Chasing a "done" look pushes cooks toward one of two mistakes: stopping too soon and risking illness, or cooking well past the safe point into dry, tough meat (USDA FSIS). A hamburger patty that reads 160°F on a thermometer is safe no matter its interior color. Pink or brown, on its own, proves nothing (USDA FSIS).

This isn't a rare slip, either. USDA Under Secretary for Food Safety Dr. Emilio Esteban has cited the department's own consumer research showing that plenty of people skip handwashing or skip the thermometer altogether (USDA cookout guidance). Guessing by eye is a default habit for a lot of cooks, not an occasional shortcut.

These numbers don't shift based on cooking method, cut, or how confident the cook feels standing at the grill. They're minimums, not cooking times, and the 145°F figure for whole cuts already accounts for a three-minute rest before serving.

Most harmful pathogens die somewhere between 140°F and 165°F, which is why the targets cluster in that range instead of sitting at one universal number (USDA FSIS). Ground meat carries a higher target, 160°F, than a whole cut from the same animal (USDA FSIS).

Prerequisite: an instant-read probe thermometer, either thermocouple or thermistor style. Read the manufacturer's instructions before using it for the first time (USDA FSIS).

1. Know the target before food goes near the heat. Check the table above and settle on the right number for the cut. Deciding in advance beats trying to remember a target while food is already on the grill.

2. Take the reading near the end of the estimated cooking time. Thermocouple and thermistor probes are built for a quick final check, not for sitting in food throughout the cook (USDA FSIS). If the reading falls short of the target, put the food back on the heat and test again a few minutes later, rather than assuming it will finish on its own.

3. Insert the probe into the thickest part of the food, away from bone, fat, or gristle. Those spots heat differently than the surrounding meat and throw off the reading (USDA FSIS). For roasts specifically, place the probe midway into the thickest section, still clear of bone and fat pockets (USDA FSIS).

4. For burger patties and other thin cuts, insert the probe through the side so the tip reaches the center instead of skimming the surface (USDA FSIS; USDA cookout guidance). If using a dial thermometer instead of an instant-read model, insert it 2 to 3 inches into the food (USDA FSIS).

5. Check large or irregularly shaped food, such as a roast or a whole turkey, in several places, and check every individual piece separately. A good reading on one steak says nothing about the steak next to it on the grill (USDA FSIS).

6. If cooking stuffed whole poultry, check the center of the stuffing itself. It must also reach 165°F, and USDA recommends skipping the stuffing when possible (USDA FSIS).

Not every thermometer behaves the same way. Thermocouple-style instant-read probes reach and display a final reading in about 2 to 5 seconds, the fastest category available (USDA FSIS). Thermistors work on a similar principle and are also meant for a final check rather than continuous monitoring during cooking (USDA FSIS).

Pop-up timers, the kind built into turkeys and roasting chickens since 1965, work differently. Made from food-grade nylon, they're reliable to within 1 to 2°F (USDA FSIS). Disposable temperature indicators are another option, built from materials the FDA has approved for direct food contact (USDA FSIS). As a category, most food thermometers land within 2 to 4°F of the true temperature, close enough for home cooking when placement and technique are right (USDA FSIS).

One caution worth repeating: a standard probe shouldn't sit in food while it cooks unless the manufacturer specifically rates it for that. Some models have plastic components that melt near heat (USDA FSIS).

Checking accuracy is simple. Fill a glass with ice, add water, and submerge the thermometer's stem for at least 30 seconds without lifting it out. It should read 32°F. A second check: dipped in boiling water at sea level, it should read 212°F (USDA FSIS). For cleaning, wash with hot, soapy water or follow the manufacturer's instructions; most models shouldn't be fully immersed the way the stem is during calibration (USDA FSIS).

A thermometer confirms doneness. It does nothing to undo contamination that spreads from dirty hands, a cutting board, or a countertop before food ever hits the pan. USDA recommends washing hands with soap for 20 seconds, both before and after handling raw meat or poultry (USDA cookout guidance).

Timing matters too. Bacteria multiply fastest between 40°F and 140°F, a range USDA calls the "Danger Zone." Food should be eaten or refrigerated within two hours of cooking, or within one hour if it's 90°F or hotter outside; food left out longer can turn unsafe even if it was cooked correctly (USDA cookout guidance).

Poultry deserves the most caution of all. USDA estimates roughly 125,000 chicken-associated and nearly 43,000 turkey-associated Salmonella illnesses occur in the U.S. each year, a figure the agency cited in 2024 when it proposed new contamination standards for raw poultry (USDA). That rule was still a proposal as of that announcement, but the gap between chicken and every other meat category is exactly why the 165°F poultry target and multi-point checks matter more here than anywhere else in the kitchen.

Put the whole process together and it's short. Find the thickest part of the food and take the reading there. If it's below target, keep cooking and check again rather than guessing it's close enough. Once everything on the plate clears its number, wash the probe in hot, soapy water and run the ice-water check every so often so the next reading can be trusted. Get leftovers into the refrigerator inside that two-hour window, sooner if it's hot out. The thermometer habit handles the part of foodborne illness risk that's actually preventable at the stove; handwashing and quick refrigeration handle the rest.