A PhD interview presentation is a short talk, often around 10 minutes, in which you show a selection panel that you can do research, that you have a focused project in mind and that the project fits their group. In about ten slides you summarize your past research, present a research proposal with a clear question, method and timeline, and explain why this lab suits the work. The panel is not grading slide design. It is deciding whether to supervise you for several years, and your talk opens that conversation. Below: a slide structure with timings, how to present each part, likely questions and an AI workflow.
What PhD interview panels look for
Formats vary between countries, departments and funding schemes. Some interviews ask for a proposal of your own, some for a talk about your master's project and some for your take on an advertised project. Follow the brief in your invitation first. Across formats, panels tend to listen for three things:
- Research fit. Does your question sit within what the group works on, and could someone here supervise it well?
- A feasible proposal. Is the question narrow enough to answer within a PhD, using methods, data and resources you can realistically get?
- Evidence from past research. Have you carried a piece of research from question to result, and can you explain your own role in it?
Behind all three sits a quieter test: how you think when questioned. So build a deck that invites good questions instead of saying everything.
A 10-minute, 10-slide structure
Use this skeleton for a typical 10-minute slot and adapt it to your brief:
| # | Slide | What goes on it | Time |
|---|---|---|---|
| 1 | Title | Working project title, your name, current degree, program applied to | 15 sec |
| 2 | About you | One line on your background and the thread that connects it | 30 sec |
| 3 | Past research: question and method | Your previous project, with your role made clear | 1.5 min |
| 4 | Past research: key result | One figure or finding, and what it taught you | 1 min |
| 5 | The problem | Why the question you want to study matters | 1 min |
| 6 | Research questions | One aim, two or three specific questions | 1 min |
| 7 | Approach and methods | One diagram linking each question to a method and data source | 2 min |
| 8 | Timeline and feasibility | Phases across the PhD, data access, main risk | 1 min |
| 9 | Why this lab | Overlap with the group's work, methods or facilities | 1 min |
| 10 | Summary | Question, approach and expected contribution in three lines | 30 sec |
About half the time goes to the proposal (slides 5 to 8) and a quarter to past research. If your brief asks you to focus on previous work, flip the balance.
How long should a PhD interview presentation be? As long as the invitation says. Ten minutes is a frequent request, but slots from 5 to 20 minutes are also common, so ask if the instructions are unclear. Plan on roughly one slide per minute:
| Time limit | Slides | What to change |
|---|---|---|
| About 5 minutes | 5–6 | Drop the about-you slide, keep one past-research slide, merge problem and questions |
| About 10 minutes | 8–10 | Use the structure above |
| 15–20 minutes | 12–16 | Add a second past project or a pilot result, split methods by question |
Presenting past research in two slides
Past research is supporting evidence, not the main event. Build the two slides in this order:
- State the question in one sentence. What did you want to find out, and why?
- Say which part was yours. For shared projects, be precise: "I designed the survey and ran the analysis; fieldwork was shared."
- Show one result. Use a single figure and write the finding as the slide title. "Adding survey weights reversed the effect for rural respondents" tells the panel far more than "Master's project".
- Close with what carries forward. A method, a skill or an open question that leads into your proposal.
List any publications in a small line at the foot of the slide instead of reading them out. If you have not published yet, a clearly explained thesis project does the same job.
The proposed project: question, method, timeline
Most questions will come back to this part.
The question: narrow beats ambitious
Compare "How does climate change affect cities?" with "How does street-level heat exposure differ between social housing estates and neighbouring private housing in one city, and which design features explain the gap?" The second tells the panel what you will measure, where and why. Make one aim the headline of slide 6 and list two or three questions under it.
The method: one diagram
On slide 7, place each research question on the left, the method and data in the middle and the expected output on the right: a chapter, paper or dataset. If you depend on data or equipment you do not yet have, say how you would get access.
The timeline: phases, not promises
Sketch the plan in phases across your program: background reading and a pilot, the main studies, then writing. Name one realistic risk and its fallback, and present the proposal as a starting point you expect to refine with your supervisor. A modest plan you can defend is more reassuring than a grand one you cannot. For a fuller proposal structure, see our thesis proposal presentation guide.
Why this lab
Many applicants turn generic here. Be specific:
- Name the group's line of work that your question extends, and one or two of its papers you have actually read.
- Point to a method, dataset or facility here that your project needs.
- Say what you would bring, such as a skill or a technique that complements the group.
- If you have already spoken with the potential supervisor, reflect that discussion.
One concrete connection says more than a paragraph of praise.
Questions to prepare for, and mistakes to avoid
Panels often ask some version of these:
- Why do you want to do a PhD, and why now?
- Why this question, and why in this group?
- What exactly was your contribution to your previous project?
- How does your proposal differ from existing work?
- What would you do if your main method or data source does not work out?
- Which part of your plan are you least sure about?
Lead each answer with the point, then the reason. If you have not considered something, say so and reason through it aloud.
Common mistakes:
- Too much background. Four minutes on the field leaves no time for your plan.
- A proposal sized for a career. Scope it to what a PhD can answer.
- Reading dense slides aloud. Keep one idea per slide and talk to it.
- Blurred authorship. Be clear about what you did and what your team did.
- Expected results shown as findings. At this stage you have hypotheses, not data.
- Running over. Rehearse with a timer. Our defense speech script guide shows how to script each slide.
From research statement to editable slides with AI
Once your research statement or proposal exists, turning it into slides is mostly splitting and layout. In our AI slide studio:
- Gather your source. Combine the proposal with a short summary of your past project, under clear headings.
- Upload, paste or import. Upload a .docx, .pdf, .txt or .md file, paste up to 100,000 characters, or import a web page by URL. Text-based PDFs are read in your browser; scanned PDFs need OCR first, or paste the text.
- Edit the outline. The AI drafts a slide-by-slide outline first, and drafting it uses no credits. Rename, reorder, add or delete slides to match the structure above; what you confirm is exactly what gets generated (why outline first).
- Generate. Decks run from 3 to 40 slides at 2 credits per slide, AI images included, and the cost is shown before you start: 10 slides is 20 credits. A deck takes about one to three minutes, and credits come back automatically if a generation fails. Credit packs start at $5 for 150 credits; monthly plans start at $9.
- Export and check. Download an editable PPTX with real text boxes for PowerPoint, Keynote (via import) or WPS, or export PDF or PNG. Exports carry a small "AI-generated" label. Check every number, name and citation.
The AI organises the content you give it and does not invent results or references. What you get is a high-quality, editable first draft to shape in your own voice. Check your target university's rules on AI tools as well. To see an academic deck built from real input, open the thesis defense example.
FAQ
How long should a PhD interview presentation be?
Follow the time given in your interview invitation. Around 10 minutes is a frequent request, but slots from 5 to 20 minutes are also common, usually followed by questions. If the brief is unclear, ask the department. Plan on roughly one slide per minute and rehearse until you finish slightly under the limit.
How many slides should a 10-minute PhD interview presentation have?
Around 8 to 10 slides is a common range for 10 minutes. Give two slides to your past research and four or five to the proposed project, because that is where most questions come from. Fewer, clearer slides work better than a crowded deck.
What should I present if I am applying for an advertised PhD project?
Present your reading of the advertised project instead of an independent proposal: the question as you understand it, one or two ideas for how you would approach it, and how your skills fit. Keep it tentative. The panel knows the project better than you do and wants to see how you think, not a finished plan.
Can I make my PhD interview slides from my research proposal PDF?
Yes. Upload a text-based PDF and the text is read in your browser. You can also upload .docx, .txt or .md files, or paste up to 100,000 characters. Scanned PDFs have no text layer, so run OCR first or paste the text. The AI drafts an editable outline before any slides are generated and works from your content without inventing results or references.
Editorial Team