Three numbers appear on almost every cell product specification and they get treated as one composite score of quality. They measure different things, they can move in opposite directions, and optimising for one routinely damages another. Knowing which one your experiment actually depends on is more useful than wanting all three to be high.
Viability
Viability is the proportion of cells alive at the moment of measurement, determined by whether they exclude a dye or by a more discriminating method. Two qualifiers make the number meaningful and are frequently missing.
The first is timepoint. Pre-freeze and post-thaw viability describe different things: the first reports how well the isolation went, the second reports what you will actually work with. The second is method, because dye exclusion, dual-stain approaches and flow-based methods do not return identical numbers on the same sample. A viability figure without both qualifiers cannot be compared with anything.
Recovery
Recovery is how many cells came out relative to how many went in. It is the number most often absent from a specification and the one that most directly determines whether you have enough material.
Crucially it moves independently of viability. A preparation can report excellent viability and poor recovery, meaning the surviving cells are healthy and there are simply fewer of them than planned. Yield calculations built on viability alone are running on an assumption.
Purity
Purity is the proportion of the preparation that is the population you ordered, and it is only interpretable alongside the marker and the gating strategy used to define it. Change the gate and the number changes without anything about the cells changing.
Expectations should also scale with how abundant the target is. High purity is straightforward for an abundant, cleanly marked population and considerably harder for a rare one, which is why a CD34 purity specification and a pan T cell purity specification are not comparable figures even though both are percentages.
Related product
Cryopreserved PBMCs. Single-donor PBMCs in current catalogue formats, with the donor record attached.
How they trade against each other
| If you optimise for | What typically improves | What typically suffers |
|---|---|---|
| Purity | Cleaner population, less confounding | Recovery, through additional selection and wash steps |
| Recovery | More material to work with | Purity, since fewer depletion steps are applied |
| Viability | Healthier surviving population | Recovery, if gentle handling means accepting losses elsewhere |
Because these trade, the right target is set by the downstream application rather than by a general preference for high numbers. A functional assay tolerates lower purity and needs cells that work. A phenotyping panel tolerates lower recovery and needs a clean population.
Where published specifications fit
A published specification states what a supplier will stand behind at a defined point, not what you will observe after your own handling. The gap between the two is usually the thaw, which is covered in thawing PBMCs.
Specifications also appear only where the material supports them. OrganaBio lists post-thaw viability at 85% or above on peripheral blood NK cells, and viability and CD56+ purity each at 90% or above on cord blood NK cells, while publishing no viability threshold for whole leukopak formats because a single figure across a heterogeneous unit would not carry weight. Reading those gaps correctly is covered in reading a certificate of analysis.
What to record on your side
Measure viability and count at the same timepoint, after any rest period rather than immediately post-thaw. Keep the gating strategy with any purity figure you generate. And record recovery even when the supplier did not, because it is the number that will explain a shortfall later.
Frequently asked questions
What is the difference between viability, recovery and purity?
Viability is the proportion of cells alive at measurement, recovery is how many cells came out relative to input, and purity is the proportion belonging to the target population. They measure different things and can move in opposite directions.
Why does a viability figure need a timepoint?
Because pre-freeze viability reports how the isolation went while post-thaw viability reports what you will actually work with. For material you intend to thaw, only the second is relevant, and a figure without a stated timepoint cannot be compared.
Why is recovery so often missing from specifications?
It is the least frequently published of the three and the one that most directly determines whether you have enough material. Because it moves independently of viability, a yield calculation built on viability alone is running on an assumption.
Can purity figures be compared between products?
Only cautiously. Purity depends on the marker and gating strategy, and expectations scale with how abundant the target population is. A CD34 purity figure and a pan T cell purity figure are not comparable even though both are percentages.
Why do these three numbers trade against each other?
Additional selection and wash steps raise purity while costing recovery. Fewer depletion steps preserve recovery while lowering purity. Gentle handling protects viability while sometimes accepting losses elsewhere. The right target is set by the downstream application.
What single number should I actually plan against?
Viable cells of the target population delivered, which is the product of all three rather than any one. It is what the experiment consumes and it is rarely reported directly.
Why do my measured numbers differ from the published specification?
A specification states what the supplier stands behind at a defined point, not what you observe after your own handling. The gap is usually attributable to the thaw and to when the measurement was taken relative to any rest period.
Talk to OrganaBio
Working through this on a live program?
The scientific team works through sourcing and specification questions with cell therapy and research groups directly, including donor characterization, format selection and documentation scope.
