Every company's value comes from two kinds of capital. Human capital is what its people know: their skills, judgment, relationships, and instincts. Non-human capital is what accumulates and stays: plants, cash, patents, code, brand.
Life science tools manufacturers are very good at turning the first into the second. A scientist's insight becomes a patent. A process engineer's troubleshooting becomes an SOP and a validated manufacturing line. A product manager's conversations with customers become a product. Over 10 years of building genomics tools at QIAGEN, Bio-Rad, and Fluent BioSciences, I watched that conversion happen inside every company I worked for. In many ways, it is what a manufacturer is.
Then the product ships, and the conversion stops.
After the sale, knowledge lives in people
Once a kit leaves the loading dock, almost everything a manufacturer learns about how it performs comes through people. The field application scientist who knows which kind of lab always rushes the wash steps. The support scientist who can tell from three sentences in a ticket that the problem is the centrifuge, not the chemistry. The sales rep who hears over coffee why a lab quietly switched vendors.
That is human capital, and human capital can't be owned, only rented. When an experienced FAS leaves, the company loses years of field knowledge and often the customer relationships that came with it. After the sale, a manufacturer works much like a law firm: its understanding of its own customers walks out the door whenever someone changes jobs.
Why the old systems fall short
Manufacturers have tried to capture this knowledge with CRMs, ticketing systems, call reports, and annual surveys. Each one compresses a rich, messy event into a crude record and throws away the context.
A ticket might say "low yield." It won't say that the lab ran out of the kit's wash buffer and substituted their own, that the incubation ran long because the scientist got pulled into a meeting, or that this was the third failed run before anyone called. The outcome survives. The reasons don't, and the reasons are the valuable part, because they tell you what to change in the protocol, the documentation, or the product.
Incentives work against capture, too. Scientists have no reason to document a run for the manufacturer's benefit; they already spend too much time documenting for themselves. FAS teams are rewarded for solving the customer's problem, not for writing it up afterward. So capture depends on busy people volunteering extra work, and mostly it doesn't happen.
A third kind of capital
The capital I'm describing is created by customers every time they run a product. The unit I keep coming back to is the protocol execution record: a step-by-step account of what a scientist actually did, including the order, the timing, and every deviation from the written instructions. One record describes a single run. Thousands of them become a map of where a protocol bends under real conditions, which steps generate questions, and which deviations predict failure.
What makes this possible now is that capture can become a byproduct of the work rather than an extra task. Scientists wear gloves and move fast; nobody is going to type into a tablet mid-protocol. But a scientist can talk through a run hands-free, get help in the moment, and leave behind a structured record without doing anything extra. The context a ticket strips away gets captured as it happens.
That data has to be governed carefully. It starts with the scientist and their lab, and what a manufacturer sees has to respect that (I wrote about this in What a protocol execution record should never contain). Someone also has to curate what the system learns, because an accepted answer isn't always a correct one. The people best positioned to do that are the FAS and support scientists who carry the field knowledge today.
Capital that compounds
This is what separates customer capital from an archive of support tickets: it improves with use. Every run produces signal. The signal improves the guidance, better guidance improves the next run, and the next run produces better signal. The tenth lab to hit a problem shouldn't have to rediscover what the first nine already learned, and a scientist on their first run with a new kit can draw on everyone who came before them.
It also changes the math of scale. A manufacturer with a thousand active labs learns faster than one with a hundred. FAS teams stop answering questions the company has already answered and spend their time on problems that need their judgment. Their expertise becomes more valuable, not less, because every hour of their work adds to something the company keeps.
Manufacturers already know how to build capital inside their own walls. The next advantage goes to the ones that keep building it after the product ships.
I founded Adept Scientific, which makes Kavia, a free voice copilot that guides scientists through protocols, and Spotlight, a dashboard that shows manufacturers how customers run their products.