One page that takes you from research question to close-out, in the order the work actually happens. Each step says what to produce, what usually goes wrong, and where software helps.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Upload protocol
PDF, DOCX or DOC
Describe it
Plain-language summary
Phase 1 template
8 visits, ready to edit
The 12 steps at a glance
Orientation
A clinical trial is a planned study in people that tests a question under a written protocol. The work is the same shape whether the trial is a university pilot or a pivotal study: decide what you are testing, write it down, get independent approval, recruit and consent participants, collect data in a controlled way, and analyse it honestly. The international standard for how this should be done is ICH Good Clinical Practice, which emphasises participant protection and reliable data. Capture states its alignment with ICH-GCP E6(R2) and does not claim E6(R3) compliance.
If you are choosing between doing the work yourself and hiring a contract research organisation, see how to run a clinical trial without a CRO and how to run a clinical trial with a small team. For the list of tools involved, see tools needed to run a clinical trial. This page is the sequence that ties them together.
Phase A: plan
State the population, the intervention or exposure, the comparison and the one outcome that will decide the study. Everything else, including sample size and visit schedule, follows from it.
Randomized or single-arm, blinded or open, parallel or crossover, and how long. Estimate sample size with realistic dropout using a sample size calculator. If you are unsure, a pilot study can test feasibility first.
The protocol sets out eligibility, procedures, safety monitoring and analysis. The schedule of assessments lists what happens at every visit. Use the clinical trial protocol template as a structure.
Cost drivers are sites, monitoring, data management, and participant visits. Use the timeline calculator and read how much a clinical trial costs. Name who is responsible for each function, even if one person holds several roles.
Phase B: approve
Independent review of the protocol, consent form and recruitment materials is required before enrolling participants in most settings. Prepare the package with the IRB submission checklist. A blank eCRF PDF export from Capture can support the submission.
Work out whether your product or device needs a regulatory submission or notification, and whether the trial must be registered. For example, the journal editors' body ICMJE expects trials to be registered in a public registry before enrolment, and in the US, applicable clinical trials must be registered on ClinicalTrials.gov within 21 days of enrolling the first participant. Check the specifics for your trial with your regulator or a regulatory professional; this is not legal advice.
Approval waits are the best time to build the study. Do it in the free sandbox, with no credit card; pay only when you go live.
Phase C: build
Create the eCRFs, visits, edit checks and roles. This is where an EDC and the eCRF builder come in. Test everything with sample data. For a hands-on plan see how to set up a clinical trial database in a day.
Write consent in plain language and decide how it will be collected, on paper or electronically with eConsent. Set up questionnaires or diaries through ePRO if participants report outcomes. See how to write an eConsent form.
Train site staff, write a data management plan and a monitoring plan, and run a full dry run with test participants. File the approvals and documents in your study file.
Phase D: run and close
Screen against the eligibility criteria, record screen failures, consent before any study procedure, then follow the visit schedule. Track enrolment against target weekly.
Record adverse events and review serious ones promptly. Resolve queries as they arise rather than at the end, and use risk-based monitoring to focus source checks where they matter. See risk-based monitoring and SDV.
When data are clean and queries closed, lock the database so no further change is possible without a documented unlock, then analyse, report and archive. See database lock and the lock and unlock workflow.
In practice
Academic and early-stage teams often have one person covering several roles. That works if the roles are written down. The principal investigator owns the protocol and medical decisions. A coordinator handles screening, consent and visits. Someone owns data: building forms, reviewing queries and exports. Someone owns safety review. Where one person holds two roles, avoid reviewing your own work: have a colleague check forms you built and queries you raised.
Put the minimum documents in one place from day one: the approved protocol and amendments, ethics approvals, consent versions, a delegation log of who may do what, training records and the data management plan. The delegation log template is a simple start. Keeping this file tidy while the study runs is far easier than reconstructing it at close-out.
Pitfalls
| Step | Common mistake | Fix |
|---|---|---|
| 1 to 3 | Too many endpoints and visits | Cut anything that will not change a decision |
| 5 | Submitting incomplete documents | Use a submission checklist before sending |
| 7 | Building forms after approval, then discovering protocol gaps | Build during the approval wait and test against the protocol |
| 10 | Consent after a study procedure | Block procedures until consent is recorded |
| 11 | Cleaning data only at the end | Auto-queries and weekly review |
Software
An EDC with an audit trail, edit checks and role-based access.
Paper or eConsent with version control and signatures.
ePRO or diaries if outcomes come from participants.
Built-in randomization if you have more than one arm.
A study file maintained to your sponsor's procedures.
Large enterprise suites. Many early studies run on one all-in-one platform; Capture supports Phase 1 to 3.
Define the research question and the primary endpoint. Design, sample size, protocol and budget all follow from it.
Yes, many sponsor- and site-led studies do. You take on the functions a CRO would perform, so plan who covers data management, monitoring and safety. See the dedicated guide.
It depends. ICMJE expects registration before enrolment for journal publication, and US applicable trials must register on ClinicalTrials.gov within 21 days of first enrolment. Confirm your obligations with your regulator. This is not legal advice.
It varies widely with ethics review, regulatory steps, contracting and site readiness. Build the study database while approvals run, so it is not on the critical path.
At minimum, an EDC for data capture with an audit trail. Add eConsent, ePRO and randomization as the protocol requires. Capture combines them.
Yes. The free sandbox includes every feature with no credit card, and you pay only when you go live with real participants.
Yes. Capture supports Phase 1, 2 and 3 studies.
Keep exploring
How to run a clinical trial without a CRO
Doing the work in-house.
Tools needed to run a clinical trial
The stack behind the steps.
IRB submission checklist
Prepare step 5.
Clinical trial timeline calculator
Plan step 4.
How to run a pilot study
Test feasibility first.
How to choose an EDC system
Pick the system for step 7.
Build and test your study in the free sandbox. No credit card, pay only when you go live.