Skip to main content

Work

Conference Notes Template

Most conference notes are written once and never opened again. This template captures three points in your own words, the claims worth checking, the people to follow up with and one next step, so the session still pays off a month later. It works the same for a live talk, a podcast episode or a course video.

Updated 2026-10-27

Copy and adapt

The template

Conference notesTemplate

[Session title], {{date}}

Source: [conference and room / podcast and episode number / course and module] Speaker(s): [name, role, organization] Why I'm here: [the question I want this session to answer] Slides or recording: [link, or timestamp range 00:00 to 00:00] Related: [[Event, podcast or course name]]

Key points (three, in my words)

  1. [Point], evidence: [data, slide number or timestamp]
  2. [Point], evidence: [data, slide number or timestamp]
  3. [Point], evidence: [data, slide number or timestamp]

Numbers and quotes to check

  • "[Exact words]" ([slide or timestamp])
  • [Statistic], source the speaker cited: [source or "none given"]

What I think

  • [Where I agree, where I do not, and what does not fit my situation]

Q&A

  • Q: [question] ([who asked])
  • A: [answer]

People

  • [Name, role]: [what we talked about], follow up via [channel]

Resources mentioned

  • [Paper, book, tool, repository or link]

Actions

  • [Next step] @due(YYYY-MM-DD)

One-line takeaway

[What I would tell a colleague who asks whether it was worth it]

5 examples

Examples for different situations

The same structure, filled in for the situations people actually bring to it. Copy the one closest to yours.

1.Tech conference talk

When it fits: Use this when your company paid for the ticket, you are attending a talk that touches a decision your team is about to make, and you will be asked what you learned on Monday.

Tech conference talkExample 1

Retiring a monolith one queue at a time, 2026-11-05

Source: Cascade Platform Summit 2026, Hall C, Portland Speaker(s): Halima Rostova, staff engineer, Ledgerline (payments) Why I'm here: Is an event queue a safer way to split our order service than the API gateway approach we sketched in September? Slides or recording: slides promised on the event site by Nov 12 Related: [[Cascade Platform Summit 2026]], [[Order service split]]

Key points (three, in my words)

  1. They routed writes through a queue first and moved consumers out one at a time, evidence: 38 consumers extracted over 14 months (slide 9).
  2. Dual writes were the main source of incidents, evidence: 6 of their 9 sev-2 incidents in year one (slide 15).
  3. They kept the monolith as the source of truth until each consumer had run in shadow mode for 30 days.

Numbers and quotes to check

  • "Shadow mode found more bugs than our test suite did" (slide 17)
  • p99 latency on checkout went from 420 ms to 310 ms, source the speaker cited: internal dashboard, not published

What I think

  • Shadow mode for 30 days is realistic for us. 14 months is not; we have 6 engineers, they had 40.
  • Our order service already emits events for billing, so half the queue exists.

Q&A

  • Q: How did you handle ordering guarantees? (attendee from a retailer)
  • A: Partition by customer ID, accept out-of-order across customers.

People

  • Halima Rostova: asked about their shadow-mode diff tool, follow up via the email on her last slide
  • Viktor (SRE at a logistics company, met at lunch): same migration, 4 months in, follow up on LinkedIn

Resources mentioned

  • Strangler fig pattern write-up, linked on slide 3
  • Their open-source diff tool for comparing shadow outputs

Actions

  • Two-page proposal: queue-first split for the order service (@me) @due(2026-11-14)
  • Email Halima about the diff tool (@me) @due(2026-11-07)

One-line takeaway

Queue-first works if you budget for shadow mode; the dual-write period is where it hurts.

2.Medical CME session

When it fits: Use this when you are a clinician at a continuing medical education session, the content may change how you treat patients, and you need the credit details and the claims you must check against the guideline before acting.

Medical CME sessionExample 2

Getting HFrEF patients on all four drug classes, 2026-10-16

Source: State Academy of Family Physicians Annual Scientific Assembly, Ballroom B Speaker(s): Dr. Samuel Achterberg, cardiologist, regional heart failure clinic Why I'm here: Our practice starts patients on two classes and rarely gets further. What does a realistic titration schedule look like in primary care? Slides or recording: session recording in the CME portal for 90 days Related: [[CME 2026]], [[Heart failure registry project]]

Key points (three, in my words)

  1. Start all four classes at low doses within weeks rather than maxing one before adding the next, evidence: speaker cited the current US heart failure guideline (slide 6).
  2. Cost and prior authorization block the SGLT2 inhibitor more often than side effects do, evidence: his clinic's audit, 41 percent of stalled patients (slide 11).
  3. Nurse-led titration visits every two weeks moved his clinic's rate on all four classes from 22 to 58 percent in one year.

Numbers and quotes to check

  • 22 to 58 percent, source the speaker cited: his clinic's internal audit, not published
  • Potassium threshold for holding the MRA: check against the guideline and our pharmacy protocol before using

What I think

  • The nurse-led visits fit our practice. We have one RN who already does diabetes titration.
  • The audit numbers are one clinic. Useful as a direction, not a benchmark.

Q&A

  • Q: What about patients over 80 with eGFR around 30? (attendee)
  • A: Individualize, start one drug at a time, check labs at one week.

Resources mentioned

  • Current ACC/AHA/HFSA heart failure guideline
  • His clinic's titration flowsheet (handout, photographed)

Actions

  • Run an EHR report: HFrEF patients and how many classes each is on (@me) @due(2026-11-06)
  • Ask our pharmacist about SGLT2 inhibitor coverage on our top three plans (@me) @due(2026-10-30)
  • Claim 1.0 CME credit and complete the evaluation (@me) @due(2026-11-15)

One-line takeaway

The gap is process, not knowledge: a two-week titration visit is the change worth testing.

3.Academic conference session

When it fits: Use this when you are a graduate student at a discipline conference, most talks are 12 minutes, and the useful output is a list of papers to read, a dataset to ask about and a sharper sense of where your chapter fits.

Academic conference sessionExample 3

Urban pollinator networks session (4 talks), 2026-08-06

Source: Annual ecology society meeting, Session 112, Room 220 Speaker(s): four talks, 12 min each; noted here: K. Lindgren (postdoc, urban ecology lab), "Bee foraging distance across 60 city parks" Why I'm here: Is anyone else measuring foraging distance in parks smaller than one hectare? That is the gap I claim in chapter 2. Slides or recording: no recording; abstract number 112-3 in the program Related: [[Thesis chapter 2]], [[Ecology meeting 2026]]

Key points (three, in my words)

  1. Foraging distance dropped sharply in parks under 2 hectares, evidence: regression on slide 8, 60 parks, 3 summers.
  2. Their smallest park was 0.8 hectares. Nothing below that, so my gap still holds.
  3. Mowing frequency explained more variance than park size (slide 10).

Numbers and quotes to check

  • "Mowing explains twice the variance of area" (slide 10), check in the paper when published
  • Sample: 2,140 marked bees, source the speaker cited: preprint, not yet reviewed

What I think

  • Mowing is a confound I have not controlled for. Need to add it before my committee meeting.

Q&A

  • Q: Did you control for nest site availability? (Dr. Haverford)
  • A: Partly, using soil cover as a proxy. Acknowledged as a limit.

People

  • K. Lindgren: asked whether mowing data is shareable, follow up via email after the meeting
  • Pilar (PhD student, poster 44): same species, different city, follow up to compare protocols

Resources mentioned

  • Lindgren et al. preprint, link in the abstract
  • City parks department mowing schedules (public records request)

Actions

  • Read the preprint and add it to my literature notes (@me) @due(2026-08-14)
  • Email Lindgren about the mowing data (@me) @due(2026-08-11)
  • Add mowing frequency to my field protocol before next season (@me) @due(2026-09-30)

One-line takeaway

My gap holds, but mowing is a confound I have to measure.

4.Podcast episode

When it fits: Use this when a podcast episode said something you want to act on at work, you were listening on a walk, and you need the timestamps so you can quote it accurately instead of from memory.

Podcast episodeExample 4

Episode 212, usage-based pricing for small SaaS, 2026-10-22

Source: The Margin Notes podcast, episode 212 (58 min) Speaker(s): host Corinne Abara, guest Ikaika Mahoe (former head of pricing at a developer tools company) Why I'm here: We charge per seat and our biggest customers use the product least. Is usage-based pricing worth a test? Slides or recording: episode page, timestamps below Related: [[Pricing experiments 2027]]

Key points (three, in my words)

  1. Usage pricing works when usage tracks value the customer can see, evidence: 14:20 to 17:05.
  2. A hybrid (small platform fee plus usage) reduced churn in their case because bills stopped jumping, evidence: 23:40.
  3. Customers hate surprise invoices more than high invoices. Spending caps were the fix, evidence: 31:15.

Numbers and quotes to check

  • "Net revenue retention went from 104 to 118 percent" (24:10), source the guest cited: none given
  • "Most usage-priced companies over-index on the wrong metric" (28:50), opinion, not data

What I think

  • Our value metric is probably reports generated, not API calls.
  • 118 percent is one company's number with no source. Will not quote it internally.

Resources mentioned

  • The guest's essay on value metrics, linked in the show notes

Actions

  • Pull report-generation counts per account for the last 6 months (@me) @due(2026-10-30)
  • One-page memo: hybrid pricing test on new accounts only (@me) @due(2026-11-13)

One-line takeaway

Find the metric customers already watch, price on that, and cap the bill.

5.Recorded course video

When it fits: Use this when you are working through an online course alone, you can pause and rewind, and the risk is passive watching: notes need exercises you tried and the points where you got stuck.

Recorded course videoExample 5

Module 4, lesson 3: SQL window functions, 2026-11-02

Source: online data analysis course, module 4 of 8, video 3 (18:40) Speaker(s): course instructor Why I'm here: I keep writing self-joins to get running totals at work. Want to replace them. Slides or recording: lesson page, timestamps below Related: [[Data analysis course]], [[Weekly sales report]]

Key points (three, in my words)

  1. A window function computes across rows without collapsing them, unlike GROUP BY, evidence: 02:10 diagram.
  2. PARTITION BY restarts the calculation per group; ORDER BY inside OVER sets the running order, evidence: 06:45.
  3. ROW_NUMBER, RANK and DENSE_RANK differ only on ties, evidence: 11:30 table.

Numbers and quotes to check

  • Instructor says window functions are always faster than self-joins (14:05). Test on our warehouse before believing it.

What I think

  • Paused at 09:00 and rewrote the weekly sales running total. Worked first try, 22 lines down to 9.
  • Got stuck on the frame clause (ROWS BETWEEN). Rewatched 15:20 twice.

Q&A

  • Posted in the course forum: does the frame default differ between our warehouse and the one in the course?

Resources mentioned

  • Practice dataset in the lesson downloads
  • Warehouse documentation page on window frames

Actions

  • Replace the self-join in the weekly sales report (@me) @due(2026-11-06)
  • Time both queries on last quarter's data (@me) @due(2026-11-06)

One-line takeaway

Running totals and rankings are one OVER clause, but read the frame default for your database.

What belongs in conference notes

The top conference note templates share a header (event, session, speaker, date), a notes area and a takeaways box. That covers the session. What they leave open is what happens to the notes afterwards, which is where most of the value is lost. This template adds four things for that:

  • The question you came with. It decides what is worth writing down.
  • Numbers and quotes to check. Speakers round, cite internal dashboards and repeat figures from other talks. Anything you might repeat at work gets its source or a "none given."
  • Your own reaction, kept separate. What fits your situation and what does not.
  • One action and one sentence. The action is what makes the ticket worth it. The sentence is your answer when someone asks how it was.

Skip the transcript. Slides and recordings are usually published within a week. Your notes are the part nobody else can publish: what mattered to you and what you will do about it.

One note per session, one hub per event

Long events defeat single-document notes. Fifteen sessions in one file become a scroll nobody rereads. Write one short note per session and link each to a hub note for the event. The hub holds the schedule, the people you met and, at the end, the trip report. The same shape works for a podcast you follow (a hub per show) and an online course (a hub per course, a note per lesson).

Live talks, podcasts and recorded videos

The same structure works as a podcast notes template or a video notes template: swap slide numbers for timestamps. The habits differ.

  • Live talks. Write fragments with slide numbers. Fill in What I think in the five minutes between sessions, while you still remember why you reacted.
  • Podcasts. You are often walking or driving. Note the timestamp on your phone, or record a short voice memo, and write the note when you stop. A quote you cannot replay is a quote you will misremember.
  • Course videos. Pause and try the exercise before writing about it. "Worked first try" or "stuck on the frame clause at 15:20" is more useful later than a summary of what the instructor said.

The trip report

Within three working days, write one page for your team from the hub note:

  1. The two or three sessions worth their time, with links to your notes.
  2. What you propose to try, and the smallest version of it.
  3. People worth knowing, and why.
  4. Anything you heard that you are not sure is true yet.

If each session note ends with a one-line takeaway, this takes twenty minutes. If it does not, it usually does not get written.

Common mistakes

  • Copying slides. You will get the deck. Write what you would not get from it.
  • Repeating a statistic without its source. "The speaker said 118 percent" becomes "it is 118 percent" by the second retelling.
  • No owner or date on actions. "Look into the diff tool" is not an action. "Email Halima by Friday" is.
  • Waiting until the flight home. By then three sessions have blurred into one. Five minutes after each session is enough.

When to use a different template

For a semester of classes rather than a one-off talk, the lecture notes template adds review questions and spaced revision. For a paper you heard about and then read, move to the research notes template or the literature review template. For a meeting where decisions are made, use the meeting notes template. If you searched for conference notes because of a parent-teacher conference, the IEP meeting notes template includes a filled-in parent-teacher conference example.

Customization tips

  • Write the question before the session starts

    The "Why I'm here" line takes thirty seconds in the hallway and decides what you write down. Without it, notes become a transcript of slides you can download later. With it, you write the three things that answer your question and skip the rest.

  • Keep a hub note for each event, course or podcast

    One note per session, all linked to one note for the event. The hub holds logistics, the list of sessions, and the trip report you write at the end. A three-day conference with fifteen sessions stays navigable instead of becoming one long scroll.

  • Use timestamps for anything recorded

    For a podcast or a course video, write the timestamp next to every point and quote. It costs two seconds and means you can replay the exact moment before quoting it to a colleague. For live talks, the slide number does the same job.

  • Separate what was said from what you think

    Key points are the speaker's argument in your words. "What I think" is your reaction. Keeping them apart stops a speaker's opinion from turning into a fact in your notes, and makes your trip report more honest.

  • Drop sections that do not apply

    A course video rarely has people to follow up with. A hallway conversation has no slides. Delete empty sections instead of leaving them blank, so the note is quick to reread.

In the app

How to do this in Scribelet

  1. Open Desk Settings, then Templates, tap Add Template, paste the template above and set the category to Research. Add #conference or #podcast as a default tag.

  2. Create a hub note like [[Cascade Platform Summit 2026]] and link each session note to it. The hub's backlinks panel lists every session you attended, and the auto-linker (Pro) suggests links between related notes.

  3. Photograph slides or handouts and attach them to the session note. Scribelet extracts the text, so searching for a term from slide 15 finds the note it came from.

  4. Walking out of the room, record a voice note (Pro) with your three points and next step. Scribelet transcribes it and structures it into a note with headings and action items.

  5. Turn on background verification for notes with statistics you plan to repeat. It checks factual claims against the web and proposes corrections you accept or dismiss, which is the honest way to handle a number like "118 percent" with no source.

  6. Weeks later, ask the AI chat "which sessions at the summit talked about dual writes?" It answers from your notes with citations.

Templates, template variables and AI generation are documented in Note templates.

Frequently asked questions

How do you take notes at a conference?

Decide what question you want each session to answer, write three points in your own words rather than copying slides, note any number you might repeat with its source, and write one next step before you leave the room. Slides are usually published; your reaction and your follow-ups are not.

How do I take notes from a podcast or a video?

Use the same template with timestamps instead of slide numbers. Pause to write rather than writing while listening, and for a course video, try the exercise before writing about it. The podcast and course examples above show both.

What is a conference trip report?

A one-page summary for your team or manager written within a few days: the sessions worth sharing, three takeaways, what you propose to try, and who you met. It is easy to write if each session already has a one-line takeaway, as this template does.

Should I take conference notes on paper or a laptop?

Either works if the notes end up in one searchable place. Paper keeps you listening; a laptop is faster to search later. A common compromise is paper or phone during the session, then a few minutes typing the template that evening.

How many notes should I take per session?

For a 30 to 45 minute talk, three key points, one or two numbers to check and one action is usually enough. If you are writing more than a page, you are probably copying slides you can download.

Conference notes you will actually open again

Save this template in Scribelet, link every session to one hub note for the event, attach your slide photos, and ask your notes next month which talk said what. Free to start.

Free plan, no credit card. See pricing

Study

Class Notes Template

One note per class session, linked to one note per course. The template captures what the lecture was trying to answer, the key points in your words, the questions you still have, and what is due next. Copy it, or start from the example closest to your class.

Research

Research Notes Template

Research notes go wrong in two ways: you cannot tell where a claim came from, or you cannot tell whether it is still true. This template ties every finding to a source and a confidence level, keeps observation apart from interpretation, and records when you last checked.

Study

Book Summary Template for Book Notes

A book summary is worth writing when it answers three questions a year later: what was the argument or story, what did I make of it, and what did I change because of it. Copy the template, or start from the example closest to the book on your nightstand.

Meetings

Meeting Notes Template

One structure that captures the three things people come back for: what was decided, who owns what by when, and the context that made the decision make sense. Copy it as-is, or pick the example closest to your meeting.

Research

Literature Review Template

A literature review is an argument about what is known, built from sources, not a list of summaries. This template takes you from scope and search, through a source matrix, to one section per theme with agreement, disagreement and gaps. Pick the example organized the way your sources group.

Article

How to remember what you read

From the Scribelet blog

We use cookies for analytics to improve your experience. Learn more