Rebuild Last Year's 60-Slide Lecture (Without Starting Over)

Every teaching attending has the folder. `lecture_FINAL_v3.pptx`. Sixty slides. Built two years ago for a talk you gave once, half of it copied from a fellow who copied it from someone who has since left the program.

It is technically yours. It is also unteachable: walls of text, a citation style that is just journal names in parentheses, no speaker notes, and three slides whose data you no longer trust.

The instinct is to start a new deck. That is the wrong move. The structure already encodes hours of clinical thinking. The problem is not the content — it is the packaging. You rebuild it, you don't rewrite it.

Why Rewriting From Scratch Is the Trap

Rewriting a 60-slide lecture from a blank page feels productive because it is visible work. It is also the slowest possible path.

The old deck already contains the answers to the expensive questions: what order the concepts go in, which cases illustrate which point, what depth the audience expects. Throwing that away to retype it means re-deciding everything you already decided.

The faster path treats the old deck as a source, not a draft. You feed it in, and the rebuild keeps the skeleton — section order, case selection, teaching arc — while fixing the layer that actually fails an audience: density, citations, and notes.

What Actually Breaks a 60-Slide Inherited Deck

Long decks fail in predictable, structural ways. Naming them is most of the fix:

• Density — slides built to be read, not presented. Six bullets where one line and a spoken sentence would land.
• Phantom citations — "(NEJM 2019)" with no DOI, no first author, no way for a learner to find it.
• Stale data — an incidence figure or a guideline grade that was current when the slide was made and is not now.
• No speaker notes — the reasoning that connects slide 14 to slide 15 lives only in the head of whoever first gave the talk.
• No editable output — a flattened PDF or an image-export deck you cannot adjust when the clinical nuance changes.

Notice that none of these are knowledge problems. They are all packaging problems — which is exactly why a rebuild beats a rewrite.

The Rebuild Workflow for a Long Deck

A 60-slide deck is large enough that the rebuild has to be deliberate. The workflow:

• Upload the existing .pptx or .pdf as the source — the whole deck, not a trimmed version. The structure is the value.
• Set the real teaching parameters: audience (residents vs. fellows vs. faculty) and target talk length in minutes. A 60-slide deck for a 30-minute slot needs to be cut, not just polished — say so up front.
• Let the rebuild work the deck in batches. A long lecture is processed in segments rather than one pass, so a 60-slide source doesn't silently time out or get truncated.
• Review the slide-level flags: dense, needs-citation, outdated-stat, low-confidence. These point you straight at the slides that actually need a human decision.
• Read the generated speaker notes as a continuity check. If the notes for two adjacent slides do not connect, the structure has a gap the original deck was hiding.
• Export the teach-ready .pptx. It stays editable — the final clinical nuance is yours to set, not frozen into an image.

Trust the Flags, Not the Whole Deck

With a long lecture, the temptation is to re-review all 60 slides line by line. Don't. That is the rewrite trap wearing a different hat.

The slide-level flags exist so your attention goes where judgment is actually required: a low-confidence rebuild on a mechanism slide, an outdated-stat flag on an epidemiology figure, a needs-citation flag on a claim you are about to defend in front of faculty.

A flagged slide is a 30-second decision. An unflagged slide is one the rebuild is confident it preserved faithfully. Spend your review budget on the flags and you turn a midnight rewrite into a coffee-length pass.

The Honest Limits

A rebuild does not invent clinical truth. If the source deck cited a 2019 trial wrong, the rebuild flags the citation as weak — it does not silently substitute a paper it was not given. That is deliberate: a confident wrong citation in front of faculty is worse than a flagged uncertain one.

It also will not make a 60-slide deck into a good 10-minute talk by magic. If you ask for a length the content cannot support, the cut shows. The parameters you set at the start are doing real work — set them honestly.

The Point

The lecture you dread rebuilding is the one you already gave. The structure is done. The thinking is done. What's left — density, citations, notes, an editable export — is exactly the mechanical layer worth handing off.

Rebuild the deck you have. Save the blank page for the talk you have never given.