L0product

Demand Signal Scoring

Close the SPCL feedback loop — score audience response signals into typed demand (latent/active/validated) that patches creator PowerIndicators. The return path that makes content strategy a closed loop instead of a guess.

360

Priority Score

Pain × Demand × Edge × Trend × Conversion

Customer Journey

Why should I care?

Five cards that sell the dream

1Why

Content without feedback is gambling.

If you can't measure what landed, how do you know what to publish next?

The friction: You publish content scored by SPCL. Audience responds. Saves, shares, DMs, clicks. Those responses go unmeasured. Your next content decision runs on the same stale data.

The desire: Every content piece scored within 48 hours. Know whether demand is latent (they saved it), active (they reached out), or validated (they paid).

The proof: Hormozi built $100M on this insight. Facebook keeps the scoring proprietary. We make it transparent and portable.

The stories that prove this matters

2Evidence

Three demand types, one progression.

What separates a bookmark from a purchase order?

Not all signals are equal. A save means "later." A DM means "now." A conversion means "done." The algorithm classifies, so you focus on what's moving.

Type	Signal	Validates
Latent	Saves, bookmarks	Pain exists
Active	DMs, replies, clicks	Demand exists
Validated	Conversions, payments	Conversion works

How demand typing works

4Loop

Signals decay. Honesty compounds.

Is a save from last month still a demand signal?

A save at hour 1 means something different than a save at hour 72. The algorithm applies time-weighted decay so stale signals don't masquerade as fresh demand.

Without decay	With decay
Viral spike = forever high	Viral spike = decaying signal
Stale data looks fresh	Fresh signals weigh more
Decisions lag reality	Decisions track reality

Signal decay stories

5People

Transparent scoring beats black boxes.

Who benefits when demand scoring is proprietary?

Facebook scores your demand signals internally. LinkedIn scores them internally. You never see the algorithm. You optimize for their metrics, not yours.

This algorithm is pure, deterministic, on-chain ready. Everyone sees how it works. The creator owns the signal. The platform doesn't extract — it compounds.

Who this serves

1 / 5

Same five positions. Different seat. The customer asks "will this help me?" The builder asks "can we prove it works?"

Feature Dev Journey

How does this get built?

Five cards that sell the process

1Job

One algorithm. Two feature IDs.

What's the smallest thing that closes the loop?

DSS-001: calculateDemandSignal() — pure function, 6 signal types, time-weighted decay, typed demand output.

DSS-002: Feedback wire — extracts powerFeedback, patches PowerIndicators.trackRecord for next SPCL run.

Build contract details

1 / 5

Situation

A creator publishes content scored by SPCL. Audience responds — saves, shares, DMs, clicks, conversions. Those responses go unmeasured. The creator's next SPCL score uses stale PowerIndicators because no feedback path exists.

Intention

Every content piece has a demand score within 48 hours of publish. The score classifies demand type (latent/active/validated), computes signal strength, and patches the creator's PowerIndicators for the next SPCL run.

Obstacle

Platform response signals are scattered across APIs (LinkedIn, YouTube, email, web analytics). No unified schema. No decay function. No mapping from raw signals to typed demand.

Hardest Thing

Signal decay. A save at hour 1 means something different than a save at hour 72. Getting time-weighted scoring right determines whether the demand signal is actionable or noise.

Priority (5P)

3/5

Pain

3/5

Demand

4/5

Edge

5/5

Trend

2/5

Convert

Readiness (5R)

Principles4 / 5

Performance1 / 5

Platform3 / 5

Process2 / 5

Players2 / 5

What Exists

Component	State	Gap
SPCL scoring algorithm	Working	One-directional. No return path from audience response.
PowerIndicators type	Working	trackRecord fields exist but never updated from content performance.
content-strategy index	Working	3 Future slots reserved. None implemented.
Lead magnet template	Working	Describes demand signals conceptually. No scoring.
validate-demand skill	Working	Manual 4-gate process. No scored signal input.
explore-exploit algorithm	Working	Could drive content experimentation once demand signals exist.

PRD	Contributes
Autoresearch Loop	Demand signals feed overnight autonomous loops. The autoresearch loop chains SPCL + demand-signal into a content optimization cycle.
Agent Platform	Algorithm lives in libs/agency/. Agent platform provides the orchestration layer.
Sales Dev Agent	Demand signals inform which content the sales agent should produce. Active demand = warm leads.
Marketing Sites	Marketing sites capture leads. Demand signal scores the response. Closed loop.
Prompt Deck	PromptDeck compression uses demand signals to prioritize which cards resonate. The 5P Demand dimension gets scored evidence.

Kill Signal

If after 30 days no content piece has been scored through the full SPCL -> publish -> demand-signal -> feedback loop, the algorithm is solving a theoretical problem. Kill it.

Questions

If Facebook's algorithm already measures latent demand but keeps it proprietary, what happens when that scoring is transparent and portable?

What decay curve shape best represents signal freshness — exponential, linear, or step function?
Should the feedback wire increment AND decrement PowerIndicators, or only increment with natural decay?
At what point does demand signal scoring replace manual 5P Demand scoring versus just informing it?