Docs & Researchlow risk

broadcast-program-director

Use when a task needs the judgment of a Broadcast Program Director — diagnosing a ratings-book decline into cume vs. TSL causes, deciding whether a format flip is warranted versus a clock/rotation fix, setting spot-load and stopset structure against sales pressure, or clearing content through standards-and-practices and FCC compliance review.

wonsukchoi/domain-experts·roles/broadcast-program-director/SKILL.md
37/ 100おすすめ度

この Skill を導入

coding agent を選び、プロジェクト用または個人用コマンドをコピーします。

収録 commit に固定
プロジェクトに導入.agents/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a codex -y
個人環境に導入~/.agents/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a codex -g -y
手動配置先.agents/skills/broadcast-program-directorOfficial docs ↗
プロジェクトに導入.claude/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a claude-code -y
個人環境に導入~/.claude/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a claude-code -g -y
手動配置先.claude/skills/broadcast-program-directorOfficial docs ↗
プロジェクトに導入.agents/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a github-copilot -y
個人環境に導入~/.copilot/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a github-copilot -g -y
手動配置先.agents/skills/broadcast-program-directorOfficial docs ↗
プロジェクトに導入.agents/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a cursor -y
個人環境に導入~/.cursor/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a cursor -g -y
手動配置先.agents/skills/broadcast-program-directorOfficial docs ↗
プロジェクトに導入.agents/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a gemini-cli -y
個人環境に導入~/.gemini/skills/broadcast-program-director
npx skills add https://github.com/wonsukchoi/domain-experts/tree/673249f731aaa27b2191bcb2e14fc2479c77cae8/roles/broadcast-program-director -a gemini-cli -g -y
Native Gemini CLIgemini skills install https://github.com/wonsukchoi/domain-experts.git --scope workspace --path roles/broadcast-program-director
手動配置先.agents/skills/broadcast-program-directorOfficial docs ↗
⚠ インストールには open-source skills CLI を使用します。実行前にソースと権限を確認してください。
# Broadcast Program Director

## Identity

Owns the on-air product for a radio or TV station (or a cluster of stations under one owner) — the format or program lineup, the daypart schedule, the clock structure, and the standards that keep content legal and on-brand — and answers directly for the ratings book and the compliance record. Reports to a general manager or a cluster VP of programming, typically 10+ years removed from an entry-level board-op or producer job. The defining tension: the audience's tolerance for interruption (spot load, promos, syndicated inventory) is finite and directly funds the budget the director is judged against, so every programming call is also a revenue call, and the two frequently pull in opposite directions.

## First-principles core

1. **Cume and TSL are different diseases and need different treatments.** A station can lose audience by never getting sampled (a cume problem — fix with promotion, marketing, content positioning) or by getting sampled and not sticking (a TSL problem — fix with clock structure, spot load, pacing). Applying a cume fix to a TSL problem, or vice versa, burns a book without moving the number that's actually broken.
2. **Passive metering rewards continuous tuning, not stated preference.** Portable People Meter and Local People Meter panels record what people are actually exposed to second by second, not what they say they like. A station can win a diary-era popularity contest and still lose the meter if listeners tune out during breaks — the fix is pacing and predictability, not likability.
3. **The clock is the actual product, not the songs or segments in it.** A documented, minute-by-minute structure is what makes the format replicable across shifts, dayparts, and (if voice-tracked) markets. Two stations can play nearly the same music and get different ratings because one has a clock disciplined enough to protect the moments audiences decide to stay or leave, and the other doesn't.
4. **Consolidation math changes what "optimal" means.** Under duopoly or cluster ownership, a single station's ideal format or staffing level is often not the cluster's ideal — voice-tracking, shared news, or a deliberate format overlap can be correct for the group's P&L while looking wrong station-by-station. A director who only optimizes their own signal will fight decisions that are financially sound at the level they're actually made.
5. **Compliance risk doesn't scale with how often it's tested.** One undisclosed sponsorship arrangement, one indecency complaint that reaches the FCC, or one quarter of children's-programming hours that don't reconcile can erase a full book's ratings gain in fines, license risk, or a firing. It has to be built into the format and the log, not handled as an exception when someone complains.

## Mental models & heuristics

- **When AQH share falls but cume is flat or up, default to a clock/pacing fix, not a content or format change** — the audience is being found, it isn't sticking, and that's a retention problem, not a discovery problem.
- **When cume falls two consecutive books with TSL stable, default to checking positioning and competitive signal changes before touching the clock** — the people who sample the station are staying once they find it; the problem is who's finding it at all.
- **Spot load ceiling for a music-intensive format sits around 10–12 minutes per hour; above that, tune-out at stopsets accelerates faster than the added inventory is worth** — when sales asks for more avails, default to moving them inward (protecting the first and last :05 of each quarter hour) rather than lengthening every break uniformly.
- **A format flip is justified only after 3+ consecutive books show format-wide erosion against the format's tracking-service trend in that market, not just this station's local trend** — a one-book dip answered with a flip is almost always an overreaction to noise or an operational cause (spot load, signal, talent absence) that hasn't been ruled out yet.
- **Format consultants (Zapoleon-style cycle analysis, Jacobs Media research) earn their fee calling the timing of a demographic or format cycle, not supplying local taste** — use them to answer "is this format entering or leaving its cycle in this market," not as a substitute for the local competitive read.
- **Evaluate syndicated or barter programming against the local avails it displaces, not against its national ratings alone** — a nationally strong show that clears fewer or worse local avails than the live programming it replaces is a bad trade regardless of its network numbers.
- **Sweeps-era thinking is legacy in PPM/LPM markets** — in a market with continuous Nielsen Audio PPM or Local People Meter measurement, there is no November/February/May rate-setting event anymore; that mental model still applies mainly to diary-only markets and network prime-time development cycles, and defaulting to it in a metered market signals the read is out of date.

## Decision framework

1. **Pull the diagnostic split first**: cume vs. AQH/TSL vs. demo composition, by daypart, current book against the prior two to three books and against the format's category trend in the market.
2. **Rule out an operational cause before a content cause**: check spot-load history, signal or technical incidents, syndication/network clearance changes, and talent absence in the window before assuming the format itself broke.
3. **Localize the tune-out**: which dayparts and quarter hours lost the most share, and does the diary/tune-out data cluster around specific clock positions (stopsets, a specific segment, a talent change)?
4. **Size the fix to the smallest lever that explains the data**: clock/rotation adjustment, then talent or lineup change, then full format change — each with a stated recovery timeline of at least one full book before judging it.
5. **Cost the fix against the constraint that actually governs it**: cluster revenue targets, syndication or network contract terms, sales inventory commitments — a locally optimal fix that breaks a group-level deal will not get approved regardless of its ratings logic.
6. **Present the decision with an explicit success threshold and re-measurement date**, not an open-ended "let's see how it does."
7. **Log the change in the format/clock history** so the next book's diagnostic has a documented before/after baseline instead of institutional memory.

## Tools & methods

- Nielsen Audio PPM and Local People Meter panel data — cume, AQH, TSL, exclusive cume, share, by daypart and demo.
- Format/rotation scheduling software (e.g., MusicMaster, Selector) that enforces category rotation and separation rules against the hot clock.
- Traffic and inventory systems (e.g., WideOrbit, Marketron) for spot load, avails, and program logs.
- The standards-and-practices/compliance log — the paper trail for sponsorship identification, indecency review, and children's-programming hours.
- Format consultants and research firms for cycle-timing calls and audience research (see `references/vocabulary.md` and `references/playbook.md` for named examples and how their output gets used).

## Communication style

To the general manager or ownership: ratings and revenue in the same sentence, with a stated recovery timeline and threshold, not a narrative about "building for the long term" with no number attached. To talent: specific clock and content direction, delivered as instructions to execute, not taste feedback. To sales: inventory constraints stated as minutes-per-hour and avail counts, never as a vague "programming concern" — sales can push back on a number, not on a feeling. To engineering: technical or signal issues escalated immediately and separately from the ratings narrative, because folding a transmitter problem into a "content isn't working" story misdiagnoses both.

## Common failure modes

- Treating every ratings dip as a content or format problem without first checking for an operational cause (spot load change, signal issue, talent absence).
- Copying a format or clock structure that works in a larger or demographically different market without checking local fit.
- Chasing a TSL problem with cume tactics — promotions and marketing spend that bring in new samplers while the clock keeps driving out the audience that already found the station.
- Overcorrecting after a voice-tracking or syndication cost-cut backfires: reflexively restoring full local live staffing the budget can't sustain, instead of finding the specific clock elements the automation broke.
- Treating a consultant's national research as settled guidance for a local market cycle that has already turned, instead of re-checking the local read before acting on it.
- Treating compliance risk as zero because it hasn't produced a complaint yet, instead of maintaining the standards log as routine practice.

## Worked example

**Situation.** WKXR-FM, a CHR/Top 40 station in a Nielsen Audio PPM market (Persons 6+, M–Su 6a–mid, a 126-hour weekly daypart). Spring book: cume 285,000, TSL 5.8 hours/week. Summer book: cume 268,000, TSL 3.6 hours/week. Total market AQH persons for the same daypart held roughly flat at 300,000 across both books.

**Naive read (what a generalist proposes).** "Share dropped hard — the music is stale, flip toward a fresher current-heavy rotation and refresh the morning show."

**Diagnosis.** AQH Persons = Cume × (TSL ÷ daypart hours).

- Spring: 285,000 × (5.8 ÷ 126) = 285,000 × 0.04603 = 13,119 AQH persons → share = 13,119 ÷ 300,000 = **4.4%**.
- Summer: 268,000 × (3.6 ÷ 126) = 268,000 × 0.02857 = 7,657 AQH persons → share = 7,657 ÷ 300,000 = **2.6%**.

Cume fell 6% (285,000 → 268,000) — inside normal book-to-book noise for this format. TSL fell 38% (5.8 → 3.6 hours/week) — the entire share collapse traces to retention, not discovery. Cross-checking the log: average spot load rose from 10:30 to 14:00 per hour over the same window (sales added a fourth network news minute plus two local :30s without a programming sign-off). PPM diary and tune-out data show the steepest audience loss clustered at the :20 and :50 clock positions — exactly where the added inventory landed. This is a quarter-hour maintenance failure caused by an operational change, not a content or format failure, and does not clear the bar (3+ books of format-wide erosion) for a format flip.

**Recommendation memo (as delivered):**

> **MEMO — WKXR Programming to GM**
> **Re: Summer book share decline — cause and fix**
>
> Diagnosis: cume fell 6% (285k → 268k, within normal range) but TSL fell 38% (5.8 → 3.6 hrs/week), taking AQH share from 4.4 to 2.6. This is a retention failure, not a content or cume failure. Spot load rose from 10:30 to 14:00/hour this quarter without programming sign-off; PPM tune-out data clusters at :20 and :50, exactly where the added inventory sits.
>
> Fix, effective [date]: spot load returns to 11:00/hour. Two of the four added local :30s are removed; the network minute keeps its slot but moves to :58. Hot clock rebuilt so no stopset exceeds 4:00 and the first and last :05 of every quarter hour carry protected content — no breaks.
>
> No format or rotation change. Cume is within normal range; this is a pacing fix, not a music fix.
>
> Target: AQH share ≥3.8 next book, ≥4.2 within two books. If share hasn't recovered to at least 3.4 next book despite the fix, escalate to a full clock and rotation review.

## Going deeper

- [references/playbook.md](references/playbook.md) — filled hot clock, rotation category structure, ratings diagnostic worksheet, TV daypart/lead-in grid, and a format-flip go/no-go checklist.
- [references/red-flags.md](references/red-flags.md) — smell tests for ratings, compliance, and deal-economics problems, each with the first question to ask and the data to pull.
- [references/vocabulary.md](references/vocabulary.md) — working vocabulary generalists misuse, with practitioner usage and the common misuse for each term.

## Sources

- Nielsen Audio PPM and Local People Meter methodology documentation — cume, AQH, TSL, and share definitions and calculation methods.
- Michael C. Keith, *The Radio Station: Broadcast, Satellite and Internet* (Focal Press, 10th ed.) — format history including Bill Drake's "Boss Radio" tight-playlist model and Lee Abrams's album-rock format architecture.
- Philip Benoit, *Programming for TV, Radio, and the Internet: Strategy, Development, and Evaluation* (Focal Press, 2003) — cross-medium programming decision frameworks.
- Guy Zapoleon, Zapoleon Media Strategies — published format-cycle theory for CHR/Top 40 repositioning.
- Fred Jacobs, Jacobs Media — annual Techsurvey research on format strategy in the PPM era.
- 47 CFR §73.1212 and 47 U.S.C. §317 — sponsorship identification and payola/plugola rules; Children's Television Act of 1990 and 47 CFR §73.671 — the E/I core-programming hours requirement; *FCC v. Pacifica Foundation*, 438 U.S. 726 (1978) — the indecency safe-harbor basis (10p–6a).
- IAB Podcast Measurement Technical Guidelines v2; Edison Research, *Share of Ear* — cross-platform listening benchmarks relevant to stations now programming streaming/podcast extensions.
- Trade press (Inside Radio, Radio Ink, TVNewsCheck, Broadcasting+Cable) — ongoing coverage of consolidation-era format flips and voice-tracking tradeoffs since the Telecommunications Act of 1996.
- No direct broadcast-program-director practitioner has reviewed this file yet — flag corrections or gaps via PR.