Everything you need to know about using CollabCoder for your research.
CollabCoder is built with research data in mind. Here's what we provide so you can make an informed decision for your IRB submission:
When AI tag suggestions are used, the selected passage is sent to Anthropic's API for processing. Anthropic does not use API inputs to train models, and data is not retained after processing.
FERPA applies when student education records are involved. CollabCoder itself doesn't store student records, but if your research data contains student information, FERPA considerations may apply.
The same security infrastructure described above protects all data in CollabCoder. We recommend consulting your institution's compliance office if your research involves FERPA-protected data. They can assess whether our security measures meet their requirements based on the specifics of your study.
All data is stored on Google Cloud infrastructure (Firebase/Firestore) in the United States. Google Cloud maintains SOC 1, SOC 2, SOC 3, and ISO 27001 certifications. Your data is encrypted both in transit (TLS) and at rest (AES-256).
Yes. CollabCoder runs automated daily backups of all Firestore data to Google Cloud Storage. Backups are retained so your data can be recovered in the event of an issue.
Only authenticated users who have been explicitly added as members of your project can access its documents, tags, and highlights. There is no public access, no shared links, and no admin backdoor. Project owners control who has access.
When you select a passage of text and request suggestions, CollabCoder sends that passage along with your codebook (tag names and descriptions) to Anthropic's Claude API. The AI analyzes the passage against your tags and returns the most relevant matches, each with a brief rationale.
You review each suggestion and decide whether to apply it. The AI never auto-tags, never modifies your data, and never makes coding decisions on your behalf. It's designed to speed up your workflow while keeping your analytical judgment at the center.
No. CollabCoder uses the Anthropic API, which does not use API inputs or outputs for model training. Your research data is processed for the tag suggestion and is not stored or retained by Anthropic after the response is generated.
Absolutely. AI tag suggestions are optional and only triggered when you explicitly request them. You can do all your coding manually by selecting text and choosing tags from your codebook. The AI is there when you want a second opinion, not as a requirement.
The more descriptive your codebook tags are, the better the suggestions. When you add descriptions to your tags (e.g., "Frustration: Participant expressing annoyance or dissatisfaction with a process"), the AI has more context to work with and provides more accurate recommendations.
CollabCoder supports TXT, DOCX, PDF, RTF, CSV, XLS, XLSX, Markdown, and HTML files. All files are converted to text for coding. Note that scanned or image-based PDFs won't work since there's no text to extract.
Open your project, scroll to the Collaborators section, and enter your teammate's email address. They'll need to have a CollabCoder account with that email. Once added, they can see all documents, tags, and highlights in the project.
Yes. CollabCoder is fully responsive and works on iOS and Android browsers. You can select text, apply tags, add notes, and review coding on mobile. No app download required.
Yes. You can export all coded excerpts for a document as a CSV file. The export includes the document name, tag, coded text, notes, coder name and email, position offsets, and timestamps.
Yes. CollabCoder supports bulk codebook import. You can paste your tags one per line, and optionally include descriptions separated by a colon or pipe character. You can also import tags from a CSV file.
The free plan includes 1 project with up to 2 documents and 2 collaborators. You get access to all core features including AI tag suggestions, bulk codebook import, CSV export, and mobile support. It's a full version of the tool, not a limited trial.
Your existing data is never deleted. If you downgrade from a paid plan to free, you'll still be able to access all your existing projects and documents. You just won't be able to create new projects beyond the free tier limits until you upgrade again.
Yes. The Academic plan is $7/month or $59/year for users with a .edu email address. It includes everything in the Pro plan at a reduced price.
Yes. You can cancel your subscription at any time. You'll continue to have access to your paid features until the end of your billing period, then your account reverts to the free plan. No data is deleted.
We'd love to hear from you. Reach out and we'll get back to you as soon as we can.
Contact Us