Stability Testing Staffing: Contract Lab Techs for a Pull-Schedule Spike

Last updated: October 30, 2026

By Dave Manchester, Scientific Recruiting, KORE1

Can you see a stability testing spike before it lands on your bench? You can, because pull points are fixed the day a study starts. ICH Q1A(R2) calls for long-term testing every three months in year one, every six in year two, then yearly. Stack every study’s pulls, convert them to analyst-hours, and start a contractor early enough to finish site training before the heaviest month. Data as of October 2026.

One thing before the math. KORE1 places contract lab techs, I’m the person who recruits them for our pharmaceutical staffing and laboratory staffing work, and this page ends with a way to hire through us. So check my reasoning against the ICH text. I link it.

Why Is Stability Testing the One QC Spike You Can See Coming?

A stability pull is the scheduled removal of samples from a stability chamber at a fixed time point, such as month three or month nine, so the lab can test them and show how the product holds up over time. Each pull sends a defined set of tests to the bench on a date somebody wrote down months or years earlier.

Most QC work shows up uninvited. A batch finishes, a complaint arrives, an auditor wants something by Friday. Stability is different. It sends a calendar first.

The calendar comes straight from the guideline. Fixed from day zero. For a product with a proposed shelf life of at least 12 months, ICH Q1A(R2) says the frequency of long-term testing “should normally be every 3 months over the first year, every 6 months over the second year, and annually thereafter through the proposed shelf life.” The accelerated arm of a six-month study needs at least three time points, such as 0, 3, and 6 months. In the United States the written program is also a legal requirement, because 21 CFR 211.166 calls for test intervals based on statistical criteria and says the written program “shall be followed.” Not optional.

Source: ICH Q1A(R2), Step 4 version dated February 6, 2003, drug product testing frequency. Your protocol governs and can add pulls, such as an intermediate condition.

A caveat on that source. ICH is folding the Q1 stability series and Q5C into a single Q1 guideline. Public consultation ran from April to September 2025, and ICH’s own Q1 work plan expects final adoption in November 2026. As I write this in October 2026, Q1A(R2) is still the text I’m reading from. If you lock protocols after that date, check ICH before you trust me.

So what does a fixed calendar buy a staffing plan? A forecast you already own. Somewhere in LabWare, STARLIMS, a stability module, or the spreadsheet nobody admits to maintaining, every pull date for the next 18 months already exists. You can read it today. Hiring to the average month fails twice, because you’re short when the pulls stack up and paying for slack when they don’t. Both cost money.

How Do You Turn the Pull Calendar Into Analyst-Hours?

Four inputs and one sum. Nothing exotic. Studies, times pull points, times tests per pull, times hours per test, added up by calendar month. “Hours per test” means everything an analyst does for that test at the bench or the keyboard, which covers prep, the attended part of the run, data processing, documentation, and second-person data review. Unattended instrument time stays out. The HPLC works alone while the analyst starts the next sample set.

Here is a made-up portfolio of six studies to show the arithmetic. Every number below is invented, none of it is KORE1 data, and the shape is the point.

Lay each study’s pull months over the 2027 calendar, using the 0, 3, 6, 9, 12, 18, and 24 month points, and total the hours.

Illustration only. The year totals 784 analyst hours, an average of about 65 a month. Darker shading means a heavier month. Swap in your own studies, start dates and test loads.

Read it by month and the year stops looking busy and starts looking lumpy. March, June, and December each carry 140 hours, a little more than double the monthly average. Four months carry nothing. A permanent team sized for the average is short in three months and has slack in four.

I used 120 bench hours per analyst-month, about 28 a week once you take out training, logbooks, waiting on a reviewer, and an instrument that’s already busy. Your number will differ. Ask whoever builds your bench schedule, then trim what they say, because the first guess runs generous.

Two things shrink the spike before you hire anyone. Your protocol sets how far a pull can drift from its nominal date, and that window is scheduling slack you may be able to use, so read it before you pay anyone to cover a gap you could have shifted by a week. Then there’s the hardware. One HPLC runs one sequence at a time. Contractors add hands. Not detectors. If dissolution baths or LC-MS time limit your heavy month, a second analyst may add nothing.

I worked through the same arithmetic for a release backlog in clearing a batch release backlog with contract QC analysts. That one is a pile that already exists. This one is a pile you can put a date on.

When Should the Contractor Start?

Earlier than feels necessary. The hire is quick. KORE1’s offer-to-start time is a median of 3 days (n=1,997), 86% of people start within a week, and 37% start the same day they accept. How fast a contractor can start has the full split.

Training is the slow part. Under 21 CFR 211.25, anyone working on a drug product needs training “in the particular operations that the employee performs and in current good manufacturing practice.” So a contractor shouldn’t report GMP data until your quality unit has a signed training record covering those methods. The onboarding checklist for GMP contractors shows what that file holds. Book the time early.

Work backward from the heavy month. The top row is fixed. Every row under it is a clock somebody owns.

That 3-day figure starts at the offer. It leaves out finding the person, and I don’t have a requisition-to-offer number for lab seats that I’d stand behind, so I won’t print one. What I’d say is that a strong stability analyst is like a professional athlete. If they’re good, several teams are already talking to them, and a slow interview process loses them to whoever moves faster. Speed matters.

Devin Hornick, a partner at KORE1, puts the appeal of contract labor this way. “You can get them quick… flexible… to scale up, to scale down as needed.” Stability is a rare case where both halves of that sentence come with a date on them.

What Does a Stability Contractor Cost?

KORE1’s lab and QC contract placements from 2019 to 2026 (n=54) billed a 25th percentile of $33 an hour, a median of $40, and a 75th percentile of $56. Ryan McManus of KORE1 describes the reason we publish ranges like this as “the structured data of what people can expect to pay.” Here is one stability seat at 40 paid hours a week.

Source: KORE1 placement data, 2019 to 2026, life sciences, lab and QC family, n=54 W2 contract placements (14 since 2022, and their bill median is also $40), exported October 16, 2026. The wider dataset is 79% California. Anomalies and payroll-service deals are excluded, and burden is modeled at 24%. Bill rates only. Each cell is paid hours times the hourly bill rate.

Let’s Connect. Send me next year’s pull calendar and I will tell you when to bring someone in. Send me the calendar with the number of active studies, your heaviest month, the methods per pull, whether the work is GMP or GLP, the site city, and the start date you’re aiming for.

Run my invented portfolio through it. March through June carries 356 testing hours, about three analyst-months. One seat across those 17 weeks is 680 paid hours, or $27,200 at the median. The gap between 680 and 356 is real. In May the seat has no pulls to test, so it goes to routine QC work, data review, or the backlog your permanent team keeps stepping around. No such work? Plan around it before you sign anything.

Lab and QC contracts ran a median 193 days (n=54), against 116 days for all roles, so six months is closer to normal than three. Six months at 40 hours a week is exactly 1,040 hours, the point where a KORE1 contractor can move to your payroll with no conversion fee. How long contract assignments last shows the spread.

What’s inside the hourly rate? Wages, payroll taxes, workers’ compensation, liability insurance, paid sick leave, and the firm’s margin, billed as one number. For scale, the Bureau of Labor Statistics reports a national median hourly wage of $29.03 for chemical technicians, the closest federal category for bench analysts, with a 25th percentile of $23.33 and a 75th of $36.49 (OEWS, May 2025, SOC 19-4031). Our lab and QC contractor bill rates page sets the two side by side by role.

Devin’s worked example uses a $100 hourly pay rate. Add a 24% burden and the loaded cost is $124. A typical bill rate on that pay lands around $150 to $160. It isn’t lab-specific, and lab pay sits well below $100, so scale it down. Pay, then burden, then margin. If a firm quotes you a markup percentage, ask what it sits on, because 35% of the pay rate is not 35% of the loaded cost. Our guide to comparing staffing quotes line by line goes through it, and how KORE1 models burden shows the method.

Which Stability Seats Do You Contract, and Which Do You Keep?

Keep two. The stability program owner holds the protocols, the chamber inventory, the pull calendar, and the shelf-life argument. The investigation lead takes the calls nobody wants. An out-of-specification (OOS) result falls outside a limit in the specification. An out-of-trend (OOT) result is still inside the limit but doesn’t fit the pattern the earlier time points drew. Both need someone who knows the product’s history, and 21 CFR 211.192 requires a thorough, written investigation of any failure to meet specification. Rent that seat and you rent your product’s memory.

One habit I’d add. On every heavy month, name a permanent analyst who already knows the product as the person the contractor asks. A contractor with nobody to ask will guess. Cheap insurance. Our contract QC analysts page covers the wider bench if the spike turns out to be more than pulls.

What Happens If a Pull Is Missed?

A missed pull becomes a deviation. The written program “shall be followed” under 21 CFR 211.166, and under 21 CFR 211.160(a) any departure from the written sampling plans and test procedures “shall be recorded and justified.” So a pull that slips past its allowed window turns into a record your quality unit has to assess, plus a hole in the time-point series that the shelf-life argument stands on. Whether the data survive that hole depends on your protocol’s window, the statistics, and the attribute, and that’s a call for your stability owner and QA. I’m a recruiter. What I can tell you is that the cheapest deviation is the one a trained pair of hands prevented. Prevention wins.

Questions Stability Managers Ask

How do I forecast QC analyst hours for an upcoming stability pull schedule?

List every active study, mark each pull month from the protocol, multiply tests per pull by attended hours per test, and total by month.

The peak month, not the average, tells you when to bring someone in. Add the study starts you expect, because a new study pulls at month zero and again at month three. The worksheet above runs the arithmetic, and the contract lab and QC cost calculator prices the seat once you know the hours.

Do stability analysts need HPLC and LC-MS experience, or can a general lab tech run the pulls?

A general lab tech can pull and log samples, but running and reviewing the assay and degradant methods takes an analyst with HPLC experience.

HPLC is the workhorse for assay and related substances in most small-molecule stability programs. LC-MS comes in when someone has to identify an unknown peak, so it’s a narrower skill and worth asking about directly. Ask which chromatography data system the candidate has used, whether that’s Waters Empower, Thermo Chromeleon, or Agilent OpenLab, because relearning one during your heaviest month is expensive. Ask early. KORE1’s life sciences team recruits for pharma, biotech, and CDMO labs, and analysts with those methods are what we place.

How soon can a contract analyst start once I say yes?

Three days at the median, counted from signed offer to first day (n=1,997), and 86% of KORE1 placements start within seven days.

That is the hire, not the training. Site SOPs and your stability methods still have to be taught and signed off before the contractor reports GMP data, so start that clock the day the offer is accepted.

Can I keep the analyst once the spike is over?

Yes, with no conversion fee once the contractor has worked 1,040 hours, which is 26 weeks at 40 hours a week.

Lab and QC contracts ran a median 193 days (n=54), so a seat that starts ahead of a heavy quarter can reach that mark in the second half of the year. The calendar decides it. If next year’s pulls are as heavy as this year’s, you may be describing a permanent seat.

Six months of one stability analyst, what does that run?

$41,600 at the KORE1 lab and QC median bill rate of $40 an hour, for 40 hours a week over 26 weeks.

The middle half of our placements billed between $33 and $56 an hour (n=54), which puts six months between $34,320 and $58,240, and because our data leans California, a quote for Cambridge or Research Triangle Park gets built for that market’s pay and not copied from ours. Tell me the city and I’ll tell you whether we can fill it.

Send Me Next Year’s Pull Calendar

Six details do it. The number of active studies, your heaviest month, the methods per pull, GMP or GLP, the site city, and the date you want someone at the bench. If you’d rather price it yourself first, the contract lab and QC cost calculator does the seat math, and our contract staffing page explains the model.

When you’re ready, Let’s Connect. Expert in finding you the right talent, right away. You can also find me on LinkedIn.