iSectra — Believe IT
Pharmaceutical

IT From Seed Stage to IPO: What Changes at Each Stage

What changes at each stage, which cheap early decisions prevent expensive remediation, and what auditors want to see before an S-1.

Santiago Aristizabal Sep 16, 2026 8 min read
IT From Seed Stage to IPO: What Changes at Each Stage

Across the life of a biotech, the technical requirements change less than people expect (but the evidentiary requirements change a lot. You move from "it works" to "prove it worked, and prove nobody could have changed it," and the move happens in discrete steps tied to regulatory and commercial milestones, not to headcount. We took CinCor Pharma from its first employee through IPO and on to acquisition by a major pharmaceutical company, and we have supported another pharmaceutical client from three employees through their IPO and every audit since (sixteen years). What follows is drawn from watching those arcs and the same arc in others.

What actually forces the change?

Free IT Risk Assessment

See how your IT actually stacks up.

Twelve questions, about four minutes, and your score appears on screen immediately — no sales call required.

Take the assessment

Headcount is a poor proxy. A 12-person company running a GMP process has obligations a 60-person company outsourcing everything does not. What forces change is the milestone, and there are five that matter: first GxP activity, first regulatory submission, first partner or customer audit, first commercial product, and the S-1. Each one raises the evidentiary bar (what you have to show, to whom, and over what period), and the cheap decisions that prevent expensive remediation are the ones you made before the milestone, not during it.

Seed and preclinical

What matters here: identity and access from day one, a real backup with a tested restore, and not putting research data in personal accounts. What doesn't matter yet: validation, formal change control, most of what vendors will try to sell you.

The cheap decisions at this stage that prevent expensive remediation later are almost entirely about attribution. The single most important is a proper tenant with named individual accounts (and no shared logins, ever. Retrofitting attribution onto a system that ran on shared credentials is close to impossible: you cannot reconstruct who did what from a login that was four people. The second is multi-factor authentication on everything from day one, which is free or near-free and prevents the one breach that ends a preclinical program. The third is a documented offboarding process, so a departing co-founder or contractor's access is gone the same day they are, not discovered in a SOC 2 review two years later. None of these cost money. All of them cost real money to retrofit.

First GxP activity

This is the step change. The moment a system touches a GxP process (GMP manufacturing, GLP study data, GCP clinical records), the expectations shift from "it works" to "it works and we can prove it." That means a computerized system inventory, a risk classification of what's in scope and why, validation for the systems that are, change control on changes to them, and audit-trail review as an operational practice rather than a feature you bought.

What companies typically go back and fix here: spreadsheets that were never classified as systems but perform GMP calculations; audit trails that were enabled but never reviewed; shared instrument logins retrofitted to individual accounts; and backup systems that exist but have never had a restore tested. The reason is always the same (the system was built to run, not to be evidenced, and the milestone forces the second frame onto the first. Read the Part 11 regulation for the electronic-records framework and FDA's Data Integrity and Compliance With Drug CGMP guidance for how the agency applies it; together they define what "prove it" actually requires.

Clinical stage

Clinical brings data volume and external parties. The new problems are CRO and vendor access, data transfer agreements, retention obligations that now outlive the study, and the first serious questions about where data physically sits.

The decisions that matter: where CRO data lives (your tenant or theirs), how access is granted and revoked, and what the data transfer agreement actually says about ownership and retention at study close. A CRO that holds the primary copy of study data in its tenant, with access to your systems as a courtesy, is a very different posture from the reverse (and which one you have is decided in the contract, not discovered at exit. Retention obligations for clinical data routinely exceed the life of the study by years; a backup that was adequate for a 50-person company is not adequate for a dataset you are legally required to keep for a decade.

First partner or customer audit

This is usually the first time someone external asks for evidence rather than assurances. Security questionnaires arrive (often SOC 2-aligned), and the realization lands that "we use Microsoft 365" is not an answer to any of the questions being asked.

The questions a partner asks are about controls operating over time: who has access, who reviewed that access, how changes are authorized, how incidents are handled. A company that has been making the early decisions above has the answers; one that hasn't has to build the evidence in a hurry, which is the expensive version. A security questionnaire answered from documentation that already exists is a week of work; one answered from memory and reconstruction is a quarter.

Commercial stage

Commercial brings multi-site operation, manufacturing or manufacturing oversight, 24/7 operational dependency, and the point at which downtime has a revenue number attached to it for the first time. The IT decisions at this stage are about availability and recovery, not just integrity: redundant connectivity, a documented disaster-recovery plan with tested RTO and RPO, and a support model that covers the hours the business runs. A GxP environment that ran during business hours is not the same as one that ships product around the clock, and the change is structural (the systems, the staffing, and the contracts all have to reflect it.

Pre-IPO and the S-1

The S-1 brings SOX IT general controls into scope. What that means concretely: formal segregation of duties, evidence of access reviews performed on a schedule, change management with authorization and testing, and the expectation that auditors will want to see the controls operating over a period (not existing on the day they ask. A control you implemented last quarter has no operating history; a control you've been running for eighteen months does. This is why pre-IPO IT readiness is a months-long preparation, not a quarter-end task. See the PCAOB auditing standards on internal control over financial reporting for the framework external auditors apply; the practical point is that the controls have to exist and to have been run, with evidence, before the audit clock starts.

The companies that handle this well are the ones who were running access reviews and change logs as a matter of course from the commercial stage (because at that point they had to), so the SOX walk-through is a relabeling, not a build.

What to do early that pays off later

Decisions that cost nothing at seed stage and save six figures of remediation later:

  1. 1.Individual named accounts in a proper tenant, no shared logins ever (attribution cannot be retrofitted.
  2. 2.Multi-factor authentication everywhere, from day one (free or near-free, and the single control that prevents the catastrophic breach.
  3. 3.Documented offboarding (access revoked the same day a person leaves, with a record.
  4. 4.One authoritative place for research data (not a founder's personal laptop, not a personal cloud account.
  5. 5.A system inventory maintained from the first system (a spreadsheet is fine; the point is it exists and is kept current.
  6. 6.Tested restores (a backup that has never been restored is not a backup, and the test is free.

Almost nothing on this list costs money. All of it costs real money to retrofit at clinical stage, because retrofitting means rebuilding evidence for decisions that were never recorded.

When to bring in help

Honest framing: most companies don't need a full IT function early. They need someone accountable for the decisions above (someone who owns the tenant, the access, the backups, and the inventory, and who knows where the next milestone is. The mistake isn't outsourcing or hiring; it's nobody owning it. A fractional or managed arrangement with a clear owner is better than a full hire nobody defined the job for, and worse than a full hire with a defined job (but all three are better than the default, which is that the co-founder who "knows computers" holds it in their head until the first audit.

The stages, at a glance

What changesSeed / preclinicalFirst GxPClinicalFirst auditCommercialPre-IPO / S-1
What's in scopeIdentity, access, backupGxP systems in the inventoryCRO/vendor access, data transferControls operating over timeAvailability, recovery, 24/7SOX IT general controls
What auditors or partners ask forLittleValidation, audit trails, access listData ownership, retentionSOC 2-aligned evidenceDR plan, RTO/RPO, support hoursControls operating over a period
Common gapShared logins, personal accountsUnvalidated spreadsheets, unreviewed audit trailsCRO holds primary data, retention unclearNo documentation to answer fromDR plan exists but untestedControls implemented at quarter-end, no history
Decision to make earlyNamed accounts, MFA, tested restoresInventory + risk classification from day oneContract decides data ownershipRun access reviews as a matter of courseBuild DR and test it before revenue depends on itStart ICFR controls 12-18 months out

The pattern is the same across the companies we've walked through this arc: the ones who handle the transitions cleanly aren't the ones who spent early. They're the ones who made the free decisions early, so the expensive ones at each milestone were small. Almost nothing on that list costs money at seed stage; all of it costs real money to retrofit at clinical stage, and more at pre-IPO. For the regulated environment this sits inside, see pharmaceutical IT; for the electronic-records rules that govern the GxP step, 21 CFR Part 11 for executives; and for what shows up when an investigator arrives at one of these stages, what FDA investigators look at in your IT systems.

Hero photo: "black and white spiral staircase" by Nicolas Hoizey on Unsplash.

Frequently asked questions

Don't see your question? Email info@isectra.com and we'll answer it directly.

Share
Pharmaceutical
SA
President & COOSantiago Aristizabal

Santiago Aristizabal is President and COO at iSectra, where he oversees the design and operation of regulated IT environments for pharmaceutical and life-sciences clients through every stage of growth.

Get the next post in your inbox.

Signal, once or twice a month. No spam — just the posts worth reading.

Let's see what your IT could feel like.

A senior engineer will review your environment and respond within 24 hours. No sales script, no obligation.

(845) 563-0346
Call