Research Data Management Services for Grant-Funded Projects
Manage, clean, document, and hand over research project data without hiring a full-time data manager. Reproducible pipelines, FAIR-aligned outputs, archival deposition — owned by your team, not a vendor.
Where research data management gaps actually live
After delivering data work for grant-funded teams across health, neuroscience, and policy domains, the same patterns appear over and over.
Schema drift during collection. Six months in, the dataset has columns nobody documented and a date format that flips between sites.
Cleaning that isn't reproducible. The published figures cite the cleaned dataset, but nobody can re-derive it deterministically from raw inputs.
Documentation in heads, not files. The PI knows why early-2025 measurements are excluded. When they leave, the institutional knowledge of the dataset goes with them.
FAIR commitments made aspirationally. The DMP promised FAIR-aligned outputs; nothing was done to actually produce them.
Funder data requests catch the team flat-footed. 'Where can a reviewer access the cleaned dataset?' becomes a frantic two-week project.
What we deliver
A research data management engagement takes your data from 'scattered on team laptops' to 'structured, documented, reproducible' in 4–8 weeks. Specifically:
Schema validation as code
Pydantic / JSON Schema definitions that catch drift at ingestion rather than at report time. Schema versions are explicit; changes are tracked.
Reproducible cleaning pipelines
One-command runs from raw → cleaned, parameterised where useful, version-controlled. Cleaning decisions are recorded in code, not in someone's memory.
Analysis pipelines that produce report figures
Final report figures regenerable by re-running the analysis script. No screenshots, no manual Excel steps, no orphaned numbers.
FAIR-aligned metadata and documentation
Data dictionary, README, decision log, license clarity. Deposition to Zenodo or your institutional repository with a citable DOI.
Handover pack the team can operate
Setup guide, runbook, fresh-laptop test passing, data-management-plan update. The team owns and maintains the infrastructure after we exit.
How a research data management engagement runs
Four to eight weeks, three phases, no retainer.
Audit (week 1)
We inventory your current data assets: what exists, where it lives, who touched it, what's documented. Output: a clear gap analysis between current state and FAIR-aligned target state.
Build (weeks 2–6)
Schema validation, cleaning pipelines, analysis scripts, metadata structure — all delivered as code your team owns. Iterative, with weekly checkpoints.
Hand over (week 7–8)
Documentation, fresh-laptop test, deposition to your archive of choice, DMP update, training session for the team. After handover, the infrastructure runs without us.
What this typically costs vs alternatives
A research data manager hire: €40–55K/year, plus three months of recruiting, plus ongoing supervision. Most grant-funded projects can't justify that recurring cost. A consultancy with a monthly retainer: €3–6K/month indefinitely. A Pragma scoped 4–8 week engagement: finite, your team owns the output as code, no retainer, no vendor lock-in. Pricing depends on project scope — the scope review tells you the exact figure for your project, before any commitment.
What 'doing nothing' looks like: schema drift goes uncorrected, the published figures cite a cleaned dataset nobody can re-derive, FAIR commitments stay aspirational. When the funder requests reviewer access to the cleaned dataset, the answer is a frantic two-week project. The cost: a chunk of your project's reputation at exactly the wrong moment. The cost of acting now: 60 minutes for a free scope review.
Research data management questions
Need the data side of your project handled?
Tell us your dataset's current state, your funder's expectations, and your timeline. We'll reply within 2 business days with a scope or with the questions we'd need answered.
Request a Scope Review