Can Biotechs ever truly escape Microsoft Excel?

Team KC's voice on where spreadsheets belong, where they break things, and what could close the gap.

Early-stage biotech discovery runs on Excel. Every ELN/LIMS project comes with the implicit promise of replacing it. Yet spreadsheets never fully disappear. So, we put the question to our team: Can biotechs ever truly escape Excel? Among Laboratory Informatics Specialists Xandria Kovacic, Angel Kleiman, and Caitlin Tormey, none said yes, but the reasons are even less black-and-white. 

The honest answer: No, and that’s the wrong goal

Each team member landed in roughly the same place. The point isn’t to eliminate the tool entirely, but to decide what it’s allowed to hold. The goal should be to set a boundary, deciding deliberately what data belongs in the ELN for that stage, and then making sure the important data actually moves over.

“The question isn’t whether Excel disappears,” Xandria put it “it’s whether it’s still holding data that needs an audit trail, version control, or traceability, because that’s where things go wrong.” Caitlin echoed this point, saying “the goal doesn’t necessarily need to be to fully escape it, but rather to define when and where it should be used versus when more structured systems need to be in place”

Often, early-stage biotechs don’t have the resources to push what may be heavily customized analysis into an ELN or LIMS, if they have one in the first place, as Angel said, "You can move the structured data over, but the day-to-day might stay in Excel if they can’t go into an ELN, and that’s fine.

Why scientists keep reaching for it:

For early exploratory work, a spreadsheet does something enterprise systems are structurally bad at. It asks nothing of you up front. Even with an unstructured ELN entry, scientists run into limitations compared to features that exist natively in Excel, such as formulas in a table. As Xandria pointed out, most platforms have spreadsheet-like tables, "but few match Excel's Goal Seek, custom macros, or ad hoc formula flexibility. That's exactly why exploratory work still leans on Excel." Beyond that, a LIMS or ELN requires that the shape of the data be defined before you begin. Fields, types, and relationships are all reasonable demands once a workflow has settled, but early R&D is often too fluid for structure to be worth defining.

The flexibility that makes Excel dangerous as a system of record (it accepts anything and enforces nothing) is the same flexibility that makes it the right tool for discovery. Caitlin captured this appeal by saying Excel is "an unconstrained sandbox for early, exploratory R&D that rigid enterprise systems cannot replicate."

When Excel becomes a liability:

The case against Excel is not that it produces obvious or dramatic errors. It's actually the opposite. As Angel put it, "Excel tends to fail silently." Our team has seen two main types of this failure.

The first is a process problem. Excel has no audit trail; version history is not the same thing as a record of who changed which value and why. The classic result is version drift: two (or more) copies of the same sheet, numbers quietly diverging, and the wrong one making it into a final deliverable.

The second is a domain problem. Excel isn't biology- or chemistry-aware. It can't distinguish a valid construct from a mislabeled one, and it won't catch the kind of error Caitlin flagged in NGS sample sheets: "a simple copy-paste error can shift a row of index sequences, completely mismatching them with their sample IDs." It just sees text in cells. A purpose-built ELN or LIMS understands the data well enough to throw an error; a spreadsheet never will.

The answer vs. the solution:

Although the answer to the title question is “no”, that might be the right answer. The mistake would be to frame success as it’s disappearance. Success is actually much harder because it means ensuring Excel no longer holds data that can’t afford to live there. 

Great implementations will make the spreadsheet safe to use, as Angel put it “The system is the source of truth, and the spreadsheet becomes a place you pass through rather than where everything lives. That’s the whole game.”

Not sure where the line should be? Figuring out the balance between which spreadsheets should become workflows and which can be left alone is exactly the work we do. Because Karchem Consulting is vendor-agnostic, our recommendations are driven by your scientists and your data-integrity requirements. Whether you’re pre-IND or scaling toward cGMP, we can help you find that line. Get in touch today to see where to start!