PDF
Your PDF
Paste your PDF. Nothing to upload.
←PDF
Make a interview guide
Turn PDFs into an interview guide
Questions built from what the documents left unresolved, not from what they already answered.
Free to try, no card. Your PDF opens in Ancher.
Document structure in orderPage numbersTables as tables
Kept in parts
Every one of these stays separately addressable instead of collapsing into one block of text.
GoalWarm-upCore questionsProbes
Interview guide
Questions ordered to get past the answer people rehearse.
The point of reading before an interview is to stop asking things the documents already tell you. Ancher separates what the sources establish from what they leave open, builds the core questions out of the open half, and writes probes aimed at the specific gaps — a number with no method behind it, a claim only one document makes, a recommendation with no finding under it.
The honest version
Every question the documents already answer is a minute you do not get back
An interview has maybe forty minutes of real attention in it, and the fastest way to waste them is to open with things a prior report states plainly. Reading the documents to find what they do not settle is the preparation that changes the conversation, and it is the part that gets skipped because it is slower than skimming.
One page of one PDF, at p. 9, read two ways.Illustrative. The figure is invented; the gap it falls into is not.
the improvement was substantial, though the paper does not report
In Table 2 — not readThe specific never reaches the draft
Interview guide
In Table 2
no control group listed
from p. 9
Lands in the draft, with its timecode
- Core questions come only from what the documents leave unresolved.
- Probes name the specific gap — the figure with no method, the claim with one source — so the follow-up is concrete.
- Anything the documents settle is listed separately so it is not asked by accident.
What comes out
Interview guide, section by section
Questions ordered to get past the answer people rehearse.
Goal
What this conversation has to establish that the documents cannot.
Warm-up
Openers grounded in something specific from the documents, which signals you have read them.
Core questions
Questions built from the unresolved half, ordered so the hardest one is not first.
Probes
Follow-ups naming the exact gap, with the page it appears on.
Close
What to confirm before ending, including anything the documents contradict each other on.
Documents-to-interview-guide prompt
The link on its own gets a summary. This is the instruction that produces the 5 sections above, in that order, with the rules that keep them honest. Paste it with your PDFs — in Ancher, or in whatever assistant you already use.
Build an interview guide from these documents. (1) State what this conversation must establish that the documents cannot. (2) Write openers grounded in something specific from them. (3) Build core questions only from what they leave unresolved, ordered so the hardest is not first. (4) Write probes naming the exact gap and its page. (5) List what to confirm at the close, including any contradiction between sources. Separately list what the documents already settle, so it is not asked.
The detail, if you want it
How each part of a PDF becomes a section
| From the PDF | Into | Why |
|---|---|---|
| Document structure in order | Core questions | Headings arrive as headings, which is what lets a limitations section be found — and limitations are where the questions are. |
| Page numbers | Probes | A page anchor lets you cite the document in the room, which changes how a question is received. |
| Tables as tables | Probes | Tables keep their structure, so a probe can name the exact cell that has no method behind it. |
| Reading order | Warm-up | Sequence survives, so an opener can reference where in the argument something appeared rather than just that it did. |
| Embedded images and figures | Close | Figure captions state findings tightly, which makes them the cleanest thing to confirm at the end. |
Everything Ancher keeps from a PDF
Document structure in order
Headings, body and captions arrive as separate elements, so the document's own outline survives ingest.
Page numbers
A page-level anchor is what makes a citation checkable by someone else holding the same file.
Tables as tables
Tables come back as their own elements rather than as flattened lines, so a figure stays attached to the row and column that give it meaning.
Embedded images and figures
Figure captions frequently carry the tightest statement of a finding in the whole document.
Reading order
Elements come back in sequence, which is what lets a section be reassembled rather than guessed at.
Word and PowerPoint too
The same path takes .docx and .pptx, so a deck and its write-up can sit in one collection.
The four steps
- Add the PDF. Upload the file. Ancher parses it into ordered elements rather than one flat string, keeps the page number on each, and pulls out the embedded images.
- Nothing gets flattened. Document structure in order, page numbers, tables as tables, embedded images and figures, reading order, word and powerpoint too stay separately addressable instead of collapsing into one block of text.
- Ask for a interview guide. Ancher fills goal, warm-up, core questions, probes, close using the mapping above, not a generic summary pass.
- Check it before you use it. Every specific keeps a link back to where it came from in the PDF, so a claim can be verified without reopening the source.
Where this goes wrong
- A guide built from the documents' contents rather than their gaps produces an interview that reads the report back to its author.
- Documents disagree more often than people expect. Bringing a contradiction to an interview is useful; bringing it as an accusation is not.
- Probes that cite a page are powerful and can read as prosecutorial. Order them so the conversation is established before the sharpest one arrives.
Questions people ask
How is this different from just reading the documents?
Reading tells you what they say. This separates what they settle from what they leave open, and only the open half becomes questions — which is the step that usually gets compressed into a skim the night before.
Should I cite pages in the room?
Selectively. Naming a page shows you read it and makes a vague answer harder, but doing it in every question turns a conversation into a deposition and people close down.
What if the documents contradict each other?
That goes in the close, as something to confirm. A contradiction between two sources is one of the highest-value things an interview can resolve, and it is usually invisible until they are read together.
Bring your own PDFs
Everything you save lives in one workspace, so the interview guide is built from your sources — not from a model's memory of the internet.
Open Ancher →Not a PDF? The same interview guide also comes from recording, web page.