Many people rightly believe that panicking about SSD lifespan is pointless, since most PCs will never write enough data to a drive to exhaust its endurance. If you check your SSD's health with CrystalDiskInfo or a first-party tool, you'll probably see "Good" or "Normal" under "Drive Health." You'll probably see the same status for years, irrespective of what's going on physically with the NAND flash on your drive. SSD health is just a percentage that tells you how much of the drive's rated lifespan has been exhausted. It doesn't account for serious read/write errors or physical degradation of the SSD's cells. Even an SSD with 99% or 95% health could have encountered uncorrectable errors, at which point it should not be trusted with your data anymore. A deeper look at your drive's S.M.A.R.T. data will tell you if you need to start preparing for the last rites.

Your SSD's S.M.A.R.T. data records nearly everything about it

The "Drive Health" percentage is just part of the story

Every SSD (or HDD) constantly records S.M.A.R.T. (Self-Monitoring, Analysis, and Reporting Technology) attributes, storing up to 255 pieces of information about the drive's activity. Most manufacturers use only 20–30 of these attributes to measure drive health, performance, and wear, and these are what you see in tools like CrystalDiskInfo or manufacturer tools like SanDisk Dashboard (also for WD drives), Samsung Magician, Adata SSD Toolbox, and others. The simple percentage value most people use to gauge SSD health is derived from the Percentage Used metric. For instance, the Gen3 SSD in my system shows "5" next to the Percentage Used metric (I've switched from hexadecimal to decimal in CrystalDiskInfo settings), which automatically shows "95%" under Health Status at the top of the screen.

This only means that I've written around 5% of the total amount of data I can write to the SSD in its lifespan, as certified by the manufacturer in the form of the TBW rating. While it does give me a general idea about the mileage of my SSD to date, it doesn't actually tell me anything about the true health of the drive. Without checking other metrics, I can't know for sure whether the drive is free of serious errors that could threaten data integrity in the near future.

A "healthy" drive can still hide severe read/write errors

You need to look at other metrics

Despite what the Health Status section says in CrystalDiskInfo, your SSD could still have encountered severe read/write errors already. You need to look beyond the percentage lifespan used, and check metrics like Media and Data Integrity Errors, Critical Warning, and Number of Error Information Log Entries. The first of these metrics should have "0" next to it; anything else indicates that the physical cells on your SSD have started degrading. When the drive's ECC mechanisms fail to recover read/write failures, it's considered a severe failure and logged as a data integrity error. This basically means that any non-zero value in the Media and Data Integrity Errors field cannot be taken lightly.

The metrics that can trigger a non-zero entry next to the Critical Warning field are Available Spare (the number of buffer cells that can be used to replace retired cells), Temperature, and, of course, Media and Data Integrity Errors. If your SSD has entered a read-only state, that will also be logged as a "Critical Warning." A drive that's barely been used is still prone to data integrity errors due to sudden power loss events, overheating, or faulty firmware. The health status of your drive does not reflect these events — you need a deeper look at other S.M.A.R.T. metrics for that.

What should you do if your SSD is failing?

Take immediate action

So, what should you do if you actually spot serious read/write errors on your drive? Well, the first order of business is to back up your data if you haven't already done it. A compromised drive can fail at any time, taking your data with it. You may find it difficult to part with a drive that still has most of its lifespan left, but you can't keep using it if you value your data. Many users have drives with several dozen data integrity errors, but keep using the drive for years. While regular backups and an absence of glaring storage issues might convince you to "wing it," the safest recommendation is to RMA the drive (if it's under warranty) or to replace it as soon as possible.

You could also try updating your SSD firmware if a new version is available — your SSD software will clearly show you have a pending update. A firmware update can sometimes solve the underlying reason behind your drive's data integrity errors. If the error count still keeps rising in CrystalDiskInfo over the span of a few months or a year, then you should accept your fate and chuck the drive for good.

Don't rely on the SSD health percentage alone

It is easy to look at your SSD's "Good" health percentage and think everything is fine, but the real picture might be very different. The other S.M.A.R.T. metrics, in addition to Percentage Used, will show you what's really going on with your drive. Any severe physical issues with the NAND flash get logged here and indicate a failing drive. You should stop using it immediately, back up your data, and replace it. A firmware update can sometimes save your drive, but in most cases, write errors signal impending drive failure.