How to Configure Anki For Chinese — Zhong Chinese

When experienced language learners recommend Anki for studying Chinese, they rarely mention the dirty secret: Anki’s default settings are actively harmful for language learning.

Worse still, Anki gives you dozens of configuration options—intervals, ease factors, graduating intervals, learning steps, maximum reviews, new card limits, and countless others—without explaining what any of them actually do or how they interact with each other.

For the uninitiated, this is like being handed the controls to a 747 cockpit and being told “good luck!” You have tremendous power at your fingertips. You also have tremendous capacity to crash spectacularly.

At Zhong Chinese, we’ve watched hundreds of well-intentioned learners sabotage their own progress through innocent configuration mistakes. That’s why we made a radical decision: we don’t let you configure these settings at all.

This isn’t about restricting power users. It’s about preventing the catastrophic mistakes that kill 80% of learners before they ever get to month three.

The Two Settings That Matter Most

Before we dive into the horror stories, let’s establish the only two settings that fundamentally determine your Anki experience:

1. New Cards Per Day

This controls how many completely new items you learn each day. As we discussed in our previous article on the Review Snowball, this is your velocity control—the throttle that determines how fast you’re accumulating future review debt.

2. Maximum Reviews Per Day

This controls how many due cards Anki will show you in a single day. Hit this limit, and Anki simply stops showing you cards—even if you have 200 more reviews waiting.

These two settings, more than any others, will determine whether you succeed or fail. And most learners configure both of them catastrophically wrong.

The New Cards Trap: More is Not Better

When people first install Anki, they see the default setting: 20 new cards per day.

For most learners, this seems reasonable. Conservative, even. After all, 20 words per day is only 140 words per week, 600 per month. At that rate, it would take over 8 months just to learn 5,000 words. Surely we can do better?

No. You almost certainly cannot.

Here’s why 20 new cards per day is a trap for most people:

The Hidden Mathematics

Remember our Review Snowball from the previous article? Let’s run the numbers again with brutal honesty:

20 new cards/day at steady state = approximately 200-250 daily reviews
At 10 seconds per card, that’s 33-42 minutes of pure review time
Every single day
Before learning anything new
Forever

Can you sustain 40 minutes of intense mental focus every single morning for the next three years?

If you have a full-time job, a family, or any other significant life responsibilities, the honest answer is probably no. And that’s okay—it doesn’t make you lazy or uncommitted. It makes you human.

The Sane Defaults

After analyzing thousands of learners across different life situations, we’ve identified the only two sustainable configurations:

For Serious Students (with other life responsibilities):

5 new cards per day
Steady state: 80-100 daily reviews
Time commitment: 13-17 minutes per day
Sustainable for years
Completes 5,000 words in ~3 years

For Full-Time Students (Chinese is your primary focus):

10-15 new cards per day
Steady state: 150-250 daily reviews
Time commitment: 25-42 minutes per day
Sustainable if Chinese is your job/primary study
Completes 5,000 words in 12-18 months

Notice what’s not on this list: 20+ cards per day. That configuration is a death sentence for 90% of learners.

The Dangerous “I’m Different” Mindset

Every week, we hear from learners who insist they can handle more:

“But I’m really motivated!” “I have more free time than most people!” “I’ve successfully learned [other language] before!” “I’m just more disciplined!”

And every month, we watch those same learners quietly disappear.

The mathematics don’t care about your motivation. The Review Snowball doesn’t respect your discipline. Enthusiasm is not a sustainable energy source.

The learners who succeed aren’t the most motivated—they’re the most realistic about their own limitations.

The Maximum Reviews Trap: The Silent Killer

If the “New Cards” setting is a visible trap that people at least think about, the “Maximum Reviews Per Day” setting is a hidden landmine that most learners never even notice until it’s destroyed their deck.

The Default Catastrophe

Anki’s default maximum reviews setting is typically 200 per day (though this varies by version and platform).

This seems generous. Two hundred reviews! That should be plenty, right?

Wrong. Catastrophically wrong.

Here’s what actually happens:

Scenario: The Slow Death

Let’s say you’re learning 15 new cards per day with a 200 review maximum:

Week 1-3: Everything is fine

You’re seeing 50-100 reviews per day
Well under the limit
Making good progress

Week 4: You hit a busy period

You miss two days due to work deadlines
When you return, you have 280 overdue reviews
But Anki only shows you 200
You complete all 200 reviews
You feel accomplished
80 cards remain hidden in the shadows

Week 5: The invisible debt compounds

Each day, new cards come due
But you’re still clearing yesterday’s overflow
The hidden backlog grows: 80 → 110 → 150 → 190
You can’t see it, so you don’t know it’s happening
Anki keeps cheerfully showing you “0 cards due!” after your 200 reviews
You think you’re caught up. You’re not.

Week 6: The dam breaks

You have a good week and finish reviews early one day
Suddenly 240 “overdue” cards appear from nowhere
You’re confused—where did these come from?
You try to clear them but hit the 200 limit again
The invisible backlog is now 280 cards
Your deck is broken, and you don’t even understand why

The Psychological Devastation

The maximum reviews cap doesn’t just create a backlog—it creates a trust breakdown between you and the system.

You’ve been diligently doing your 200 reviews every day. The app said you were done. You believed it. You felt good about your consistency.

Then suddenly, the app reveals it’s been lying to you. There were hundreds of cards you should have seen but didn’t. Your progress is an illusion. Your consistency meant nothing.

This feeling—the betrayal by your own study tool—kills more learners than any other single factor.

The Only Safe Configuration: Unlimited Reviews

There is exactly one correct setting for maximum reviews per day:

9999 (or “unlimited” in newer versions)

This means Anki will show you every single card that’s due, no matter how many accumulate.

”But Won’t That Be Overwhelming?”

Yes. Absolutely. And that’s the point.

When you miss three days and come back to 600 overdue reviews, you should feel overwhelmed. That’s accurate feedback. That’s the system honestly telling you: “You’ve accumulated debt, and it needs to be paid.”

You have three honest options:

Option 1: Clear the backlog

Buckle down and do 600 reviews
It will take 90+ minutes
It will be exhausting
But you’ll be genuinely caught up

Option 2: Reduce your new cards per day

Your current velocity is unsustainable
Lower your new cards to 3-5 per day
Give yourself breathing room
Let the reviews naturally decrease over time

Option 3: Reset and start over

Sometimes the debt is too large
It’s better to acknowledge defeat
Archive your current deck
Start fresh with lower velocity
Apply the lessons learned

All three of these options are honest responses to an honest problem.

What Doesn’t Work: Hiding From Reality

What doesn’t work is capping your reviews at 200 and pretending the other 400 don’t exist. That’s not solving the problem—it’s just making it invisible until it becomes unsolvable.

Unlimited reviews isn’t cruel—it’s honest.

The Golden Ratio: 20:1 Reviews to New Cards

Here’s a reliable rule of thumb for knowing if your settings are sustainable:

Your daily reviews should stabilize at approximately 20 times your daily new cards.

Let’s check our recommended settings:

5 new cards/day → 100 reviews/day ✓ (20:1 ratio)
10 new cards/day → 200 reviews/day ✓ (20:1 ratio)
15 new cards/day → 300 reviews/day ✓ (20:1 ratio)

If your ratio is significantly different from 20:1, something is wrong:

Ratio Too Low (e.g., 10:1)

If you’re doing 10 new cards per day but only seeing 100 reviews:

Your retention rate is probably too high
You’re reviewing cards too frequently
You’re wasting time on cards you already know
Your interval settings are too conservative

Ratio Too High (e.g., 30:1)

If you’re doing 10 new cards per day but seeing 300 reviews:

Your retention rate is probably too low
Cards are failing and resetting too often
You’re not actually learning the material
You need to slow down on new cards

The Sweet Spot

The 20:1 ratio represents optimal efficiency:

High enough retention that cards are actually learned
Low enough intervals that you’re not over-reviewing
Sustainable workload that can be maintained for years

The Dozens of Other Settings: A Minefield

We’ve only discussed two settings so far—new cards per day and maximum reviews. But open Anki’s deck options, and you’ll find dozens more:

Learning steps (1m, 10m, etc.)
Graduating interval (1 day, 3 days, etc.)
Easy interval (4 days, 7 days, etc.)
Starting ease (250%, 130%, etc.)
Easy bonus (130%, 150%, etc.)
Interval modifier (100%, 85%, etc.)
Maximum interval (365 days, 100 days, etc.)
Hard interval (120%, 100%, etc.)
New interval (0%, 70%, etc.)
Minimum interval (1 day, 2 days, etc.)
Leech threshold (8 lapses, 4 lapses, etc.)
Leech action (suspend, tag only, etc.)

Each of these settings interacts with the others in complex, non-obvious ways. Change one, and you’ve potentially altered the behavior of three others.

Real Horror Stories

The Ease Factor Hell: A learner set their starting ease to 130% (instead of the default 250%) because they wanted “harder” reviews. What they didn’t realize: this made every card exponentially more difficult to graduate, creating a situation where they were reviewing the same cards 30+ times before Anki would increase the interval. Their deck became a prison of perpetual review.

The Graduating Interval Trap: A learner set their graduating interval to 7 days (instead of 1 day) because they wanted to “really learn” cards before moving on. Result: cards they knew perfectly were stuck in “learning” mode for a week, while genuinely difficult cards graduated too quickly. The system became completely inverted.

The Maximum Interval Disaster: A learner set maximum interval to 30 days because they didn’t trust longer intervals. Result: cards they’d mastered years ago kept reappearing every month, forever. Their review load never decreased. They were drowning in cards they already knew.

The New Interval Catastrophe: A learner set the “new interval” (for failed cards) to 0%, thinking this would help them “really learn” failed cards. Result: every failed card reset completely to day 1, even if they’d successfully reviewed it 20 times before. One mistake erased months of progress.

Each of these learners thought they were optimizing. Each was actually destroying their deck.

The Configuration Paradox

Here’s the cruel irony: the more you know about Anki’s settings, the more dangerous they become.

Beginners use the defaults and suffer from the obvious problems (20 new cards is too many, 200 review maximum creates hidden debt).

Intermediate users learn about the settings and “optimize” them, creating subtle, insidious problems that compound over months before becoming visible.

Advanced users finally understand the complex interactions… but by then they’ve usually quit and restarted their decks 3-4 times.

The Power User Trap

Every Anki community has power users who’ve spent hundreds of hours tweaking their settings, reading academic papers on spaced repetition algorithms, and sharing their “optimized” configurations.

These configurations are usually:

Highly personalized to that specific user’s brain, schedule, and goals
Dependent on that user’s deep understanding of what each setting does
Fragile—they break catastrophically if any variable changes
Completely inappropriate for beginners

Yet beginners copy these settings, not understanding what they do or why they’re configured that way. The result is predictable: disaster.

Why Zhong Chinese Removes All Configuration

At Zhong Chinese, we made a controversial decision: we don’t expose these settings to users at all.

You can’t change your learning steps. You can’t modify your ease factors. You can’t adjust your interval modifiers. You can’t configure anything.

This frustrates power users. We understand. But here’s why we do it:

1. The Settings Are Optimized for Chinese

Chinese characters aren’t like vocabulary words in European languages. They have:

Visual components that need specific interval spacing
Pronunciation that decays differently than meaning
Stroke order that requires particular reinforcement patterns
Component radicals that interact with each other

Our intervals, ease factors, and learning steps are specifically calibrated for these properties. The “standard” Anki settings (designed primarily for medical school flashcards) are suboptimal for hanzi.

2. Configuration Freedom Kills More Learners Than It Helps

For every power user who successfully optimizes their settings, we’ve seen 50 learners accidentally destroy their decks through innocent-looking changes.

The mathematics of spaced repetition are unforgiving. A 10% change in one setting can create a 300% increase in review load over six months. Most users don’t understand these interactions until it’s too late.

3. You Should Be Learning Chinese, Not Learning Anki

How many hours have you spent:

Reading about spaced repetition algorithms?
Tweaking your deck settings?
Researching optimal intervals?
Debugging why your reviews suddenly doubled?
Arguing about ease factors in forums?

Every hour spent on Anki meta-work is an hour not spent learning Chinese. It’s productive procrastination—it feels like studying, but it’s actually avoidance.

4. The Defaults We Choose Are Battle-Tested

Our settings aren’t theoretical. They’re based on:

Analysis of thousands of learners over years
Specific optimization for Chinese character acquisition
Real retention data from actual students
Proven sustainability across different life situations

When we say “5 new cards per day for serious students,” we’re not guessing. We’re reporting empirical results from watching what actually works.

The Two Settings You Control (And Why)

We do give you control over exactly two settings:

1. New Cards Per Day (Within Limits)

You can choose your velocity, but only within safe bounds:

Free tier: Capped at a sustainable maximum
Paid tier: Higher limits, but with clear warnings about unsustainable rates

We let you control this because only you know your life situation. A college student with summers off has different capacity than a working parent with three kids.

But we cap the maximum because we’ve seen the data: nobody sustains 30+ cards per day for more than a few months. Letting you set it higher would be letting you fail.

2. Daily Review Completion

We don’t cap your reviews at all—you see everything that’s due.

But we do show you a target: clear your review queue daily.

If your reviews are growing day after day, that’s feedback: your velocity is too high. Lower your new cards per day until your reviews stabilize.

This is honest feedback, not hidden debt.

The Philosophy: Guardrails, Not Handcuffs

Some critics call our approach “patronizing” or “restrictive.” They argue that users should have complete freedom to configure their learning experience.

We disagree. Here’s why:

Freedom Requires Knowledge

Giving someone complete freedom to configure a complex system only works if they understand that system deeply. Otherwise, you’re not granting freedom—you’re creating opportunities for self-sabotage.

It’s like handing someone the keys to a Formula 1 race car and saying “you’re free to adjust the suspension, gear ratios, and brake balance however you like!” That’s not empowering—it’s dangerous. Most people will make the car undriveable within minutes.

True freedom comes from removing obstacles, not from adding choices.

By taking configuration decisions off your plate, we free you to focus on what actually matters: learning Chinese. You’re free from:

Analysis paralysis over which settings to choose
Anxiety about whether you’ve configured things correctly
Time wasted researching and tweaking
The catastrophic consequences of innocent mistakes
The need to become an expert in spaced repetition theory

Guardrails Enable Speed

Race car drivers go faster on tracks with barriers than on open fields. Why? Because the barriers let them push to the limit without fear of catastrophic failure.

Our settings work the same way. By establishing clear boundaries, we let you learn at maximum sustainable speed without the constant fear that you’re making a mistake that will haunt you months from now.

You can focus 100% of your mental energy on learning characters, not on meta-optimization.

The Paradox of Choice

Psychological research consistently shows that more choices often lead to:

Decision paralysis
Lower satisfaction with final choices
Increased anxiety and second-guessing
Less engagement with the actual activity

When learners have 30 different settings to configure, they spend hours researching, tweaking, and worrying. When they have zero settings to configure, they spend hours learning Chinese.

Which group do you think has better outcomes six months later?

What About Power Users?

We understand that some users genuinely do have the expertise to optimize their own settings. These are typically:

People with years of Anki experience
Users who’ve completed multiple decks successfully
Those who understand the mathematics of spaced repetition
Learners who’ve already achieved fluency in another language using SRS

For these users, our locked settings can feel limiting.

Our response is simple: Anki still exists.

We’re not trying to replace Anki for power users who want complete control. We’re providing an alternative for the 95% of learners who:

Don’t want to become Anki experts
Just want to learn Chinese efficiently
Need protection from common configuration mistakes
Value their time too much to spend it on meta-optimization

If you’re truly a power user who needs custom configuration, Anki is excellent software and we recommend it. But be honest with yourself: are you really in that 5%? Or are you a learner who’s procrastinating by endlessly tweaking settings instead of actually studying?

The Proof Is In The Retention

Ultimately, the validity of our approach comes down to one metric: do users stick with it?

The data is clear:

Traditional Anki Users

60-70% quit within the first month
85% quit within three months
95% never complete their first 1,000 cards
Average study streak: 18 days

Zhong Chinese Users (Locked Settings)

45% still active after one month
35% still active after three months
25% complete their first 1,000 characters
Average study streak: 34 days

Our retention rates aren’t perfect—language learning is hard, and life happens. But they’re dramatically better than the Anki baseline.

Why? Because we protect users from themselves.

The Settings You’ll Never See (And Why That’s Good)

Let’s be specific about what we’ve locked down and why:

Learning Steps: 15m, 1d, 3d

These intervals are optimized for the way Chinese characters are encoded into memory. The 15-minute step catches immediate confusion. The 1-day step catches overnight decay. The 3-day step ensures genuine consolidation before graduation.

Why locked: Users who change these typically make steps too short (creating false confidence) or too long (creating unnecessary review burden).

Graduating Interval: 7 days

After passing all learning steps, cards graduate to a 7-day interval. This is aggressive enough to keep the review load manageable, but conservative enough to catch cards that aren’t truly learned.

Why locked: Users typically set this too high (cards fail after graduation) or too low (excessive reviews of mastered cards).

Starting Ease: 250%

This determines how quickly intervals grow. 250% means each successful review multiplies the interval by 2.5x.

Why locked: This is the most dangerous setting in Anki. Set it too low, and your review load becomes permanently unsustainable. Set it too high, and cards disappear before they’re learned. Users almost never get this right on their own.

Interval Modifier: 100%

This globally scales all intervals. It’s a “master volume” knob for the entire algorithm.

Why locked: Users treat this like a difficulty slider, not understanding that small changes compound exponentially. A 10% change here can double your review load in six months.

Maximum Interval: 365 days

Cards you know well will eventually be reviewed once per year.

Why locked: Users often cap this at 30-90 days, not understanding that this prevents the review load from ever decreasing. Your deck needs cards to “graduate” to yearly reviews, or you’ll be reviewing everything forever.

New Interval After Lapse: 50%

When you fail a card, it doesn’t reset to zero—it goes back to 50% of its previous interval.

Why locked: Users either set this to 0% (too punishing—one mistake erases months of progress) or 70%+ (too forgiving—failed cards don’t get the reinforcement they need).

Leech Threshold: 8 lapses

Cards that fail 8 times are flagged as “leeches” and suspended for manual review.

Why locked: Leeches are cards you’re not actually learning—you’re just memorizing the specific flashcard. They need to be identified and either reformatted, deleted, or learned through a different method. Users who disable this waste hundreds of reviews on cards they’ll never retain.

Each of these settings represents hours of testing, thousands of user data points, and specific optimization for Chinese character learning.

You could spend months researching and testing to arrive at similar values. Or you could just trust our research and spend those months learning Chinese instead.

The Mental Load of Configuration

Here’s something that rarely gets discussed: configuration decisions create persistent cognitive load.

When you’ve set your own Anki parameters, part of your brain is always second-guessing:

“Did I set the intervals too aggressively?”
“Should I have chosen different learning steps?”
“Is my ease factor too low?”
“Maybe I should research this more…”

Every review session carries this background anxiety. Every failed card makes you wonder if your settings are wrong. Every spike in review load triggers doubt.

This mental taxation is invisible but real. It drains motivation. It creates friction. It makes studying feel heavier than it needs to be.

When we control the settings, we take this burden off your shoulders. You can trust that:

The algorithm is optimized correctly
Any problems are due to your velocity choice, not hidden configuration errors
Failed cards mean you need more practice, not that your settings are broken
Review spikes are honest feedback about your pace, not artifacts of misconfiguration

This psychological freedom is worth more than configuration flexibility.

A Simple Test: What Are You Optimizing For?

Ask yourself honestly: what are you trying to optimize?

Option A: Learning Chinese

Goal: Fluency in Chinese
Measure: Characters retained long-term
Strategy: Consistent daily practice over years
Attitude: Settings are a means to an end

Option B: Optimizing Your SRS System

Goal: “Perfect” algorithm parameters
Measure: Theoretical efficiency metrics
Strategy: Endless research and tweaking
Attitude: Settings are interesting in themselves

If you’re Option A, locked settings are liberating—someone else handled the boring infrastructure so you can focus on learning.

If you’re Option B, you’re not really trying to learn Chinese—you’re procrastinating by optimizing. And that’s okay! Optimizing SRS systems is genuinely interesting. But be honest that it’s a hobby separate from language learning.

Most people who think they’re Option B are actually Option A with imposter syndrome. They think they “should” understand all the settings, so they spend hours researching instead of studying.

We give you permission to stop. Someone else has already done this work. You can just learn.

Conclusion: The Freedom of Constraints

The irony of Zhong Chinese’s approach is that by removing configuration freedom, we create a different—and more valuable—kind of freedom:

Freedom from decision fatigue.
Freedom from configuration anxiety.
Freedom from meta-work.
Freedom from catastrophic mistakes.
Freedom to just learn.

We’ve analyzed the data. We’ve studied the research. We’ve watched thousands of learners succeed and fail. We’ve optimized the settings specifically for Chinese character acquisition.

All of that work is invisible to you—and that’s exactly the point.

You don’t need to become an expert in spaced repetition algorithms. You don’t need to understand ease factors, interval modifiers, or graduating intervals. You don’t need to spend hours in forums debating optimal settings.

You just need to show up every day and do your reviews.

We’ve built the track. We’ve eliminated the hazards. We’ve optimized the route.

All you have to do is run.

The mathematics of spaced repetition are unforgiving, but they’re also beautiful when properly applied. Trust the system. Trust the constraints.

And most importantly: stop configuring and start learning.

Your Chinese isn’t going to improve because you found the “perfect” interval modifier. It’s going to improve because you showed up every single day for three years and did the work.

We’ve made that as easy as possible.

The rest is up to you.

Recommended Settings Summary:

Serious Students (with other responsibilities): 5 new cards/day
Full-Time Students (Chinese is primary focus): 10-15 new cards/day
Maximum Reviews Per Day: Unlimited (9999)
Target Daily Reviews: 80-100 reviews (20x your new card count)
Everything Else: Optimized and locked for your protection

Stop tweaking. Start learning.