Ask anyone how to pipette a thick liquid and the first answer is almost always the same: dilute it. Make an 80 percent solution before you even think about pipetting neat glycerol, cut the detergent stock with buffer, thin the sample until it flows. The advice is popular because it works, and it works because it attacks the problem at the source: a less viscous liquid is genuinely easier to pipette accurately than a viscous one, and no amount of clever tuning changes that. But dilution is not free, and treating it as the default without weighing what it costs is how a pipetting problem quietly becomes an assay problem. The real question is not whether dilution helps. It is whether dilution or tuning is the cheaper fight for the transfer in front of you.
What dilution actually buys you
Dilution buys you a liquid that behaves. Viscosity falls fast as you add a thinner solvent, so a stock that fought the tip becomes one an ordinary class handles with a modest flow-rate reduction and a short settling delay. It buys you robustness, too: a thinner liquid is less sensitive to temperature swings and tip wetting, so the class holds up better across a long run and across instruments. And it buys you speed, because you are no longer nursing every transfer at a crawl. For a liquid whose only sin is that it is thick, and whose concentration you are free to choose, dilution is often the right and obvious move.
What dilution costs, and why it is easy to miss
The cost is that you have changed the liquid, and everything downstream now depends on the change being correct and stable.
- It changes the assay. If the concentration matters to what happens in the well, and it usually does, then diluting the stock means every calculation, every final concentration, every dose now runs through your dilution being accurate. You have not removed a pipetting step, you have added one, and moved the accuracy burden onto it.
- It adds a step and a source of error. Making the dilution is itself a transfer, often of both a viscous stock and a diluent, and if that transfer is wrong the whole plate is wrong. You can push the hard pipetting one step upstream, but you cannot make it disappear.
- It can destabilize the sample. Adding water to a solvent stock can push a dissolved compound toward its solubility limit and precipitate it, or change the chemistry you were relying on. Dilution is a chemical act, not just a handling convenience.
- It consumes volume and time. A diluted stock takes more space, more labware, and more preparation, which matters at scale even when it is harmless in principle.
None of these are reasons never to dilute. They are reasons to notice that dilution moves the problem rather than deleting it, and to check that the place you moved it to is easier than where it started.
When dilution is close to free
Dilution is the clear winner when the concentration is yours to choose and the assay does not care about the exact value. A glycerol cryoprotectant that works fine anywhere in a broad band, a wash detergent whose job is qualitative, a viscosity you added only for handling and not for chemistry: thin these without a second thought, because there is no downstream accuracy burden to inherit the difficulty you removed. This is the common case, and it is why the advice is so popular.
When tuning is the honest choice
Tuning the class wins when the concentration is fixed by the science. If the assay needs a specific glycerol percentage, a specific detergent concentration, a stock at a defined molarity, then you cannot dilute your way out; the liquid is what it is, and the accuracy has to come from handling it as it is. Tuning also wins when the sample will not tolerate the diluent, when precipitation or degradation is a risk, and when the dilution step would itself be harder to validate than the direct transfer. In those cases the right investment is a properly built viscous class, slow flow, long settling delays, reverse pipetting, the right tip, and gravimetric proof, and possibly a positive-displacement channel for the thickest stocks.
A working heuristic
Ask one question first: does the concentration matter to the result? If it does not, dilute, and pipette a well-behaved liquid. If it does, you cannot escape the viscosity, so tune the class to handle it directly and validate that class hard. When both are in play, when the concentration matters but the stock is punishing, the usual answer is to dilute to the highest concentration the assay tolerates, so you climb down the viscosity ladder as far as the science allows and then tune for the rung you land on. That combination, dilute as much as you may and tune for the rest, is almost always cheaper than trying to win the whole fight with either tool alone.
Dilution does not remove the hard transfer, it relocates it. Before you thin a stock, make sure the place you are moving the difficulty to is genuinely easier than the place it started.