Design once, deliver in three modes: Video, audio, and text that actually moves readiness

Why multi-format matters for readiness, not just preferences

Busy learners don’t all engage the same way, but the goal isn’t to appease preferences—it’s to increase access and retrieval strength while assessing the same competency under realistic constraints.

What we see across SageStepper cohorts (n=3,200 learners, 2024–2025):

• Readiness score lift: Learners who practiced in at least two formats saw +22 points average lift vs. +11 in single-format tracks.
• Offer rate: Candidates maintaining a 10+ day multi-format streak had a 1.4x higher offer rate in tech, healthcare, and consulting pathways.
• Mentor hours saved: Programs that shipped text + audio variants alongside video reduced live mock load by 28–36% without lowering rubric rigor.

The takeaway: Deliver the same skill in multiple modes to increase reps and lower friction, while keeping assessment consistent. That’s the bar.

Step 1: Map competencies to formats (design once, atomize responsibly)

Start with competencies, not content types. For each competency, decide which modes expose the skill authentically and where alternatives are equivalent.

Use a simple content map so you don’t reinvent:

# content-map.yaml
cohort: "Spring 2025 Product + Data"
competencies:
  - id: COMM_STRUCT
    name: Structured Communication (STAR/CAR)
    primary_format: video
    alternates: [audio, text]
    assessment: "Clarity, logical flow, evidence-based examples"
  - id: PROB_DECOMP
    name: Problem Decomposition
    primary_format: text
    alternates: [video]
    assessment: "Identify constraints, subproblems, trade-offs"
  - id: SQL_QUERY
    name: SQL Querying Under Time Pressure
    primary_format: text
    alternates: [video walkthrough]
    assessment: "Correctness, performance, reasoning"
  - id: PROD_SENSE
    name: Product Sense (Ambiguity Handling)
    primary_format: audio
    alternates: [video, text]
    assessment: "Hypothesis framing, prioritization, metrics"

Action steps:

1. List 5–7 competencies and tag a primary format plus at least one alternate.
2. Define one rubric per competency that applies across formats.
3. Store maps in your repository or in SageStepper’s action boards for shared visibility.

Step 2: Script once, publish in three modes (SageStepper workflow)

You don’t need three separate productions. Create a master script, then atomize.

Core workflow inside SageStepper:

1. Use Nova to draft 3–5 scenario variants per competency. Edit tone and difficulty to match your level.
2. Record a crisp video explaining the prompt and constraints (2–4 minutes). Upload or record natively.
3. Auto-generate the audio cut and transcript. SageStepper produces captions.vtt, clean transcript.txt, and flashcards from key points.
4. Publish a text-only version with the same prompt and timebox using Interview Sprint templates.
5. Attach the shared rubric and link items to the competency on your progress radar.

Template to keep outputs consistent:

# Master Prompt Template (Copyable)
- Competency: {{competency_id}}
- Scenario: {{30–60 word brief with role, audience, constraint}}
- Task: {{exact deliverable: explain, draft, query, framework}}
- Timebox: {{3–7 minutes}}
- Evidence to show: {{bullets}}
- Common traps: {{bullets}}

If you need offline alternatives, export with one click. For teams without native tools, here’s a fallback:

# Optional: manual audio extraction if working outside SageStepper
ffmpeg -i prompt_explainer.mp4 -vn -acodec libmp3lame prompt_explainer.mp3

Quality guardrail: Before publishing, run an equivalence check—two mentors complete the audio and text variants and score with the same rubric. Aim for <5% variance in average scores.

Step 3: Keep rigor high with shared rubrics and cross-format equivalence

Rigor comes from consistent evidence and scoring, not the medium. Build one rubric per competency and apply it across formats.

Sample rubric structure:

# rubric.yaml
id: COMM_STRUCT_R1
competency: COMM_STRUCT
criteria:
  - id: clarity
    label: Clarity & Brevity
    levels:
      1: "Wanders; unclear ask or outcome"
      3: "Mostly concise; minor tangents"
      5: "Crisp, high-signal, no filler"
  - id: logic
    label: Logical Structure (STAR/CAR)
    levels:
      1: "No structure; events out of order"
      3: "Partial structure; skips key step"
      5: "Clean STAR/CAR with transitions"
  - id: evidence
    label: Evidence & Metrics
    levels:
      1: "Anecdotal only"
      3: "Some metrics; limited linkage"
      5: "Specific metrics tied to outcomes"
scoring:
  aggregation: weighted_mean
  weights: {clarity: 0.35, logic: 0.35, evidence: 0.30}

Attach this rubric to video, audio, and text prompts alike inside SageStepper. Use item analytics to check discrimination and difficulty by format.

Example item analysis (export from SageStepper):

{
  "itemId": "PROD_SENSE_A2",
  "format": ["video","audio","text"],
  "n": {"video": 184, "audio": 219, "text": 203},
  "avgScore": {"video": 3.2, "audio": 3.1, "text": 3.3},
  "stdDev": {"video": 0.7, "audio": 0.8, "text": 0.7},
  "discrimination": {"video": 0.41, "audio": 0.39, "text": 0.42}
}

If one format consistently scores higher with similar discrimination, adjust prompts (not rubrics) to restore equivalence.

Step 4: Deliver with lightweight rituals that build streaks

Content is potential energy; rituals turn it into reps. Ship a cadence your learners can keep.

Recommended weekly rhythm:

• Mon: 20-minute Interview Sprint (text + flashcards). Focus on problem decomposition.
• Wed: 20-minute Sprint (audio). Practice executive summaries while commuting.
• Fri: 25-minute Sprint (video). Simulate panel stamina with Nova follow-ups.

What to track in SageStepper:

• Streak length by learner and by format. Trigger outreach at 0-day streak dips.
• Time-on-task per format (target 45–60 minutes/week across modes).
• Progress radar movement per competency (goal: +6–8 points/month for core skills).

Playlists keep it simple:

// sprint-playlist.ts
export const Week8Playlist = [
  { id: "COMM_STRUCT_V3", format: "video" },
  { id: "PROD_SENSE_A2", format: "audio" },
  { id: "PROB_DECOMP_T4", format: "text" }
];

Assign this playlist to your cohort via action boards. Auto-flashcards from each item drive spaced retrieval between Sprints.

Step 5: Accessibility and UDL—bake it in, don’t bolt it on

Multi-format done right doubles as accessibility. Minimum standards:

• Captions + transcripts for all video/audio. SageStepper auto-generates VTT and TXT; review for jargon.
• Alt text + slide notes if you share decks. Keep core prompts in plain text within the platform.
• Flexible timeboxes with a clear default. Allow +50% time for approved accommodations.
• Low-bandwidth options: Always publish the text variant first; keep audio files under 10MB.

Release checklist you can paste into your action boards:

- [ ] Captions reviewed (names, acronyms correct)
- [ ] Transcript posted and searchable
- [ ] Text variant live with identical rubric
- [ ] Audio file < 10MB, normalized volume
- [ ] Visuals have alt text; color contrast passes
- [ ] Accessibility statement linked

Programs that met this bar saw a 19% increase in weekly active learners and a 24% reduction in “can’t access” tickets.

Step 6: Measure what matters and iterate monthly

Keep your analytics focused on learning signals and operational savings.

Core metrics in SageStepper:

• Readiness score lift per competency and per format.
• Attempt distribution by format: are learners actually using alternates?
• Item quality: discrimination, difficulty, and time-to-complete by format.
• Mentor hours saved: time displaced by Nova and asynchronous review.

Monthly review ritual:

1. Export the item bank report and flag bottom 10% by discrimination.
2. Replace two underperforming items; add one new audio-first prompt.
3. Nudge non-users of alternates via action boards tasks.

Sample snapshot:

{
  "cohort": "Spring-25",
  "readinessLift": {"multiFormat": 23.4, "singleFormat": 11.2},
  "streakMedian": 9,
  "mentorHoursSaved": 132,
  "attemptShare": {"video": 0.44, "audio": 0.31, "text": 0.25}
}

If text attempts lag, add a 5-minute “priming” text micro-task before video sprints to build comfort.

Step 7: Case snippets you can copy tomorrow

Three real scenarios and how to deploy across formats in under an hour each.

1. Nursing new-grad: SBAR handoff

• Video: Demonstrate SBAR handoff with a patient change; include vital sign anomalies.
• Audio: 90-second SBAR voicemail simulation; learner records response.
• Text: Write SBAR note in the chart (structured template).
• Outcome: +21-point lift on communication rubric; reduced faculty paging by 33%.

2. Data analyst: SQL join debugging

• Video: Walk through a failing query with explain plan.
• Audio: “Whiteboard verbally” the approach to optimize joins.
• Text: Provide the schema and ask for the corrected query + rationale.
• Outcome: 1.3x faster time-to-correct; mentors saved 2 hours/week via async review.

3. Product manager: Metric definition under ambiguity

• Video: Frame success metrics for a new onboarding flow.
• Audio: Executive summary to a VP: trade-offs and next step.
• Text: Write a one-pager PRD metric section.
• Outcome: Offer rate up 1.5x among candidates with 12+ multi-format reps.

Use Nova for follow-up questions and progress radar to visualize competency deltas after each mini-scenario.

Common pitfalls (and simple fixes)

• Pitfall: Overproduced video that delays shipping.
  • Fix: Favor 2–4 minute “rough cut” explainers. Let Nova handle polish in follow-ups.
• Pitfall: Different prompts per format.
  • Fix: Use the master template and keep constraints identical.
• Pitfall: Rubrics drift by format.
  • Fix: One rubric per competency. Audit monthly for equivalence.
• Pitfall: Learners binge one mode.
  • Fix: Playlist rules that require at least two formats per week.
• Pitfall: Faculty time crush.
  • Fix: Batch-create with Nova, schedule via action boards, and review data bi-weekly.