Every quantitative result in a modern laboratory rests on the assumption that liquid volumes are delivered correctly, and that assumption lives or dies with pipette performance. Accuracy describes how close a delivered volume is to the intended target, while precision describes how consistently the pipette delivers that same volume across repeated attempts. A pipette can be precise but inaccurate, accurate on average but imprecise, or, ideally, both accurate and precise. The distinction matters because the sources of each type of error differ, the fixes differ, and the downstream consequences for experimental results can be severe in ways that are not always obvious.
What Accuracy and Precision Actually Mean in Practice
In pipetting, accuracy is often expressed as percent bias or systematic error: if you set a pipette to 100 µL and it consistently delivers 98 µL, it has a -2% bias. This kind of error shifts every measurement in the same direction and can sometimes be corrected by recalibrating the instrument. Precision, on the other hand, is expressed as the coefficient of variation (CV) across repeated dispenses. A pipette that delivers 99.8 µL, then 100.3 µL, then 99.5 µL has decent precision because the spread is narrow, even if the average drifts slightly from the target.
The two properties are independent. You can have tight clustering around the wrong value (precise but inaccurate) or a scattered spread that happens to average out near the correct value (accurate on average but imprecise). For most laboratory work, precision is the harder property to maintain because it depends heavily on operator technique and cannot simply be dialed away with a calibration adjustment. Determination of both properties is straightforward using either gravimetric or spectrophotometric readouts, and standardized protocols exist so that any lab can run competency checks on its own pipettes and operators.1Journal of Applied Laboratory Medicine. Assessing variations in manual pipetting: An under-investigated requirement of good laboratory practice
Factors That Degrade Pipette Performance
A pipette fresh from calibration, used perfectly, still exists in a world full of variables that erode its numbers. Some of those variables are properties of the instrument, some belong to the liquid being handled, and some belong to the person holding the pipette. Understanding which factor is responsible for a given error is the first step toward fixing it.
The Instrument Itself
Air-displacement pipettes work by moving a piston that pushes or pulls a column of air above the liquid. Any change in the air column’s behavior changes the delivered volume. Temperature matters: if the pipette body is warmer than the liquid, the air inside expands slightly and the pipette under-delivers. Altitude and barometric pressure also play a role, because air compresses differently at different pressures. Worn seals, contaminated pistons, and damaged tips all introduce random or systematic volume shifts. Over time, springs lose tension and O-rings degrade, which is why calibration drift is not a question of if but when.
The Liquid
Water is the reference liquid for calibration, but labs pipette buffers, viscous glycerol solutions, organic solvents, and protein-laden samples that behave nothing like water. Viscous liquids cling to the inner wall of the tip and resist aspiration, causing the pipette to pick up less than the set volume. Volatile solvents evaporate inside the tip, increasing air pressure and pushing liquid out prematurely. Surface tension differences change how liquid wets the tip and how droplets release. None of these effects are captured by a standard water-based calibration, so the “accurate” reading on a calibration certificate may not reflect real-world performance with your actual reagents.
The Operator
Human variability is often the largest single contributor to pipetting error, and it is the most neglected. How fast you press the plunger, the angle at which you hold the pipette, how deeply you immerse the tip, whether you pre-wet the tip, and how consistently you release liquid all affect the delivered volume. Fatigue compounds these issues: after hours of repetitive pipetting, thumb force and plunger speed become less consistent. Even the simple act of gripping a warm pipette body for an extended period changes the internal air temperature enough to shift volumes at very small scales.
Forward Versus Reverse Pipetting
Most routine pipetting uses the forward technique: you press the plunger to the first stop, immerse the tip, release the plunger to aspirate liquid, and then press to the first stop again to dispense. The second stop (a further press past the first) is used to blow out any residual liquid in the tip. This is the standard technique and the one pipettes are calibrated for.
Reverse pipetting flips the aspiration step. You press the plunger past the first stop to the second stop before aspirating, which draws in more liquid than the set volume. On dispense, you press only to the first stop, leaving an intentional excess in the tip. The idea is that viscous or foamy liquids are dispensed more reproducibly this way because the blow-out step, which introduces the most variability, is eliminated from the delivery.
A comparative study of the two techniques among laboratory students found that forward pipetting yielded an average precision of about 99.3% and accuracy of roughly 96.6%, while reverse pipetting achieved precision of about 99.7% and accuracy of around 95.9%.2Jurnal Laboratorium Khatulistiwa. Precision and Accuracy Analysis of Forward Method Pipetting and Reverse Method Pipetting by Technology of Medical Laboratory Students Pontianak The takeaway is subtle: reverse pipetting was slightly more precise (tighter clustering) but slightly less accurate (further from the target volume on average). For most applications, forward pipetting is appropriate. Reverse pipetting earns its place when you are handling viscous solutions, surfactant-heavy buffers, or volatile solvents where tip wetting and blow-out inconsistencies dominate the error budget.
How Pipette Performance Is Verified
Two main approaches exist for checking whether a pipette is performing within specification, and they do not always agree with each other. Understanding their differences helps explain why a pipette can pass one test and fail another.
Gravimetric Verification
The gravimetric method is the gold standard referenced in ISO 8655, the international standard for piston-operated volumetric instruments. You dispense water onto an analytical balance, record the mass, and convert it to volume using the density of water at the measured temperature. Repeat this ten times (or more), and you have both accuracy (average volume versus nominal) and precision (spread of the measurements). The method is simple, traceable, and widely accepted. Its main limitation is that it tells you only the total mass of liquid that arrived on the balance. It cannot detect what the liquid is or whether it was diluted along the way.
Photometric Verification
Photometric methods use dye solutions instead of plain water. You pipette a known dye into a microplate well, read absorbance on a plate reader, and back-calculate the volume from the absorbance value. Dual-dye ratiometric photometry, one of the more refined versions of this approach, uses two dyes at different wavelengths to self-correct for pathlength variation, and it can verify aqueous volumes from 2 to 200 µL with results traceable to national measurement standards.3JALA: Journal of the Association for Laboratory Automation. Multichannel Verification System (MVS): A Dual-Dye Ratiometric Photometry System for Performance Verification of Multichannel Liquid Delivery Devices
The photometric approach has a practical advantage that gravimetry lacks: it can catch sample dilution. In one study involving an automated liquid-handling system with fixed stainless-steel tips, the gravimetric method reported the expected volume and raised no flags, while the dual-dye photometric method detected that the sample had been diluted by the system liquid, reporting a lower apparent volume and alerting the user to the contamination problem.4PubMed. The use of a dual dye photometric calibration method to identify possible sample dilution from an automated multichannel liquid-handling system This is a real-world scenario with fixed-tip systems where carryover between samples is a concern. For disposable-tip manual pipettes, the dilution issue is less relevant, but the photometric method still offers the advantage of speed: you can verify an entire 96-channel head in a single plate read rather than weighing channels one at a time.
When Pipetting Errors Cascade Into Experimental Results
Pipetting errors are quiet. They do not trigger alarms or produce error messages. They simply shift your numbers, and because the shift is small on any individual dispense, it is easy to assume it does not matter. But in workflows that involve serial dilutions, the errors multiply.
Quantitative PCR is a vivid example. A standard curve in qPCR is built by serially diluting a known concentration of DNA, typically tenfold at each step. If the pipette delivering each dilution has a consistent 2% bias, that bias compounds across steps. By the time you reach the fifth dilution point, the actual concentration is meaningfully different from the intended concentration, and the slope of your standard curve shifts. Modeling of this effect has shown that a more rigorous approach to calibration curve production is needed, and that carefully choosing pipette conditions or using a single validated curve as the calibrator can substantially improve curve stability.5PubMed. Investigation of reproducibility and error associated with qPCR methods using Quantifiler Duo DNA quantification kit
The problem goes further. Systematic pipetting error creates an unintended actual dilution that differs from the intended dilution, and this discrepancy distorts the calculated amplification efficiency of the qPCR reaction. If the efficiency is reported without correction, every downstream quantification built on that curve inherits the error.6Clinical Chemistry. Efficiency Correction Is Required for Accurate Quantitative PCR Analysis and Reporting – Section: Estc and Eamc Differ Because of Dilution Errors In forensic DNA analysis, clinical diagnostics, and gene expression studies, this can mean the difference between a correct quantification and a result that is off by a factor of two or more. The irony is that most researchers blame the PCR chemistry or the thermal cycler long before suspecting the pipette.
Automated Liquid Handling Versus Manual Pipetting
Automated liquid handlers promise to remove operator variability from the equation, and they deliver on that promise to a meaningful degree. A robot presses the plunger with the same force, at the same speed, at the same angle, every single time. For high-throughput work involving hundreds or thousands of transfers, the consistency gains are substantial.
But automation does not eliminate volumetric error; it shifts the sources. An open-source liquid handler built from commodity linear-motion components achieved an average pipetting error of about 2.5%, compared to the roughly 1.5% error in the micropipette’s own specification.7Nature. Principles of computer-controlled linear motion applied to an open-source affordable liquid handler for automated micropipetting – Section: Speed and accuracy That gap matters. Automated systems trade one set of error sources (operator fatigue, inconsistent technique) for another (mechanical backlash, tip-geometry effects, aspiration-mode settings). The error profile is more predictable and easier to characterize, but it is not zero.
Tip choice and aspiration mode settings can make a significant difference on automated platforms. Testing on a commercial liquid handler showed that using a smaller-capacity tip combined with a post-aspirate air gap achieved the best results for small-volume transfers: a bias of about -2.2% and a CV at or below 1% for 4 µL target volumes.8PubMed Central. A practical and affordable photometric method for routine low volume verification of fixed-head automated liquid handling systems The post-aspirate air gap acts as a buffer that prevents liquid from dripping out of the tip between aspiration and dispense, a trick that manual pipette users accomplish instinctively by keeping the pipette vertical, but that robots need to be told to do explicitly.
Small Volumes Are the Hard Part
Everything about pipetting gets harder as volumes shrink. At 1000 µL, a 1% error is 10 µL, which is large enough to see on a balance and easy to control. At 1 µL, a 1% error is 10 nanoliters, which is smaller than the evaporation happening at the tip in real time. The physics conspire against you: surface tension dominates at small volumes, meaning the liquid’s tendency to cling to the tip wall rather than release into the receiving vessel becomes the primary error source. Air displacement becomes less reliable because the compressibility of the air column matters more when the liquid plug is tiny.
Manufacturer specifications reflect this reality. A typical adjustable pipette rated for 0.5 to 10 µL might guarantee ±2.5% accuracy at 10 µL but only ±5% at 1 µL. At the bottom of the range, you are operating near the instrument’s physical limits, and even small technique errors translate to large percentage deviations. For critical work at sub-microliter volumes, positive-displacement pipettes, which use a disposable piston that directly contacts the liquid and eliminates the air gap entirely, offer better performance than air-displacement designs. The tradeoff is higher consumable cost and fewer tip options.
Verification at these volumes is also harder. Gravimetric methods struggle because the mass of 1 µL of water (about 1 mg) is near the noise floor of most analytical balances, and evaporation during the weighing process can exceed the measurement itself. Photometric methods have an advantage here because they do not require the liquid to sit exposed on a balance; it goes directly into a dye solution and is read by absorbance. This is one reason photometric verification has gained traction for low-volume liquid-handling quality control.
How Often Should You Calibrate
The honest answer is that there is no single correct interval. Calibration frequency should be driven by risk, not by a blanket schedule. Major quality frameworks, including ISO 9001, GLP, and GMP, all describe risk-based approaches in which the scope and frequency of pipette performance verification are determined by the consequences of an inaccurate result.9Measurement: Sensors. Good pipetting practice – A science based approach for quality management of pipetting – Section: 5. Pipetting performance verification strategy A pipette used daily in a clinical diagnostic lab performing patient testing carries higher risk than one used weekly for non-quantitative buffer preparation. The former might warrant monthly verification and annual full calibration; the latter might be fine with quarterly checks.
In practice, many labs default to annual calibration because that is what their service contract covers, regardless of risk. This is better than nothing, but it leaves a wide window for drift to go undetected. A practical middle ground is to perform quick in-house checks (ten dispenses onto a balance, recording the average and spread) whenever a pipette returns from travel, gets dropped, changes users, or produces results that look off. These spot checks take five minutes and can catch problems months before the next scheduled service visit. The gravimetric and spectrophotometric protocols described in published laboratory practice guides make this accessible to any bench scientist, not just metrology specialists.1Journal of Applied Laboratory Medicine. Assessing variations in manual pipetting: An under-investigated requirement of good laboratory practice
Multichannel Pipettes and Channel-to-Channel Variation
Multichannel pipettes add a dimension of complexity that single-channel instruments do not have. Beyond the overall accuracy and precision of the device, each individual channel can perform differently. One channel might deliver 99 µL while the adjacent channel delivers 101 µL, and both are within the manufacturer’s specification for the instrument as a whole, but the 2 µL difference between them can introduce systematic patterns across a microplate.
In high-throughput screening, where you are comparing wells to each other rather than to an external standard, channel-to-channel variation matters more than the absolute accuracy of any single channel. If channel 4 consistently under-delivers, every well filled by channel 4 will have a slightly lower reagent concentration, potentially creating a stripe of anomalous readings across every plate. The dual-dye photometric verification approach was designed in part to address this, since it can assess every channel individually in a single plate read.3JALA: Journal of the Association for Laboratory Automation. Multichannel Verification System (MVS): A Dual-Dye Ratiometric Photometry System for Performance Verification of Multichannel Liquid Delivery Devices With gravimetric testing, verifying all eight or twelve channels of a multichannel pipette requires weighing each channel separately, which is slow enough that most labs simply do not do it.
Common Misconceptions About Pipette Performance
One persistent myth is that expensive pipettes are inherently more accurate than cheaper ones. Premium pipettes tend to have tighter manufacturing tolerances and more durable components, which means they hold their calibration longer and feel smoother to use. But a well-maintained budget pipette, freshly calibrated, can deliver volumes just as accurately as a top-tier instrument. The difference shows up over time and under heavy use, not on day one.
Another misconception is that pre-wetting the tip (aspirating and dispensing once before your actual transfer) is optional. Pre-wetting equilibrates the humidity inside the tip, reducing evaporation-driven volume loss on the first real aspiration. Skipping it can introduce a 1-2% error on the first transfer from a dry tip, especially with volatile solutions. Most published pipetting technique guides consider pre-wetting a baseline requirement, not an optional refinement.
A third misconception is that reverse pipetting is always more accurate. As the comparative study discussed earlier showed, reverse pipetting improved precision but slightly reduced accuracy compared to forward pipetting.2Jurnal Laboratorium Khatulistiwa. Precision and Accuracy Analysis of Forward Method Pipetting and Reverse Method Pipetting by Technology of Medical Laboratory Students Pontianak Reverse pipetting is a specialized technique best suited for problem liquids. Using it universally can actually introduce a small accuracy penalty because the pipette is calibrated for forward-mode dispense volumes, and the reverse technique intentionally deviates from that calibrated condition. Match the technique to the liquid, not to a blanket rule.
Tip Quality and Fit
The tip is the only part of the pipette that contacts the liquid, and its quality affects performance more than most users realize. A poorly fitting tip creates a weak seal with the pipette shaft, allowing air to leak past during aspiration and reducing the delivered volume. Tips from different manufacturers vary in inner diameter, wall thickness, and material composition, all of which change the dead volume (liquid retained inside the tip after dispense) and the wetting characteristics.
Using third-party tips is common and usually fine, but it introduces a variable that the pipette manufacturer did not account for during calibration. If you switch tip brands, it is worth running a quick gravimetric check to see whether performance changes. Some premium pipettes use proprietary tip-locking mechanisms designed to ensure a consistent seal, and mixing these with generic tips can degrade performance in ways that are not immediately visible. The same logic applies to automated systems: the combination of tip geometry and aspiration settings determines the final volumetric performance, and changing one without re-optimizing the other can introduce unexpected error.