Early Access Doesn’t Pay for a Postdoc

The Alpha Site Deal Hasn’t Changed. The Labs Have.

Jupiter Hospitalis was the Roman god who watched over hospitium, the bond between host and guest, often translated as hospitality. The bond worked like a contract. Host and guest made promises, clasped hands and exchanged a written agreement or a token called a tessera, which they divided so that each side kept half. Breaking the bond offended Jupiter himself, and it could only be ended by a formal declaration.

The lesson is that a relationship lasts when both sides owe each other something real. A guest who expected shelter and offered nothing was not in hospitium. He was asking for a favor.

Product Managers (PMs) of research use only (RUO) life science tools ask alpha sites for favors all the time, and then we wonder why so few labs say yes.

Why willing labs stopped saying yes

The alpha site offer has stayed the same while lab funding and staffing have changed. It still looks like it did a decade ago: a loaner instrument or free kits, early access, a possible co-authorship on an application note, or a travel stipend to present a poster at a conference. Every item on that list assumes the lab has people and grant money to spare.

That assumption no longer holds. NIH issued 60,924 grant awards in fiscal year 2026 through late September, the lowest count in six years and about 11% below the five-year average, and The Scientist expects the squeeze to fall hardest on early-career investigators. Fewer grants mean fewer funded hours. The hours a lab does have are also harder to staff, because academic labs have struggled to fill postdoc positions as more PhD graduates go straight to industry (The Physiologist).

Here is how that shows up in your alpha program. You have found the two labs whose sample types match your primary claim. You sent the alpha site agreement three weeks ago. One principal investigator (PI) has not replied, and the other said yes and handed the project to a first-year graduate student. You suspect your alpha site offer is not enticing enough, but you have no clean way to show what a better one would cost. Your VP wants your recommendation on when the project will move to the next stage in the new product development process, and whether it is still on track for an on-time launch.

Before you can fix the offer, you need to recognize what the lab is telling you. The symptoms you are seeing all trace back to that offer.

Four symptoms of a one-sided deal

Each symptom gets blamed on something else. Each one is the lab telling you it gives more than it receives.

• The yes comes slowly. Weeks pass before you get a signed agreement, because the lab is looking for someone who can spare the hours and finding no one.
• The yes comes from your friends. When the offer gives the lab little, the only labs that accept have another reason to say yes, such as a long relationship with your sales team, a key opinion leader (KOL) who already uses your platform, or the hope of a discount. Those sites are the least likely to look hard for failures, so an offer like that builds confirmation bias into your alpha site list before the first sample is run.
• The middle goes quiet. Around week five, a grant deadline or a paper revision pulls the postdoc back, because the lab’s own work always comes first.
• The ending is polite. A lab that received little has little reason to spend another afternoon documenting why your qPCR master mix failed on degraded RNA, so even the structured debrief I described in Your Beta Programme Is Lying to You comes back thin.

All four symptoms share one cause, and you can measure it before you send your next request.

The alpha offer audit

Jobs-to-be-Done theory holds that people hire a product to make progress in a specific circumstance (Christensen et al., HBR). An alpha site hires your program too, perhaps for preliminary data for a grant renewal. The audit asks whether your program does that job at a price the lab can pay.

List every item you ask the site for and state its cost in the lab’s terms: hours, materials bought on its own grant, bench time and risk to its own experiments. Then list what you give back. The gap between the two columns is why labs say no or go quiet.

Here is an example audit for a 10-week alpha site evaluation of a benchtop live-cell imaging system.

*76 hours at the FY2026 NIH entry-level postdoctoral stipend of $63,480 a year (NIH), multiplied by 1.3 for fringe benefits. Industry-sponsored agreements often add university indirect costs on top. UCSF, for example, applies 64% to industry-sponsored research.

The audit shows three things:

• The largest cost, 76 postdoc hours, is cheap for you to cover and expensive for a lab whose grant pays for them.
• The loan ends with the alpha, but for all 10 weeks the instrument occupies bench space the lab needs for its own experiments.
• Two rows give nothing back. Weekly contact still matters, because silence from a site usually means the work has stopped, but it doesn’t need a full hour.

An alpha site for a new reagent may result in different rows and the same gap. For a qPCR kit, the lab supplies its own RNA, often from limited patient samples, plus run time and a postdoc to compare your kit with the one it trusts, in exchange for free kits. Industry sites, such as a diagnostic manufacturer or contract research organization (CRO), have budgets, but they still pay in validation runs and in the change control your product would trigger, so add those costs to your audit just as you would postdoc hours.

In Roman terms, you are asking the lab to hold both halves of the token.

Pay it, trade it or trim it

Each row gets one of three actions.

• Pay for it when the ask is the lab’s time or materials, through a fee-for-service agreement or a university core facility, which already bills outside customers. At UW-Madison, for example, external commercial users must pay at least full cost recovery, up to the market rate.
• Trade for it when the lab values something you control: rights to publish its own data, extended access to the instrument after the alpha ends, or field application scientist (FAS) support for the lab’s next experiment.
• Trim the ask when part of it serves your convenience more than your evidence. Send a short agreement the contracts office has approved before. Keep one standardized core protocol across sites, because comparable data is evidence. Keep weekly contact too, because people respect what you inspect, but shorten each call to 30 minutes. That halves the lab’s time to 5 hours.

Before: “We’d value your feedback on our new imaging system during a 10-week evaluation.”

After: “We will pay for 76 postdoc hours and the donor cells. You keep the right to publish your data and the instrument for three months after the alpha. Weekly check-ins take 30 minutes, and we want to hear where the system fails.”

Your VP will ask what the better offer costs. Put it next to the cost of a site that goes quiet.

Example figures. The delay column assumes a product forecast at $2 million in first-year revenue and a one-quarter slip in the decision to move it to the next stage.

Run the audit on your next alpha site request before it goes out. If more than half the rows say Pay for it and your alpha site budget is zero, you have found why your alpha sites stall.

Hold your half of the token

Jupiter Hospitalis guarded a bond in which each side held half the token and owed the other something real. Most alpha site programs run on a one-sided version of that bond, and the cost arrives as slow agreements, biased sites, stalled data and polite debriefs, just when you need evidence to move the product to the next stage. If you lead a team of PMs, run the audit across the portfolio, because five PMs asking the same ten labs for favors wears out a whole company’s welcome. Before your next alpha site request goes out, decide row by row what you will pay for, trade for or trim.