Zendory
Archived framework Browse frameworks See active products
Framework

Signal Decay Model (SDM)

This framework describes how outbound signals lose relevance over time and how freshness affects actionability.

It is intentionally rules-based, observable, and non-opinionated.

Purpose

Define how outbound signals lose relevance over time, and why freshness matters more than precision past a point.

This framework exists to explain:

  • why old lists fail
  • why “evergreen leads” don’t exist
  • why timing is a first-order variable in prospecting

This is not about urgency or manipulation.

It is about information entropy.

Core Principle (canonical)

All prospecting signals decay. The only question is how fast.

A signal’s value decreases as the probability that it still reflects reality decreases.

What This Framework Governs

This framework determines:

  • how long a signal remains valid
  • when a prospect should be downgraded or removed
  • why recency beats completeness
  • why stale accuracy is still noise

It does not predict outcomes.

Canonical Inputs

For any signal, the model considers only:

  • signal_type
  • signal_timestamp
  • current_date
  • visibility = {explicit | inferred}

No engagement metrics.

No performance data.

Core Output

For each signal, output:

  • age_days
  • freshness_bucket
  • confidence_modifier
  • actionability
  • recommended_action

Signal Classes and Decay Rates (authoritative)

Different signals decay at different speeds.

Class A — Fast-Decay Signals

These reflect temporary intent or transition.

Examples:

  • Hiring signals
  • Campaign launches
  • New ads detected
  • Temporary landing pages

Decay behavior

  • High initial value
  • Rapid drop-off

Freshness buckets

  • 0–30 days → high confidence
  • 31–60 days → medium confidence
  • 61–90 days → low confidence
  • >90 days → expired

Rule

After 90 days, Class A signals are no longer prospectable.

Class B — Medium-Decay Signals

These reflect structural state, but can change.

Examples:

  • Conversion paths
  • CTA structure
  • Landing page alignment
  • Funnel clarity

Decay behavior

  • Moderate initial value
  • Gradual decay

Freshness buckets

  • 0–60 days → high confidence
  • 61–120 days → medium confidence
  • 121–180 days → low confidence
  • >180 days → stale

Class C — Slow-Decay Signals

These reflect deep structural characteristics.

Examples:

  • SEO structure
  • Service page presence
  • Information architecture
  • Indexation gaps

Decay behavior

  • Lower initial volatility
  • Slow decay

Freshness buckets

  • 0–180 days → high confidence
  • 181–365 days → medium confidence
  • >365 days → review required

Visibility Modifier (important)

Signals decay faster if they are inferred, not explicit.

Explicit signals

Directly observable, verifiable, timestamped

Examples:

  • Job posting date
  • Ad currently running
  • Page visibly missing

Inferred signals

Derived from absence, contextual interpretation

Examples:

  • “Probably scaling”
  • “Likely investing in growth”

Rule

Inferred signals decay one freshness bucket faster than explicit signals.

Freshness Buckets (canonical labels)

Use neutral, non-marketing labels only:

  • recent
  • ongoing
  • stale
  • expired

These are intentionally vague but operationally precise.

Confidence Modifier Rules

  • high → signal is explicit and recent
  • medium → signal is explicit but aging
  • low → signal is inferred or near expiry
  • none → expired signals

Confidence modifies whether to outreach, not what to say.

Actionability Rules

Freshness Actionability Recommended Action
recent high pursue
ongoing medium pursue or hold
stale low hold or downgrade
expired none remove

Expired signals must be deleted or archived, not reused.

Canonical Axioms

Old data is not neutral — it is misleading.
Precision decays faster than relevance.
Freshness is a signal.
Timing is not urgency; it’s accuracy.

Common Failure Modes (explicit)

This framework explains why these fail:

  • Reusing old prospect lists
  • Treating hiring signals as evergreen
  • Assuming structural issues never change
  • Optimizing for accuracy while ignoring time

Zendory Positioning (safe, generic)

This framework describes how outbound signals lose relevance over time and how freshness affects actionability.

Zendory is designed to support time-bounded, observable signals, but the framework itself is implementation-agnostic.