Forty-Five Seconds to Make It Clear

Release mode Beta

Level
Beginner
Format
individual, pair, or small group

Turn one real idea into a spoken explanation for a real audience. By the end, you can select the details that matter to a listener, compare audience-specific versions, and explain what you removed, protected, accepted, or rejected rather than assuming shorter always means better.

Choose what matters when time is short

What you will make

Turn one real idea into a spoken explanation for a real audience. By the end, you can select the details that matter to a listener, compare audience-specific versions, and explain what you removed, protected, accepted, or rejected rather than assuming shorter always means better.

Lab materials

Use during the lab

Go deeper

Before you start

Level and format: Beginner; individual, pair, or small group

What you need

A timer, a place to read aloud, and—if you want drafting support—an AI chat tool.

Harness
Any general-purpose AI chat interface. You can also do the first pass without AI.
Subscription
Free tier is enough; no separate model or extended-thinking setting is required.

Privacy boundary: Do not put confidential plans or private client information in a public tool. Use a hobby, volunteer role, creative project, fictional service, or the classroom example if you would rather not use a personal or workplace topic.

The core idea

An elevator speech is selection, not just compression. A useful 45-second explanation gives one audience a clear reason to care and an appropriate next step. AI can suggest wording, but you decide the claim, the audience, and whether the script sounds like you.

The name comes from the familiar elevator-pitch situation: a short elevator ride—or a quick hallway conversation—may be the only opening you get to make your case. Forty-five seconds is a useful practice constraint, not a claim about the length of every elevator ride. The gallery’s timed examples are 93–105 words and run about 42–45 seconds at a conversational pace. Use roughly 90–105 words as a starting band, not a pass/fail rule; your pace, pauses, and phrasing decide the result. Read aloud and revise for breath, clarity, and natural voice.

Make your 45-second explanation

Complete the activity here. Open the optional Lab Notebook beside this page when you want a place to record your work.

  1. 🎯 Choose one useful thing to make clear. Pick a project, service, proposal, role, or topic you can describe honestly. Name one listener and what you want them to understand. If you need a low-risk starting point, open the topic starters, use a hobby, volunteer role, creative project, or fictional service. The optional 45-second worksheet can hold the work from this point forward.

  2. 🧭 State the goal before asking for help. If you use a chat tool, open with: “We are going to create a 45-second elevator speech, let’s start by gathering our thoughts on this topic: [topic].” Add the listener and, if useful, the next step you hope for. Do not ask for a speech yet.

  3. 🔎 Gather the thought through material questions. Ask the tool to pose one useful question at a time. Explore who the listener is, what gets difficult for them, what you actually do or mean, and why it matters. Answer only what is relevant and appropriate to share; stop after a handful of questions unless something essential remains.

  4. 🧩 Confirm alignment. State what the explanation is really about, the distinction or boundary it must protect, and what you do not want the tool to invent. Do not request a draft until you can say why the listener should care and what must remain true.

  5. ✍️ Request the first 45-second draft. You may ask: “Based on what we have agreed, draft a plain-language spoken 45-second elevator speech for [audience]. Do not add facts. End with [next step].” Keep the first version; it gives you something concrete to evaluate.

  6. 🛠️ Refine with human ownership. Compare the draft against clarity, natural language, specific value, authenticity, and a clear next step. Give a targeted refinement request—what to keep, remove, protect, or make more direct—rather than asking the tool to take over the message. Then make two additional versions for different audiences so you have three versions to compare.

  7. ⏱️ Time one version aloud and decide what belongs. Read one refined version at a conversational pace. Record the duration and underline any phrase you stumble over. Ask for the strongest reasonable question or objection a listener may have, then decide whether it belongs in the speech, a follow-up answer, or outside the scope.

  8. 🛑 Stop deliberately. Separate necessary changes from optional polish. Keep the opening prompt, a useful exploration exchange, the alignment statement, the first draft, a refinement exchange, all three audience versions, the final timed script, and a note about why you stopped.

Why this process works

The conversation has distinct moments. Keeping the human and chatbot responsibilities visible helps you use the tool without handing it ownership of the message.

Moment Human responsibility Chatbot contribution
Open the conversation State that the goal is a 45-second elevator speech and name the topic Reflect the task without inventing a purpose
Gather thoughts Answer only relevant, shareable questions and protect the claim Ask material questions that surface distinctions or overclaims
Confirm alignment Decide when the conversation has enough useful material Summarize the agreed direction for a human check
Request the draft Specify audience, boundaries, and the next step Draft spoken language without adding facts
Refine Keep, reject, or change wording deliberately Make targeted revisions rather than taking over the message
Test and stop Time the explanation aloud and decide when remaining changes are polish Offer constructive friction when asked

Check your result

Use this checklist to inspect your choices:

Directly check the time and ask a listener to say back the main point and next step. A fluent script is not proof that a claim is accurate, accessible, or appropriate for every listener. Human review still owns factual claims, organizational promises, and the decision to record or publish audio.

You are done when

One last check: A polished sound is never a substitute for accuracy, consent, or audience fit.

Try another version

These are optional next experiments, not additional completion requirements.

  1. 🏢 Change the room. Adjust the tone and detail for your boss, their boss, or the C-suite.
  2. 👟 Change the audience. Explain the same idea to a fifth grader, high-school student, or grandparent without talking down to them.
  3. 🎨 Change the medium. Create an infographic that carries the same central distinction.
  4. 🎬 Add a voice. Use that infographic with a voiceover to create a short video.
  5. 🤔 Invite friction. Ask a skeptical collaborator to name the strongest reasonable objection, then decide whether it belongs in the speech or the follow-up conversation.
  6. 🔁 Show the difference. Make a paired before-and-after version that shows what changed when you named a different listener.

Learner Map

Tracks

Modules

Help us improve this Activity

Tell us what helped, what slowed you down, or what you needed but could not find.

🎓 Give learner feedback