Take a flat 8% nominal rate and ask what it actually yields in a year. There is no single answer. Compounded once a year it yields exactly 8%. Compounded monthly it yields 8.300%. Compounded daily it yields 8.328%. Same headline number, three different real returns, and the only thing that moved was how often the interest got added back to the balance. That single dial, the compounding frequency, is the most overlooked input in any rate comparison, and it is the one that quietly decides whether two quotes you are holding side by side are actually comparable.
Most people treat the nominal rate as the rate and stop there. But the nominal rate is a per-period rate wearing an annual costume. That 8% is really "0.6667% per month, twelve times over" or "0.0219% per day, 365 times over," and the slicing changes the outcome because each slice earns interest on the interest the earlier slices already added. The frequency isn't a footnote. It's half the calculation.
What the frequency number actually is
In the standard effective-yield formula, the frequency shows up as n, the number of compounding periods in a year:
APY = (1 + APR/n)^n − 1The APR gets divided into n equal slices, each slice compounds, and the whole thing is annualized. Crank n up and you cut the rate into thinner slices that compound more often, which pushes the effective yield higher. So the question "what is the APY?" has no answer until you have answered "what is n?"
The eight values that show up in most converters are not arbitrary. They map to real billing and crediting conventions that financial products actually use:
| Frequency | n | Where it shows up |
|---|---|---|
| Annually | 1 | Some bonds, simple-interest CDs |
| Semi-annually | 2 | Treasury bonds, many corporate bonds |
| Quarterly | 4 | A lot of CDs, some money-market accounts |
| Monthly | 12 | Most savings accounts, mortgages |
| Semi-monthly | 24 | Some payroll-linked products |
| Bi-weekly | 26 | Bi-weekly mortgage payment plans |
| Weekly | 52 | A handful of high-frequency savings promos |
| Daily | 365 | Credit cards, most online savings accounts |
The reason daily and monthly dominate the list is that they match how institutions keep their books. A savings account that credits interest monthly is literally running n = 12. A credit card that posts a daily periodic rate is running n = 365 (or 360, depending on the issuer's day-count convention, which is its own small rabbit hole). Picking the frequency is not a preference. It is a fact about the specific product, and getting it wrong is how a comparison goes sideways.
The spread, laid out
Here is the full fan-out for that 8% nominal rate across every frequency:
| Frequency | n | Effective APY |
|---|---|---|
| Annually | 1 | 8.0000% |
| Semi-annually | 2 | 8.1600% |
| Quarterly | 4 | 8.2432% |
| Monthly | 12 | 8.3000% |
| Semi-monthly | 24 | 8.3143% |
| Bi-weekly | 26 | 8.3154% |
| Weekly | 52 | 8.3220% |
| Daily | 365 | 8.3278% |
The first jump is the big one. Going from annual to semi-annual buys you 16 basis points. Quarterly buys another 8. By the time you're arguing about whether interest compounds weekly or daily, you're fighting over six thousandths of a percent. Steep at the start, nearly flat at the end. That pattern is worth keeping in your head, because it tells you when the frequency matters and when it's just marketing.
The ceiling nobody mentions
Push n toward infinity and the formula does not run off to infinity with it. It converges. The limit of (1 + r/n)^n as n grows without bound is e^r, Euler's number raised to the rate. That is continuous compounding, the theoretical case where interest is added in infinitely small increments at every instant.
For our 8% example, continuous compounding yields e^0.08 − 1, which is 8.3287%. Daily compounding already yielded 8.3278%. The difference between compounding 365 times a year and compounding an infinite number of times a year is less than one thousandth of a percentage point. Daily is, for all practical consumer purposes, the ceiling. This is why no savings account advertises "hourly compounding" as a feature: there is almost nothing left to squeeze. The continuous-compounding limit was worked out by Jacob Bernoulli in the 1680s while studying exactly this question of interest, and it is the reason e exists as a named constant at all.
The practical takeaway: anyone selling you on the magic of more-frequent compounding past the monthly-to-daily range is selling you basis-point dust. The real money is in the rate itself, not the slicing.
Where the wrong dial actually bites
The frequency stops being academic the moment you compare two products that compound differently. That's the common and expensive mistake.
Say one bank quotes a CD at 5.00% APR compounding monthly and another quotes 5.05% APR compounding daily. Eyeball the nominal rates and the second looks better by 5 basis points. Convert both to APY at their real frequencies and the monthly product lands at 5.116% while the daily product lands at 5.179%. The daily one is still ahead, but by more than the nominal rates let on, because daily compounding piled on extra. Now flip it: 5.05% monthly versus 5.00% daily. The nominal rate still favors the first, the monthly product yields 5.169%, the daily product yields 5.127%, and the first one wins by a wider margin than the headline implied. Which one wins isn't the lesson. The lesson is that you can't know until you've put both on the same frequency-aware footing.
Credit cards are where this gets uncomfortable. A card quoting 19% APR almost always compounds daily, because issuers post a daily periodic rate. Run 19% through n = 365 and the effective annual cost is 20.92% APY, nearly two full points above the number on the disclosure. The APR is the honest legal figure, but it is not the number describing what a carried balance actually costs you over a year. If you assume the 19% is your real annual cost, you are off by an amount that compounds against you every month.
When the rate is already the yield, leave it alone
The inverse trap is double-compounding a number that's already been compounded. APY is, by definition, the post-compounding figure. If a savings account advertises 4.60% APY, that 4.60% already has the frequency baked in. Feed it back into the (1 + r/n)^n formula as if it were a nominal rate and you inflate it a second time, producing a number that means nothing.
The two formulas are inverses of each other for exactly this reason:
APY = (1 + APR/n)^n − 1
APR = n × ((1 + APY)^(1/n) − 1)If you have a nominal rate and want the yield, use the first. If you have a yield and want the underlying nominal rate (to compare against a loan quoted as APR, say, or to feed into a calculator that expects nominal input), use the second. The frequency n has to match the product in both directions. Run a 5.00% APY back through the reverse formula at daily compounding and you recover a 4.879% nominal APR, the rate that, compounded daily, produces that yield. Mismatch the frequency and both directions lie to you.
This is the quiet failure mode in a lot of free online calculators: they expose a single rate field with no frequency control, silently assume annual compounding or none at all, and hand back a number that is right only by accident. The frequency dial is not a complication to be hidden. It is the thing that makes the answer true.
If you want to watch one nominal rate fan out across all eight frequencies, or convert a yield back to its underlying rate at the right n, Parify does both directions live and prints the formula it used underneath the result, so the frequency you picked is never a hidden assumption.