Eight questions worth asking before you sign a scope, and why each one actually matters.
A sales call can make almost any lab sound capable. The questions that separate a real answer from a polished one are usually specific, sometimes a little uncomfortable, and rarely the ones a vendor volunteers on their own. Here are eight worth asking directly, and what a good answer to each one actually sounds like.
What’s your experience validating this specific specimen type?
Validation doesn’t transfer cleanly across specimen types. A lab with deep experience in venous blood assays can still hit real surprises the first time it validates a dried blood spot or swab-based test, since collection technique, degradation behavior, and storage stability all change with the matrix (Prosperi et al., mSphere, 2021). Ask for a specific example of a specimen type they’ve validated that’s close to yours, not just a list of matrices on a capabilities page.
What does your self-collection rejection rate actually look like, and why do specimens usually get rejected?
This is one of the more revealing questions on this list, because most labs don’t track the number precisely enough to answer it. Nexus runs a 0.89% sample rejection rate year to date, well under the roughly 3% industry average for mail-in kit programs. The gap isn’t luck: expired samples, missing labels, insufficient quantity, and damaged swabs account for most rejections industry-wide, and all four are preventable with the right intake protocol. A lab that can name its rejection rate and its most common causes has clearly looked at the problem. One that can’t hasn’t.
Can your documentation support our IRB and procurement requirements?
Human subjects research protections under 45 CFR 46, the federal Common Rule, shape what your IRB expects to see from every part of a study that touches a specimen, lab work included (HHS Office for Human Research Protections, 45 CFR 46). If a lab partner has never produced documentation an IRB has actually reviewed and accepted, that’s worth knowing before it becomes your problem mid-study, not after.
How transparent is your timeline, and what happens when something changes it?
A validation program runs through several discrete stages: feasibility work, assay development, pre-validation, formal validation, technical transfer, and go-live. A lab that can name those stages specifically, and tell you roughly how long each one takes for work like yours, is a lab that’s actually tracking its own timeline rather than estimating from memory. Ask what they do when a stage runs long, not just how long it’s supposed to take.
Can you take on a specimen type or matrix you haven’t validated before?
Some labs will say yes to almost anything during a sales conversation. The better answer is a specific one: what their process looks like for a genuinely new matrix, including feasibility testing and a real equivalency study against an established reference method before it goes anywhere near a live program (Prosperi et al., mSphere, 2021). A lab that’s honest about needing that runway is more trustworthy than one that implies it can skip it.
What does your data delivery actually look like?
Ask how results actually reach your team: a portal login, a PDF report, or a structured feed your own systems can consume directly. Nexus supports API and SFTP integration in CSV and JSON formats, so results can land directly in a client’s own data pipeline rather than requiring manual re-entry. If a lab can’t describe its delivery format precisely, assume it means a PDF, and plan your own data handling around that.

What’s actually in your final data package?
A credible deliverable includes a clearly defined study design with acceptance criteria set before testing began, raw data that traces back to instrument output, lot numbers, and sample IDs without gaps, documentation of any deviations and why they happened, and interpretation of what the data means, not just a table of numbers. A rushed deliverable shows that something worked once. A credible one shows whether it’s reliable, reproducible, and ready to be relied on.
How do you run validation work without disrupting your existing clinical testing?
Validation studies compete for the same instruments, staff time, and reagents as day-to-day patient testing. A lab that’s honest about this tension, and can describe how it schedules validation work around clinical volume rather than treating the two as separate worlds, is one that’s less likely to let your project quietly slip because a patient sample queue took priority that week.
Ready to put these questions to a specific program? Talk to our team and see how we’d answer each one for your project.
