The Airtable base that runs your clinic intake was built by someone who left Mizzou two years ago
Replacing the Retool, Airtable, and spreadsheet stack that quietly runs a Columbia operation usually costs $35,000 to $120,000 over 2 to 5 months, depending on how many workflows you are consolidating. The trigger is rarely a feature gap. It is that the person who built the critical base left, the data now touches PHI or study records, and the tool everyone depends on has no owner, no audit trail, and no permissions worth the name.
Every Columbia clinic, lab, and insurance team has them: the Airtable base that schedules study visits, the Retool app that triages intake, the spreadsheet that tracks specimen consent. They were fast to build and they work right up until they handle real patient or research data and somebody asks who can see it and who changed it last.
The honest answer is usually nobody knows. Retool and Airtable are excellent for prototypes and small teams, but when a tool becomes load-bearing for regulated work, their permission models, audit gaps, and single-builder fragility turn into genuine risk. The clinic does not notice until an auditor asks for the access log.
Why the usual tools struggle in Columbia
- A research-visit-scheduling Airtable base with no owner since the grad student who built it graduated
- Retool intake apps where every staffer effectively sees every record because the permission model is shallow
- Specimen or consent tracking in spreadsheets that no auditor would accept
- Critical workflows with zero audit trail, so you cannot answer who changed what and when
What a custom internal tools build changes
Custom internal tools give the load-bearing workflows what the no-code stack cannot: real role-based permissions, a complete audit trail, validation that stops bad data at entry, and an owner who is your vendor or your team rather than a departed grad student. You consolidate the five fragile bases into one tool that your compliance office can actually defend, and you keep the speed by scoping tightly to what the team does every day.
- A no-code tool now handles PHI, consent, or regulated records
- The original builder is gone and nobody fully understands the tool
- Auditors or your IRB ask for access logs you cannot produce
- Five fragile bases should obviously be one maintained system
- The workflow is genuinely simple and non-regulated; Airtable is fine
- You are still prototyping and requirements change weekly
- The team is small and the data carries no compliance weight
- No budget or owner exists for a maintained custom tool
- Real role-based access so a coordinator sees their cohort, not every record in the system
- Complete audit trails that satisfy IRB, HIPAA, and insurance-regulator questions
- Validation at entry that stops the duplicate and malformed records spreadsheets let through
- One maintained tool replacing five orphaned Airtable bases and shadow spreadsheets
- Integrations to your EHR, study database, or claims system instead of copy-paste
- Slower to change than dragging a column in Airtable; iteration goes through a dev cycle
- Upfront cost where the no-code version felt free, even though it was not
- You need an owner for the tool, internal or vendor, or you recreate the orphan problem
- Over-engineering a genuinely simple workflow wastes money the spreadsheet was handling fine
The features that matter for Columbia
Internal Tools services we deliver in Columbia
Digital Heroes builds the full internal tools stack for Columbia teams. Typical engagements span:
Internal Tools pricing in Columbia: the real numbers
| Project scope | Typical cost | Timeline |
|---|---|---|
| Single workflow tool with permissions + audit | $25k to $50k | 1 to 2 months |
| Consolidated multi-workflow internal app | $55k to $95k | 3 to 4 months |
| Integrated tool with EHR/study-DB sync | $95k to $140k | 4 to 6 months |
From kickoff to launch: the schedule
Exactly what you get
The five fragile bases become one tool with real permissions, a full audit trail, and validation that keeps bad data out. Coordinators get the speed they liked from Airtable plus the controls your compliance office demanded. It connects to your EHR or study database so nobody re-keys, and it has an owner who is not about to graduate. These tools often pair with a custom-software core, a helpdesk for the team that runs them, and business-intelligence dashboards over the cleaned-up data.
How to choose a developer in Columbia
Find a team comfortable inheriting messy no-code systems and turning them into something auditable. Ask how they would migrate a live Airtable base that an IRB now cares about, and what permission model they would put around it. A good partner inventories your orphaned tools first, then consolidates, instead of cloning your spreadsheets into a slightly nicer spreadsheet.
- !A shop that just rebuilds your Airtable in Retool; ask what permission and audit model they add
- !No discovery of who owns each current tool; ask how they map the orphans before quoting
- !No data-migration plan; ask how they move live records without losing history
- !Ignoring your compliance office; ask whether they have shipped audit-grade internal tools
- !Promising a one-week rebuild of a regulated workflow; that pace is how the audit gaps got there
If internal tools is on the roadmap, custom software, wordpress, accounting usually follow within the year. Budget them as one conversation.
Rohan advises mid-market and enterprise teams on ERP, CRM and custom software, and has led delivery on dozens of business-software builds.
Writes for Digital Heroes, shipping business software for 2,000+ brands across 55+ countries since 2017.
Frequently asked questions
When does Airtable stop being good enough?
When the data becomes regulated and the tool becomes load-bearing. The moment a base holds PHI, consent, or study records and an auditor could ask for an access log, you have outgrown a permission model that was never built for that.
Can you migrate our existing Retool and Airtable data?
Yes. A proper build includes mapping every current base, migrating live records with their history where it exists, and validating nothing is lost before cutover. That migration is part of the timeline, not an afterthought.
Why not just keep building in Retool with better permissions?
For some teams that works. But once you need cohort-level access, full audit trails, and tight EHR integration, you often hit Retool's ceiling, and a custom tool gives you control without the platform constraints.