You saw “SOFR + spread” and froze
The email is short, the numbers are not. “SOFR + 2.75%” sits in the middle of the quote like it’s supposed to be reassuring, and instead it’s the first thing that makes the whole deal feel slippery. You try to anchor it to something familiar—prime, fixed rates you’ve seen online, the last refinance conversation—and it doesn’t quite stick. The lender’s worksheet has a payment example, but the line that matters is the one you can’t translate into a monthly bill without guessing what SOFR will be on the day it resets.
The freeze usually isn’t about math; it’s about control. A floating benchmark suggests a moving target, and the spread looks like the lender’s “take,” but neither tells you what your actual note rate will be next month, or next quarter, or after the interest-only period ends. Add a real constraint—closing in three weeks, a rate-lock fee on the table, or a covenant tied to interest expense—and “SOFR + spread” stops being a definition to learn and becomes a risk to price.
So the first move is to treat that phrase like a placeholder, not a rate. Before you debate whether SOFR will rise or fall, you need to know what, exactly, the quote is pointing to: which SOFR, measured how, applied when, and then padded by what else. That’s the only way the rest of the term sheet starts behaving like something you can compare.
Your first expectation: SOFR is “the rate”
The next instinct is to treat SOFR like a posted number you can check once and move on. If the quote says “SOFR + 2.75%,” it’s tempting to assume SOFR is the objective part—like it’s the “real” rate—and everything else is just the lender’s markup. So you pull up a chart, see today’s SOFR print, add 2.75%, and mentally file away a note rate you can live with. Under a deadline, that shortcut feels efficient: you need to decide, and a single number is easier to commit to than a moving reference.
But SOFR doesn’t behave like a consumer-facing rate you borrow at; it behaves like a daily market reading. That gap matters because your loan doesn’t update continuously with the market. It updates on specific reset dates, using a specified SOFR method, and the note rate you pay is whatever that calculation produces at that moment—plus the spread, plus any rounding rules. When you assume “SOFR is the rate,” the surprise later isn’t that SOFR moved. It’s that the number you watched wasn’t the same SOFR your contract actually uses when the payment changes.
Reality check: SOFR is a market snapshot
Once you stop treating SOFR like a posted borrowing rate, the quote reads differently. SOFR is a daily print of where overnight cash traded in the Treasury repo market—useful because it’s broad and transparent, but still just a snapshot of one corner of the market on one day. It can jump around on month-end or quarter-end balance sheet pressures, and it can drift lower or higher without warning. If you’re trying to decide under a three-week closing clock, that means “today’s SOFR” is a reference point, not a payment estimate.
Your loan documents usually don’t say “whatever SOFR is when you check your phone.” They say the lender will look at a specific SOFR setting (often an overnight-based average), over a defined window, then apply it on a reset date—sometimes with business-day conventions that pull the effective rate forward or back around weekends and holidays. The practical friction is timing: you can watch SOFR every day and still miss what matters, because the note rate changes only when the contract’s measurement window closes.
So the reality check is simple: you’re not forecasting a single number; you’re evaluating a method. And that’s where the quote can quietly diverge from the chart you’re using.
Where your quote diverges: term, lookback, average
The first place borrowers get tripped up is realizing the contract can reference “SOFR” and still land on a different number than the one they pulled from a chart. Some loans use an overnight-based compounded or simple average; others use a “term SOFR” setting that’s published for 1-, 3-, 6-, or 12-month tenors. The practical constraint is comparison: two lenders can both say “SOFR + 2.75%,” but if one resets off 3‑month term SOFR and the other uses a 30‑day average, the starting rate and the way it drifts between resets won’t match.
Then comes the lookback. Many documents bake in a lag—often a few business days, sometimes longer—so the rate used for your new period is based on SOFR observations from an earlier window. That sounds minor until you’re trying to plan cash flow around a reset date or a covenant test. Finally, check the averaging method and rounding. Daily simple average, compounded average, observation shift, and “rounded to the nearest 0.001%” all seem like fine print, but they can change the reset outcome enough to matter when margins are tight.
The mismatch that matters: spread and fees
After you sort out which SOFR the contract actually uses, the quote still has a second moving part: the spread. It’s easy to treat it like a harmless add-on because it’s stated upfront, but it’s the piece that doesn’t float down just because SOFR does. In practice, two offers with the same SOFR method can diverge fast if one has a 2.75% spread and the other has 3.50%, especially when you’re already staring at a payment that’s brushing up against a budget cap.
Then the fees quietly change the “real spread” you live with. An upfront discount point, an origination charge, or a renewal fee on a business line doesn’t show up in the headline rate, but it increases your cost if you refinance, sell, or pay down sooner than planned. The friction is timing: a slightly lower spread can be more expensive if it’s purchased with points you won’t recover. So the comparison that matters is not SOFR-to-SOFR; it’s all-in cost under your likely holding period.
Your safety rails: caps, floors, and resets
By the time the spread and fees are clear, the next anxiety shows up in the margin: what keeps the note rate from running away from you. This is where the “safety rails” matter, because the contract usually limits change per reset and over the life of the loan—but only in the direction the lender agrees to limit. The constraint is usually calendar-driven: a first reset after an intro period can hit before your income or rent plan has caught up, so a cap that looks generous on paper can still produce a payment jump you feel immediately.
Start with the reset schedule and the cap structure. Many loans have periodic caps (how much the rate can change at each reset) and a lifetime cap (the maximum note rate). Then look for a floor, which can keep the benchmark component from falling below a minimum, effectively turning “SOFR + spread” into “no lower than X% + spread” even if SOFR drops. Finally, confirm what “reset” actually means operationally: the rate-setting date, the lookback window, and whether the payment recalculates right away or with a lag. Those details decide whether the rails catch you when the market moves—or only when it moves down.
A borrower’s SOFR checklist for next steps
The term sheet feels less like a riddle once you turn it into a checklist you can work through before you sign or waive contingencies. First, write down the exact SOFR basis in the contract (term SOFR vs overnight average), the tenor, the lookback/observation shift, and the rounding rule—then confirm the date the lender actually “sets” the rate versus when your payment changes. Under a closing deadline, that timing mismatch is where surprises hide.
Next, price the parts that won’t float: spread, points, origination, renewal, and any required deposits or hedging language on a business loan. Ask for an all-in APR (or a fee amortization over your expected holding period), not just a teaser payment. Finally, stress the reset: periodic cap, lifetime cap, and any floor. Have the lender show three scenarios (flat, up, down) using your cap structure, so you can decide based on range, not a single SOFR print.
