CrystalDiskInfo and the SSD Health Check That Saves You a Purchase
Windows 11 now reports an SSD health percentage, but it gives no trend, no cause and no countdown, which pushes cautious users toward buying a replacement they may not need. CrystalDiskInfo fills that gap with SMART attributes and a clearer picture of remaining life. This troubleshooting guide walks through the symptom, the ordered diagnostic steps, and what to do when a step fails.
You open Windows 11, click through Settings > System > Storage > Advanced storage settings > Disks & volumes, and see a number that ruins your afternoon: your SSD is at 77% health. The drive still works, the machine boots fine, but the figure sits there without context. There is no trend line, no explanation of what caused the drop, and no estimate of how many months of use remain. That is exactly the situation one user described before nearly buying an expensive replacement drive, only to find after a second opinion that the SSD still had plenty of life left ¹.
The symptom, then, is not a failing drive. It is an ambiguous health reading that gives you no basis for a decision. The general rule of thumb cited in that account is that once health drops below 70%, you should back up your data and start thinking about a replacement — which puts a 77% drive in the uncomfortable "getting close, but how close?" zone. CrystalDiskInfo, a free and open-source utility, is the standard second opinion because it exposes the raw SMART attributes behind the percentage instead of a single summary number.
Here is the safe diagnostic sequence, in order.
1. Back up anything irreplaceable before you touch any diagnostic tool. Checking SMART data is read-only and safe, but the point of the exercise is deciding whether the drive is trustworthy, and you should not be making that decision with unprotected data on the disk.
2. Confirm what Windows already reports. Note the estimated remaining life, available spare and temperature shown in the drive's Properties panel. Windows 11 exposes more than Windows 10 did, but the figures arrive without interpretation.
3. Install CrystalDiskInfo and read the top-level health status first. The tool presents a simple good/caution/bad verdict alongside the drive model, firmware, interface and temperature, so you can see immediately whether the drive is flagging a problem or merely aging.
4. Open the SMART attribute list and find the wear and spare-block entries. These are the attributes that explain the percentage: how much of the drive's rated write endurance has been consumed, and how much spare capacity the controller has left to swap in for worn cells. A health figure without these is just a score.
5. Compare the drive's total host writes against its TBW rating. Most SSDs publish a Terabytes Written figure as a guideline for expected lifespan, and tracking usage against it is the practical way to judge how much of the rated endurance you have actually spent ². If the tool reports host writes well under the rating, a mid-70s health percentage is far less alarming than it looks.
6. Watch the trend rather than a single reading. Record the health percentage and total host writes now, then check again in a few weeks of normal use. A drive that sits at 77% for months is behaving very differently from one dropping a point a week.
7. Check temperature under load. Sustained high temperatures accelerate wear and can also cause the controller to throttle. If the reported temperature is high at idle, investigate airflow and mounting before blaming the NAND.
When a step does not resolve the question, the fallback depends on which step failed. If CrystalDiskInfo shows a caution or bad status rather than a percentage you can interpret, treat that as the tool telling you the drive itself is reporting a problem, and move straight to backing up and planning a replacement rather than hunting for a more flattering number. If the SMART attributes are unreadable — some USB enclosures and RAID configurations pass through limited or no SMART data — the limitation is the connection, not the drive, and you should test the drive on a direct SATA or M.2 connection before drawing conclusions. If the health percentage and the host-write figure tell contradictory stories, trust the endurance math over the summary score and keep monitoring.
There is also a platform caveat worth knowing. A comparison of the same SSD health check run on Linux and Windows found that only one operating system gave the full picture, which suggests the answer you get depends partly on which tool and which OS you ask ³. That is an argument for cross-checking CrystalDiskInfo's readings against the manufacturer's own utility when one is available, particularly for firmware-specific attributes.
Two things remain genuinely unconfirmed in the sources here. Neither account establishes how CrystalDiskInfo calculates or weights its health percentage, so treat the headline figure as a summary of the underlying attributes rather than an independent measurement. And no source gives a conversion from health percentage to remaining months of use — the "how long have I got" question that sends people to the store in the first place cannot be answered precisely from a single reading. The honest position is that a health percentage plus host writes plus a trend over time gets you a defensible decision; the percentage alone does not.
If you are already shopping because the reading scared you, the buying guidance elsewhere on this site is worth reading before you spend: the tips before you buy an M.2 SSD cover the compatibility questions that matter more than a few points of wear, and if the drive in question is destined for a console, the budget PS5 SSD pick is a separate decision from whether your current drive is dying. For most readers at 77% with modest host writes, the correct action this month is a backup and a calendar reminder, not a purchase.
Sources