jar with frozen cubes.

Either way, the thing to hold onto is that you are not making a worse version of churned ice cream through incompetence. You are making a structurally different frozen dessert with the equipment available, and once the fat and the sugar are in the right place, it is very good on its own terms. If you are shopping for the machine to do it with, start with our guides to the best blenders for thick frozen loads and the best blenders for protein shakes, both of which grade the motors and jar shapes that decide whether frozen material actually moves.

Frequently Asked Questions

Can I make real ice cream in a regular blender, or do I need a high-powered one?
It depends entirely on which method you use, and this is the question most articles answer wrongly by giving one answer for both. The whip-and-freeze method asks almost nothing of the motor. You are blending cold liquid cream into a thick foam, which is a light load, and a basic three-speed jar blender does it in under a minute. I have made that version in a machine that cost less than a decent chef's knife. The freeze-then-reblend method is the opposite. You are asking the blade to break apart a solid frozen mass, and that is one of the hardest things you can ask a blender to do. Under roughly a thousand watts, a full jar of frozen cubes will stall the motor, spin a cavity around the blade while the outer material stays untouched, or trip the thermal cutout. If your blender is modest, use the whip-and-freeze method and you will get a better result than someone forcing frozen bricks through an underpowered machine. If you want the reblend method, you want real wattage, a tamper, and a pulse button, and a food processor is a genuine alternative because its wide flat blade handles frozen chunks more evenly than a blender vortex does.
Why is my blender ice cream icy or grainy instead of creamy?
Almost always one of three things, in this order of likelihood. First, not enough fat. Fat physically gets in the way of ice crystals joining up into bigger ones, so a base built on milk or half-and-half freezes harder and coarser than one built on heavy cream, no matter how well you blend it. Second, not enough sugar. Sugar lowers the temperature at which the water in your base freezes, which keeps some of that water liquid at freezer temperature and is the entire reason ice cream stays soft enough to scoop. Cut the sugar and you get a block. Third, a slow freeze. The longer your base takes to pass through the freezing range, the more time crystals have to grow, so a deep bowl of base set in a crowded freezer produces coarser ice than a shallow pan in an empty one. Spread it thin, use the coldest part of the freezer, and do not open the door for the first two hours. If all three are right and it is still grainy, you are almost certainly tasting the honest limit of the method rather than a mistake.
What is the difference between blender ice cream and a Ninja Creami?
They are not the same category of machine and they do not work the same way. A blender spins a small blade fast inside a jar and relies on a vortex to keep pulling material back through the cutting zone, which is why it needs either liquid or a tamper to work. A Creami holds a pre-frozen pint in a fixed housing and drives a spinning paddle down through it from above, shaving the frozen block into fine particles at a controlled rate. That is a much better match for the actual problem, because the block never has to circulate and the paddle contacts every part of it in sequence. The result is closer to churned ice cream than anything a blender produces, and it is why the machine sells. What you give up is a two-hundred-dollar appliance that does exactly one thing, plus a 24-hour freeze before every batch. A blender is the better answer if you want dessert tonight from equipment you already own.
Is banana nice cream the same thing as blender ice cream?
No, and treating them as the same is why people are disappointed by one of them. Nice cream is frozen fruit blended into a soft mass, usually banana because banana has enough pectin and starch to hold a smooth body. It has no dairy fat and no added sugar beyond the fruit's own, so the mechanism keeping it smooth is fruit fiber and pulp holding water in place, not fat blocking crystal growth. It is genuinely good, it is quick, and it needs nothing but a blender and a freezer. It is also unmistakably not ice cream. It melts differently, it coats your tongue differently, and it tastes like the fruit rather than like cream. Make it when you want frozen fruit, which is a perfectly good thing to want. If your blender does this job often, an acai-bowl-grade machine is the same tool, and our roundup of the [best blenders for acai bowls](/best-blender-for-acai-bowls/) covers what actually matters for thick frozen fruit loads.
How long does homemade blender ice cream stay good in the freezer?
Two to three days at full quality, a week before it is noticeably worse, and after that it is edible but grainy. This is much shorter than commercial ice cream and the reason is not preservatives, it is stabilizers and storage temperature. Every time your freezer cycles through its defrost period the surface of the ice cream warms a fraction, a little ice melts, and when it refreezes it joins existing crystals rather than forming new ones. Crystals therefore grow every single day, and a home freezer with a door that opens cycles far more than a commercial holding cabinet at a steady negative twenty. Commercial ice cream fights this with guar and locust bean gum, which bind free water so there is less of it available to migrate. You can slow it at home with a cornstarch slurry or an extra egg yolk, and you can slow it further by pressing plastic wrap directly onto the surface, storing it in a shallow container rather than a deep one, and keeping it at the back of the freezer rather than in the door. Make small batches and eat them.

Related Articles

About the Reviewer

Nora Whitfield

Nora Whitfield, CIA Culinary Arts, ServSafe Certified

A.O.S. Culinary Arts, The Culinary Institute of America

CIA Culinary ArtsServSafe Certified12 Years in Professional Kitchens

Nora Whitfield trained at the Culinary Institute of America and spent twelve years in professional kitchens, the last four running the test kitchen for a regional restaurant group where her job was to decide which equipment could survive a dinner service and which could not. She has seasoned more carbon-steel woks than she can count, boiled water in every kettle worth owning, and broken enough cheap gear to know exactly where manufacturers cut corners. She reviews kitchen equipment the way she specified it professionally: cook on it, clean it, and see what it looks like after a month.