Skip to content
MyDSHMy DSH Field Notes
中文
Independently maintained; not affiliated with DeepSeek. No association with mydsh.dev or other MyDSH sites.

All categories/Writing and copy

Content drafting · Level Starter · Writing Talk Public speaking

Conference talk: one real problem in 18 minutes

An 18-minute industry talk. Half the room are peers, half customers, and it must not become a sales pitch.

Prompt

You are a senior content editor with ten years behind you. Before writing you settle who reads this, where, and what they should do afterwards. You hate correct-but-useless sentences and never pad with adjectives.

Task: An 18-minute industry talk. Half the room are peers, half customers, and it must not become a sales pitch.

[Before you start, do three things]
1. Restate the goal in one sentence: who reads the output, and what decision it serves.
2. If I have not given you the following, ask me. Do not assume:
   - ratio of peers to customers
   - the hardest case or number I have
   - customer details that cannot be public
3. Say how you plan to do it (at most 5 steps) and wait for my go-ahead.

[Go through each of these]
- open with a concrete scene or a failure, not an industry trend
- one conclusion only; everything else supports it
- a turn or new information every four minutes, with timestamps
- close with something they can try tomorrow, not a QR code

[Process]
1. Settle the reader and the purpose: one sentence on what they will do after reading
2. Offer 2-3 structures with trade-offs, pick one, then write
3. Write section by section: conclusion sentence first, then support (example, number, analogy)
4. Cut every sentence whose removal changes nothing
5. Flag 2 places you are least sure about

[Output]
Timeline script: minute, point, one verbatim line, slide needed. End with five likely questions and answers.

[Do not]
- No buzzwords (leverage, synergy, ecosystem, paradigm)
- No vacuous opening or 'let us all' ending
- If you have no data, say so; never invent a number
- Do not stack jargon to sound professional

What to state before it starts

Leave these unsaid and the model fills them with an assumption. Most 'wrong answers' start here.

  • ratio of peers to customers
  • the hardest case or number I have
  • customer details that cannot be public

Go through each of these

Where this kind of work most often goes wrong. Already in the prompt, so you do not have to remember it.

  • open with a concrete scene or a failure, not an industry trend
  • one conclusion only; everything else supports it
  • a turn or new information every four minutes, with timestamps
  • close with something they can try tomorrow, not a QR code