The Algorithm at a Glance

Every FICO score — and roughly 90% of U.S. lending decisions use FICO — is computed from five weighted categories pulled from your credit reports at Equifax, Experian, and TransUnion. The weights have remained consistent since FICO publicly disclosed them, and understanding the exact percentages gives you a roadmap for where to spend your optimization effort.

Here's the breakdown:

The math is straightforward: the top two factors — payment history and amounts owed — account for 65% of your entire score. If you're prioritizing score improvement, that's where the data says to focus. For a deeper look at how the scoring algorithm actually processes your data, we've broken that down separately.

Factor 1: Payment History (35%)

Payment history is the single heaviest variable in the FICO algorithm. It answers one question: Do you pay your obligations on time?

How the Algorithm Processes It

FICO doesn't treat all late payments equally. The model evaluates three dimensions of delinquency:

• Severity: 30-day late, 60-day, 90-day, 120-day, charge-off, collection, bankruptcy. Each tier carries progressively heavier penalties.
• Recency: A late payment from 6 months ago damages your score far more than one from 5 years ago. The algorithm applies a decay function — most scoring impact fades significantly after 24 months.
• Frequency: One isolated late payment is treated differently than a pattern of delinquencies. The model identifies behavioral trends.

Real Impact Data

According to FICO's own published data, a single 30-day late payment can cause the following drops:

ScoreNerds Data Point: Higher scores suffer larger point drops from the same derogatory event. A single 30-day late payment can cost a 780-score consumer 90-110 points, while the same event costs a 680-score consumer only 40-60 points (FICO published data). The algorithm penalizes deviation from established patterns more heavily — a person with 15 years of perfect payments who misses one has deviated sharply from their historical behavior, and the model flags this as a significant risk signal.

What Moves the Needle Most

• Autopay everything. According to a 2025 Consumer Financial Protection Bureau (CFPB) report, consumers who use autopay for at least minimum payments are 89% less likely to have a 30-day delinquency. This now includes BNPL installments — Affirm and Klarna report missed payments to the bureaus under FICO's new BNPL models, making Buy Now Pay Later a real payment history factor in 2026.
• Negotiate goodwill removals. If you have an isolated late payment on an otherwise clean history, a goodwill letter to the creditor can sometimes get it removed. This is especially effective for first-time lates on long-standing accounts.
• Late payments fall off after 7 years. But their scoring impact diminishes long before that — most of the damage is concentrated in the first 24 months.

For anyone recovering from missed payments, our guide on how to improve your credit score walks through the fastest documented recovery strategies.

Factor 2: Amounts Owed (30%)

This factor is commonly reduced to "credit utilization," but it's actually broader than that. FICO evaluates your total debt burden across every account type.

How the Algorithm Processes It

The amounts-owed calculation includes multiple sub-metrics:

• Per-card utilization: Balance divided by credit limit on each revolving account individually.
• Aggregate utilization: Total revolving balances divided by total revolving credit limits across all cards.
• Installment loan balances: How much you still owe relative to original loan amounts (loan-to-original-amount ratio).
• Number of accounts with balances: More accounts carrying balances signals higher risk, even if individual utilization is low.

The key insight: FICO evaluates both per-card and aggregate utilization. Having one card at 80% and three cards at 0% still hurts you, even if your aggregate utilization is low. The algorithm flags high individual-card utilization as a risk factor independently.

Real Impact Data

Data from Experian's 2025 consumer credit review shows a clear correlation between utilization brackets and average scores:

Notice something interesting: 0% utilization scores lower than 1-9%. The algorithm actually rewards a small amount of active usage. Carrying zero balance across all cards can signal inactivity, which the model treats as less informative than light, consistent usage.

What Moves the Needle Most

• Target the 1-9% utilization sweet spot. We ran the numbers in our utilization sweet spot experiment — keeping reported balances between 1-9% of your limit consistently produces the highest scores in this category.
• Pay before the statement date. Most issuers report your balance on the statement closing date, not the payment due date. Paying down before the statement closes is how you control what utilization gets reported.
• Request credit limit increases. Increasing your limit without increasing spending mechanically reduces your utilization ratio. A $5,000 limit bumped to $10,000 cuts your utilization in half instantly.

This is the fastest factor to optimize. Unlike payment history (which takes months or years to rebuild), utilization resets every billing cycle. You can produce a meaningful score increase within 30 days by managing reported balances. We've documented the exact process in our credit score experiments series.

Factor 3: Length of Credit History (15%)

Credit age is the "set it and forget it" factor — the one that rewards patience and penalizes credit newbies. At 15%, it carries meaningful weight but is the hardest to actively optimize.

How the Algorithm Processes It

FICO evaluates three time-based metrics:

• Age of oldest account: The longer your oldest account has been open, the better. This is the single most influential sub-metric in this category.
• Average age of all accounts: Calculated across every open account. Opening a new card drops this average, which is why churning can hurt scores.
• Age of newest account: A very recently opened account signals you're actively seeking credit, which introduces a mild risk signal overlapping with the "new credit" factor.

Real Impact Data

FICO's research indicates that consumers with a credit history of 25+ years score an average of 40-50 points higher in this category than those with 5-year histories, all other factors being equal. According to Experian data from 2025, the average age of credit for consumers with 800+ FICO scores is 11.4 years, compared to 6.2 years for those in the 650-699 range.

The practical threshold: most scoring benefits plateau around the 7-year mark for average account age. Beyond that, additional age produces diminishing returns — the algorithm's sensitivity to credit age follows a logarithmic curve, not a linear one.

What Moves the Needle Most

• Never close your oldest card. Even if it has an annual fee, consider downgrading to a no-fee version from the same issuer. Closing it removes it from your average age calculation (on VantageScore immediately; on FICO after the account falls off your report in ~10 years).
• Become an authorized user. Being added to a parent's or spouse's old account can instantly boost your average account age. The original account's opening date typically gets inherited on your report.
• Stop opening accounts you don't need. Every new account dilutes your average age. If your average age is 8 years across 4 accounts, opening a fifth brand-new account drops it to 6.4 years.

Factor 4: New Credit (10%)

This factor measures how aggressively you're seeking new credit — and the algorithm treats a burst of applications as a risk signal.

How the Algorithm Processes It

FICO tracks two primary signals in this category:

• Hard inquiries: Each credit application triggers a hard inquiry on your report. FICO counts inquiries from the last 12 months in its scoring, though they remain visible on your report for 24 months.
• Recently opened accounts: New accounts opened in the last 6-12 months count against you here. The more accounts recently opened, the higher the perceived risk.

Rate shopping protection: FICO includes a deduplication window for certain loan types. Multiple inquiries for mortgages, auto loans, or student loans within a 14-45 day window (depending on the FICO version) count as a single inquiry. This lets you comparison shop without being penalized. Credit card applications are not protected by this window — each application counts separately.

Real Impact Data

According to myFICO's published research, a single hard inquiry typically reduces your score by 3-5 points and recovers within 12 months. However, the cumulative effect compounds: consumers with 6+ inquiries in the past 12 months are statistically 8x more likely to file for bankruptcy than those with zero inquiries, which is why the algorithm treats inquiry stacking as a significant risk signal.

We tested this directly — check our hard inquiry real impact experiment for the actual before-and-after numbers across different score ranges.

What Moves the Needle Most

• Space out credit applications. Wait at least 3-6 months between credit card applications. This lets each inquiry age past the peak scoring impact window.
• Use prequalification tools. Most major issuers offer soft-pull prequalification checks. These do NOT generate hard inquiries and give you reasonable approval odds before you formally apply.
• Bundle rate shopping. If you're shopping for a mortgage or auto loan, do all your applications within a 14-day window to benefit from the deduplication window.

Factor 5: Credit Mix (10%)

Credit mix is the smallest factor by weight, and it's the one most people should spend the least energy trying to optimize. But understanding it matters — especially because bad advice about it circulates widely.

How the Algorithm Processes It

FICO evaluates the diversity of your credit portfolio across account types:

• Revolving credit: Credit cards, store cards, home equity lines of credit (HELOCs).
• Installment loans: Auto loans, personal loans, student loans — fixed payments over a set term.
• Mortgage accounts: Home loans carry particular weight in the credit mix evaluation.
• Open accounts: Charge cards (like some Amex products) that require full monthly payment.

The algorithm rewards having a diverse mix of account types because it demonstrates you can manage different forms of credit responsibly. However — and this is critical — FICO has explicitly stated that consumers should not open accounts solely to improve their credit mix.

Real Impact Data

ScoreNerds Data Point: Data from the Federal Reserve Bank of New York's 2025 Household Debt and Credit Report shows that consumers with both revolving and installment accounts have average FICO scores 15-25 points higher than those with only one account type, after controlling for other factors. Having a mortgage in the mix adds an additional 10-15 point advantage on average. However, credit mix matters most for thin-file consumers — for those with 10+ years of history and strong payment records, the marginal contribution of credit mix drops to near-zero.

But here's the nuance: credit mix matters most for "thin file" consumers — people with limited credit histories. For consumers with 10+ years of history and strong payment records, the credit mix factor contributes only marginally to score calculations. The algorithm essentially gives it less marginal weight when other factors are robust.

What Moves the Needle Most

• Don't take on debt to "improve" credit mix. Taking out a personal loan you don't need to add an installment account is a net negative — the hard inquiry and potential interest costs outweigh the minimal scoring benefit.
• Credit-builder loans are the exception. If you have a thin file with only credit cards, a small credit-builder loan ($500-$1,000) from a credit union can add an installment account at minimal cost. The interest is typically under $30 total.
• Let mix develop naturally. Over time, most consumers accumulate a diverse mix through normal financial life — student loans, auto loans, credit cards, eventually a mortgage. Forcing it is rarely worth the cost.

For recommendations on which credit cards pair well with your current score range and help build a healthy mix, see our best credit cards by score guide.

FICO vs VantageScore: Factor Weight Comparison

While FICO dominates lending decisions (used by 90% of top lenders according to FICO's own disclosures), VantageScore — developed jointly by the three credit bureaus — is increasingly used for credit monitoring tools and some lending decisions. The two models evaluate similar categories but weight them differently.

The bottom line: Payment history dominates both models. Utilization is the second-biggest lever in both. If you optimize for FICO, you'll score well on VantageScore too — the rank ordering of factors is nearly identical, even if the exact weights differ. For a complete walkthrough of how both algorithms process your data, see our how credit scoring works deep dive.

Where to Focus First: An Optimization Framework

Not all factors are equally actionable. Here's how we'd rank them by a combined score of weight (how much it affects your score) x speed (how fast you can move the needle) x control (how much is within your power to change):

Notice that amounts owed ranks above payment history for optimization despite having a lower weight. That's because utilization is the most rapidly actionable factor — you can change your reported utilization in a single billing cycle, while payment history takes months or years to build or repair.

This prioritization framework is what drives our credit score improvement guide. We focus on the factors where effort-to-impact ratio is highest, not just raw weight. Before you start optimizing, though, make sure you're not operating on bad assumptions — we tested 12 of the most common credit score myths and several turned out to be costing people points.

How the Five Factors Are Evolving in 2026

The core five factors remain the foundation of credit scoring in 2026. But three developments are reshaping how they interact:

BNPL Payments Now Feed Into the Algorithm

FICO launched dedicated BNPL scoring models (FICO Score 10 BNPL and 10T BNPL) in late 2025. For the first time, Buy Now, Pay Later payment data affects your credit score. BNPL primarily impacts two factors: payment history (on-time vs. missed BNPL installments) and new credit (though FICO groups multiple BNPL loans together rather than counting each as a separate inquiry). FICO simulations show most BNPL users see approximately ±10 points of change.

Medical Debt No Longer Weighs on "Amounts Owed"

The CFPB's rule eliminating most medical debt from credit reports (effective 2025) removed a significant drag on the amounts-owed factor for approximately 15 million Americans. Medical collections that previously inflated your derogatory marks and compressed your score are now invisible to the algorithm. This is the single largest population-level factor shift in recent years.

ScoreNerds Data Point: According to FICO's inaugural Credit Insights report (2025), the average American's credit utilization jumped to 36.1% — well above the scoring penalty threshold and the highest level since pre-pandemic. This means the amounts-owed factor (30% of FICO) is actively suppressing scores for the majority of consumers. Dropping utilization below 10% remains the single highest-ROI optimization for most Americans in 2026.

Trended Data Adds a Sixth Dimension

FICO 10T evaluates your 24-month balance trajectory — not just your current snapshot. While this is technically distributed across the existing five factors (primarily amounts owed), it functions as a new behavioral dimension. Consumers whose balances are trending downward score higher than those with flat or rising balances, even if both have the same current utilization. This rewards consistent debt reduction over time, changing the optimization calculus for factor #2.

Key Takeaways

• 65% of your score comes from just two factors: payment history (35%) and amounts owed (30%). Master these two and you've covered the majority of the algorithm.
• Utilization is the fastest lever. It resets every billing cycle. A 50-100 point swing is possible in 30 days by managing reported balances.
• Payment history is the heaviest but slowest to fix. Prevention (autopay) is vastly more efficient than repair. One late payment takes 18+ months to fully recover from at high scores.
• Credit age rewards patience. You can't hack time. Keep old accounts open, avoid unnecessary new accounts, and let your history compound.
• FICO and VantageScore agree on priorities. Payment history and utilization dominate both models. Optimizing for one effectively optimizes for both.
• Don't over-optimize low-weight factors. Taking on unnecessary debt to "improve credit mix" is almost always counterproductive. The 10% weight doesn't justify the cost.
• BNPL is now a factor. On-time BNPL payments build credit; missed ones damage it. For the 45 million Americans using BNPL regularly, this is new territory in 2026.

Frequently Asked Questions