Arrange paired samples with a drinking-water laboratory before collecting them. Identify the same target contaminant, an actual untreated sample point and a point receiving treated water, then follow the laboratory's collection instructions. Record the filter's exact model, cartridge, service history and operating conditions. A lower result does not certify every contaminant claim.
Agree on the test before collecting
Use an appropriate drinking-water laboratory. The Arizona ADHS directory warns that not every listed facility performs every drinking-water analysis. Confirm the target substance and the facility's relevant capability before arranging the samples.
Before comparing a nitrate number with a limit, check nitrate as nitrogen versus nitrate on a laboratory report.
Explain that this is a treatment comparison, and ask for separate labeled containers and instructions for each location. The lab should define the collection method appropriate to the analyte. Do not substitute a generic flush or first-draw rule for every test; lead, bacteria and other analyses can require different preparation.
The laboratory selection guide covers those facility and collection questions. For a lead-specific concern, use the existing lead-testing guide.
Confirm what each point receives
Identify a real untreated point upstream of the device and the actual treated outlet you use. An outdoor faucet is not automatically upstream, and a kitchen tap may not share the dedicated filtered-faucet supply. Have the plumbing path confirmed if it is unclear.
Document any bypass, blending setting or separate treatment stage. A result collected while the filter is bypassed cannot show the usual treated-water performance. Conversely, a second device between the filter and sample tap changes what the comparison represents.
Record the device and conditions
Keep the exact system and cartridge identifiers, replacement date, recent sanitization or flushing, and normal usage with the results. Complete the manufacturer's prescribed commissioning before evaluating routine service. A new filter still requiring a flush is a different condition from its normal operation.
Record any indicator message too, but distinguish an RO faucet maintenance timer from a TDS monitor. That model-specific indication does not replace the laboratory’s analysis of the requested substance.
The CDC treatment guide emphasizes that treatment types remove different concerns, and the NSF standards guide warns that certification does not mean all contaminants are reduced. Choose the requested analyte to match the actual intended claim.
Interpret the comparison within its limits
Check units and the laboratory's reporting limit. A report of not detected is not a measurement of absolute zero. Ask the laboratory how the results can be compared, particularly if the untreated concentration is already below its reporting limit.
A successful paired test is evidence about those samples. It is not a substitute for checking the exact certified model and reduction claim, or for continued manufacturer maintenance. If the result fails the intended goal, review sample locations, parts, installation and operating conditions before assuming which component failed.