A fridge that's too cold can freeze your food, while one that's too warm can spoil it. Produce is particularly susceptible to errant temperatures, but nearly every ingredient you refrigerate benefits from keeping the temperature set at a sweetspot.  

I don't like to waste groceries, so I set out to find out how cool my fridge was since there's no built-in thermometer. The ultimate goal was to ensure it's set to a temperature that keeps the items as fresh as possible without freezing. 

Here's how it went down and what I learned after testing nearly every temp setting in my fridge.

What temperature should your fridge be?

The scientific consensus puts the ideal fridge temperature at 40 degrees Fahrenheit (4 degrees Celsius) or below -- cold enough to keep bacteria in check, but not so cold that it freezes delicate foods.

Your freezer should be set to 0 degrees Fahrenheit (minus 18 degrees Celsius) or lower. It might seem like anything below 32 degrees Fahrenheit (0 degrees Celsius) would do the job, but the colder target exists for good reason: it keeps food solidly frozen even when you open the door, add fresh items, or ride out a brief power outage. As a bonus, lower temperatures also speed up freezing, which means smaller ice crystals and better preservation.

I ran tests using real-world conditions

When testing the fridge temp settings, I didn't unload all of my food, though I did move more sensitive items -- fresh produce, eggs, and some glass jars -- into temporary coolers for colder tests. Likewise, I couldn't keep items outside the fridge or at warmer temperatures for longer than two hours.

This all required me to keep my tests short. Every time I opened the door, checked the temperature or changed a setting, I affected the test. These inconsistencies, while distressing to my scientific mind, yielded findings that align more closely with real-world conditions in which you might open a packed fridge multiple times an hour.

Fridge thermometers make these tests possible

As with any test, I needed a way to measure and track temperatures over time. For this test, I used three different kinds of temperature sensors: ThermoPro Refrigerator Thermometers, Hatusoku Digital Thermometer with External Sensor and the recently announced Ikea Timmerflotte smart temperature and humidity sensor.

ThermoPro is the most suitable option for most people because it can be hung from shelves or flipped back into a helpful stand. The built-in light also helps you see the LCD in low lighting -- like my freezer, which has no interior light.

Hatusoku's long temperature probe makes it a great option for a thermometer that you can stick on the outside of the fridge for at-a-glance checks. Still, the probe is extremely sensitive and almost always reacts when I open the door.

Because it's so sensitive, the min/max setting isn't as useful. Consider waiting a few minutes after a cooling cycle before checking the temperature. You also need to contend with an additional wire when placing the probe.

Timmerflotte made my testing arsenal after I floated the idea to a company representative. They were just as curious to see how the sensor performed in the fridge as I was. Still, a developer with the company said the sensors could handle the internal temperature of my fridge, but that the metal might interfere with its connection to the Dirigera smart hub. (Fortunately, I had no connection issues during my test.)

Because it's a smart temperature sensor, Timmerflotte provided me with real-time temperature measurements without requiring me to open the fridge. Ikea sent enough sensors so that I could track the temperature in both the front and back of the fridge, which was incredibly helpful for finding the perfect setting. The Ikea Home Smart app didn't have a way to track temperature over time — I couldn't spot trends beyond my manual checks.

However, I should note that the minimum temperature for Timmerflotte was around 14 degrees F (minus 10 degrees C), so it was unable to track the temperature of my freezer. It also turned off after about 12 hours because freezing temperatures and Alkaline batteries don't get along.

How I tested my fridge temps: Limitations and caveats

My refrigerator uses a simple dial thermostat to control both the fridge and freezer. The dial lists numbers 1 through 9, with 9 being the coldest setting. Starting with 1, I measured the temperature for at least 20 minutes (often longer with colder settings) to allow it to stabilize after each compressor cycle. 

Here's the rub: My results are unique to my home and fridge. 

• Although GE still makes my fridge (Model GTR15BBMRWW), it's an old model with a likely manufacture date of April 2001. The owner's manual even recommends setting the temperature control dial to 5 and adjusting from there.
• I have a lot of food in my fridge due to Thanksgiving preparations. With such a large thermal load, it takes considerable time for the temperature to stabilize after adjustments are made. A high food load also blocks airflow during cooling, resulting in cold and hot spots.
• The average temperature of my kitchen during these November tests was about 64 degrees Fahrenheit, which affects how long the fridge runs.
• Frequently opening the door during testing certainly affected the results.
• I usually use a slightly colder setting in the summer to keep foods in the safe zone, and GE's advice reflects this. Even then, I only put items that are safe to freeze in the back of the fridge, just in case.

Finally, my measurements were more consistent and reliable in the fridge than in the freezer, which fortunately stayed below 14 degrees Fahrenheit (minus 10 degrees Celsius) throughout the test. (I mostly chalk this up to forgetting to photograph refrigerator thermometer readings for Levels 2, 3 and 4.) However, the freezer thermometer wasn't consistently at 0 degrees Fahrenheit (minus 18 degrees Celsius) or below until I reached Level 4.

My test results and takeaways

Ultimately, Level 5 (actually slightly below this) was the only one that kept all of my food within the target range. The front, back and freezer all met the aforementioned food safety guidelines. The other settings either froze items in the back or kept front items too warm.

You can see why I canceled the test after Level 8, which put the entire fridge at or below freezing. This level also caused the compressor to run continuously for almost an hour before shutting off. 

So, I couldn't risk spending at least as much time trying to hit Level 9, which I suspect could turn my entire fridge into a freezer -- even if the fridge compartment doesn't quite hit 0 degrees Fahrenheit. (I don't have enough cooler space to hold a fridge's worth of food.)

For my fridge, Level 5 is right in the middle, which is likely what the manufacturer intended when curating the various settings. From a user standpoint, I found that impressive. My biggest takeaway from the results is that setting the fridge to a colder temperature isn't necessarily better, which is sometimes my instinct when I perceive that food is spoiling too quickly. 

Should you test the temperature of your own fridge?

If your food is consistently spoiling or freezing without explanation, it could be because your settings aren't calibrated properly. You may not even know which setting is intended to hit the target fridge temp of 40 degrees Fahrenheit (4 degrees Celsius) or below. If that's the case, running a similar test on your own fridge would be wise. 

I recommend getting a set of fridge thermometers as I did. I consider them essential for simple fridges like mine, which use a dial thermostat that doesn't display the actual temperature.

While many fridges have a digital thermostat with your target temperature, an independent fridge thermometer's flexibility can help you learn more about your fridge. For example, you'll be able to move the thermometer around to find cold spots or track temperatures during a power outage.