Free and low-cost qualitative research tools for PhD students include: manual coding in spreadsheets (free, works for small datasets), Quirkos (affordable, intuitive for beginners), Delve (free tier for basic projects), and Skimle (freemium, with AI-assisted thematic analysis included in the free tier). NVivo costs EUR 1000 or more per year per user and is rarely worth it for individual researchers — better, more accessible alternatives exist at every price point.
If you are a PhD student doing qualitative research, you have probably been told to use NVivo. Your department may have a site licence; your supervisor probably used it for their PhD. It has been the default tool for qualitative data analysis in academia for decades.
But NVivo's dominance is not the same as NVivo being the right tool for you. This guide is for researchers who want to do rigorous, publishable qualitative analysis without spending their entire studentship allowance on software — or without learning a tool that takes a week of training to operate at a basic level.
Why NVivo is the default (and why that is worth questioning)
NVivo became the academic standard in an era when the alternatives were manual card-sorting and basic word processors. It built an enormous feature set over 25 years: advanced querying, social network analysis, matrix coding, mixed methods integration, and much more.
For large-scale, team-based qualitative research projects — particularly those involving complex multi-method designs or very large corpora — NVivo is genuinely powerful. A research centre running a five-year funded project with a team of researchers and 200 documents has legitimate uses for its full feature set.
A PhD student with 30 interview transcripts, working alone, uses perhaps 15% of NVivo's functionality. And pays for all of it.
The cost. As of 2026, NVivo Student pricing starts at EUR 115 for a one-year licence in the EU markets with students needing to cough up an additional 300 EUR for AI and transcription features . The full academic licence for researchers with institutional access varies, but where institutions have not purchased site licences, individual researchers pay full commercial rates. The detailed NVivo pricing analysis covers the full cost structure.
The learning curve. NVivo training workshops at universities run for two or three full days. The interface is powerful but not intuitive — researchers frequently describe spending more time learning the tool than they expected before being able to use it productively.
The collaboration problem. NVivo's cloud collaboration features have a troubled history. Sync failures, version conflicts, and project file corruption are persistent complaints in user forums. For team projects, this is a serious risk.
What you actually need from a qualitative analysis tool
Before evaluating alternatives, it is worth being clear about what the tool needs to do. For most PhD qualitative research, the requirements are:
- Code text segments and assign codes with labels
- Group codes into broader themes or categories
- Store quotes with their context and source attribution
- Navigate between themes and sources — see all quotes tagged with a given code
- Support iterative refinement — add, rename, and merge codes as the analysis develops
- Export findings in a format usable for writing and presentation
That is it. A surprisingly small feature set. Most qualitative analysis tools — and some non-specialist tools — handle this adequately.
The real options
Manual coding in a spreadsheet (free)
For datasets of up to around 10 interviews, manual coding in a well-structured spreadsheet is a completely viable approach. The workflow:
- Paste each transcript into a spreadsheet, one paragraph per row
- Add columns for: participant ID, codes applied, notes, and memo content
- Use filters and sorting to group by code and see all instances of a theme together
This sounds low-tech because it is. It is also transparent, flexible, and requires no software training. The limitation is scale: beyond 20 interviews, the cognitive load of managing codes manually becomes significant.
Quirkos
Quirkos is a genuinely pleasant qualitative analysis tool designed to make the coding process intuitive rather than technical. It uses a "bubble" visualisation where codes appear as bubbles that grow as more content is assigned to them — a useful visual for understanding the relative weight of themes.
Pricing: Around 10 EUR per month for a student licence — significantly cheaper than NVivo. There is also a free trial period that covers a full project for evaluation.
Limitations: Less powerful for complex team projects or very large datasets. No AI-assisted analysis. Better for structured manual coding than automated theme discovery.
Delve
Delve is a web-based qualitative analysis tool with a free tier that covers basic coding and project management. It is approachable, cloud-based (which makes collaboration easy), and well-suited to the basic coding workflow without a steep learning curve.
Pricing: Free tier is meaningful for individual projects; paid tiers add team features and larger project capacity.
Limitations: Less powerful than NVivo for complex queries. AI features are limited. Works best for projects where you have a clear coding framework in advance.
Skimle (freemium, with AI-assisted analysis)
Skimle takes a different approach from traditional QDAS tools. Rather than providing a manual coding environment, it applies structured thematic analysis automatically — processing your transcripts and producing a category hierarchy with insights and supporting quotes. The researcher then reviews, refines, and extends the structure.
For PhD research, the practical benefit is speed: Skimle can process 30 interview transcripts and produce an initial thematic structure in the time it would take to manually code three of them. This leaves more time for the interpretive work that actually requires human judgement.
Methodological legitimacy. The question that matters for publication is whether AI-assisted analysis is methodologically defensible. This depends on which journal, which methodology, and how the analysis is documented. The AI in qualitative research guide for academic researchers covers this in detail, including how to document AI assistance transparently in your methods section.
Skimle's approach — structured thematic analysis with full traceability from every theme to every source quote — is closer to the methodological standards of reflexive thematic analysis than a black-box AI summarisation. Every claim is backed by evidence the reader can verify.
Pricing: Skimle has a free tier that includes meaningful functionality for individual researchers. See pricing.
REFI-QDA compatibility. Skimle supports REFI-QDA export, which means you can take the AI-generated structure into NVivo or ATLAS.ti for further manual refinement if your institution requires or prefers those tools.
MAXQDA
MAXQDA is a full-featured QDAS tool that competes directly with NVivo at a somewhat lower price point. It is particularly strong for mixed methods research (combining qualitative and quantitative analysis) and has a reputation for being somewhat more user-friendly than NVivo.
Making the choice
| Your situation | Recommended tool |
|---|---|
| Small dataset (under 15 interviews), solo researcher | Manual coding or Delve free |
| Moderate dataset, manual coding preferred, tight budget | Quirkos student licence |
| Mixed methods or complex project, need full QDAS | MAXQDA (cheaper than NVivo) |
| AI-assisted analysis acceptable in your field, want speed | Skimle |
| Team project, institution has site licence | Whatever the site licence covers |
One note on institutional access: before purchasing any tool, check what your university library or research office provides. Many UK and European institutions have NVivo site licences that make it free for enrolled students. If yours does, the cost argument disappears — though the learning curve and collaboration problems remain.
A note on methodological transparency
Whichever tool you use, your methods section needs to explain your analytical approach clearly enough that a reader could evaluate it. The tool is secondary to the method. "I used NVivo to code my transcripts" describes the software, not the methodology. "I conducted reflexive thematic analysis (Braun & Clarke, 2006, 2021), using iterative inductive coding followed by thematic review" describes the methodology — the software is a footnote.
If you use AI-assisted tools like Skimle, document that clearly. State what the tool did (initial thematic structure), what you did (reviewed, refined, extended the structure through iterative analysis), and how you ensured rigour (checking every theme against source quotes, reflexive memos, peer review of theme definitions). See the AI in qualitative research guide for specific language you can adapt.
Ready to run your qualitative analysis without spending your PhD funding on software? Try Skimle for free and see how far AI-assisted thematic analysis can take you.
Related reading:
- How to use AI in qualitative research: a guide for academic researchers
- Qualitative data analysis tools: a complete comparison
- How to code qualitative data: inductive, deductive and abductive approaches
About the authors
Henri Schildt is a Professor of Strategy at Aalto University School of Business and co-founder of Skimle. He has published over a dozen peer-reviewed articles using qualitative methods, including work in Academy of Management Journal, Organisation Science, and Strategic Management Journal. Google Scholar profile
Olli Salo is a former Partner at McKinsey & Company where he spent 18 years helping clients understand their markets, develop winning strategies and improve their operating models. He has done over 1000 client interviews and published over 10 articles on McKinsey.com and beyond. LinkedIn profile