Signal prioritization

Signal Prioritization

Role: Signal buyer. You answer which signals are worth paying for, which can be derived from public data, and how to weight them — before the team commits a data budget. The goal: a signal stack that surfaces accounts at the right moment, not a firehose of triggers that produces analyst paralysis.

---

Before starting

Confirm (ask or infer):

• ICP trigger event — from ICP research Layer B. The trigger event defines which signals are proxy-valid. If the trigger is "post-production incident," then job postings for "SRE" and PagerDuty stack adoption are strong proxies; generic intent data is weak.
• Sales motion — SLG outbound needs account-level signals; PLG expansion needs user-level behavioral signals.
• Signal data budget — what can the team spend per month on third-party signal data?
• Current signal stack — what data sources are already connected (CRM, product analytics, content engagement)?

IF icp_trigger_event = undefined →
  BLOCK. Return:
  "Signal prioritization without a defined trigger event produces a list of
   data sources, not a decision framework. Run /icp-research, return with
   Layer B (trigger event) populated, then select signals that proxy for it."

---

Inputs

Input	Required?	Description
ICP trigger event (Layer B)	Required	The specific event that makes an account ready to buy
ICP motion fit	Required	PLG / SLG / MLG — determines signal type needed
Monthly signal budget	Required	Even rough: $0 / <$2K / $2K–$10K / $10K+
Current data sources	Required	What's already connected: CRM, product analytics, marketing platform

---

Contract

This skill guarantees:

• Signal recommendations are grounded in the ICP trigger event — not generic best practice
• Signals are classified: own (first-party), monitor (free proxies), buy (paid)
• Each paid signal recommendation includes expected signal-to-pipeline ratio and a falsification test
• Output is a prioritized buy/build/skip list — not a feature comparison of data vendors

---

Decision logic

Phase 1 — Trigger-to-signal mapping

For each ICP trigger event, identify signals that proxy for it:

Trigger type → Signal proxy category:

  "Post-production incident" →
    - Job posting: SRE, Platform Engineer, Head of Reliability (Tier 3 proxy)
    - Tech stack: PagerDuty, OpsGenie in stack (Tier 2 technographic)
    - Content: HN/Reddit post about incident from that company (Tier 3 proxy)
    - First-party: their engineer on your pricing page (Tier 1)

  "Funding event" →
    - Crunchbase / news: Series A/B announcement (Tier 3 proxy — public)
    - Hiring surge: open roles in the funded department (Tier 3 proxy)
    - Note: funding is a budget trigger, not a pain trigger; pair with pain signal

  "New leadership hire in buying department" →
    - LinkedIn job change (Tier 3, manual) or LinkedIn Sales Navigator alert (Tier 2)
    - Note: new leaders change vendors in first 90 days more often than stable leaders

  "Competitor churning" →
    - G2 negative reviews of competitors in last 30 days (Tier 2, public)
    - Competitor customer list changes (Tier 3, manual from LinkedIn)

  "Stack migration / modernization" →
    - BuiltWith / HG Insights: competitor removal + target stack adoption (Tier 2 technographic)
    - Job postings referencing migration: "migrate from X to Y" (Tier 3)

Instruction: For each ICP trigger event, list 2–4 signals that proxy for it.
Rank by precision (how often the signal co-occurs with the trigger).

Phase 2 — Signal classification and investment framework

Classify all identified signals into three buckets:

Class	Definition	Action
Own	You already generate this signal (product behavior, CRM data, content engagement)	Activate and connect to scoring immediately. This is your highest-precision data.
Monitor	Free public signal you can track without a subscription (Crunchbase free, LinkedIn, G2 public)	Set up monitoring workflow. High effort but zero cost.
Buy	Requires a paid subscription (Bombora, G2 Buyer Intent, Demandbase, ZoomInfo intent, etc.)	Evaluate ROI before committing. Apply the buying test below.

Buying test for paid signals:

Before subscribing to any paid signal source, confirm:

1. Signal-to-trigger precision: Of accounts flagged by this signal, what % actually
   have the ICP trigger event active? If you can't estimate this, start with a trial.

2. Volume match: Does this signal fire enough times per month to feed your pipeline target?
   Rule: monthly signal volume ≥ 3× your monthly qualified meeting target.

3. Falsifiability: How will you know in 90 days if this signal is generating pipeline?
   Define: signal → account contacted → meeting booked → pipeline created.
   If you can't trace this chain, you can't measure ROI.

4. First-party alternative: Is there a first-party signal you could generate instead?
   (Example: content piece that attracts the specific ICP, replacing a trigger data subscription)

IF buying_test passes on all 4 criteria → BUY
IF 2–3 criteria pass → PILOT (30–60 day trial with explicit success metric)
IF < 2 criteria pass → SKIP (monitor the free proxy instead)

Phase 3 — Signal weighting model

Assign weights to each signal in the account scoring model (feeds into /outbound-motion):

Weight framework (total = 100 points):

  First-party signals (own): 50 points total
    - Product activation / trial: 30 pts
    - Pricing/demo page visit: 20 pts (if tracked)
    - Content engagement from target account: 10 pts

  Third-party intent (buy): 30 points total
    - Intent topic surge on primary pain keyword: 20 pts
    - Competitor review activity: 10 pts

  Trigger event proxies (monitor): 20 points total
    - Funding event (last 60 days): 10 pts
    - New hire in buying department: 10 pts

Customize this template to the ICP's actual trigger event:
  - If trigger = funding → give funding signal 15 pts, reduce other proxies
  - If trigger = production incident → give SRE job posting proxy 10 pts
  - If first-party signals are unavailable → redistribute weight to Tier 2 + Tier 3

Phase 4 — Build vs. buy decision for missing signals

For each signal gap (trigger has no current proxy):

IF signal can be generated through content or product →
  Recommend: build a first-party version.
  Example: publish a benchmark report that attracts accounts in the trigger state;
  track downloads as a first-party proxy for the trigger.

IF signal requires third-party data AND buying test passes →
  Recommend: buy. Specify vendor category (not specific vendor — use what's in your stack).

IF signal requires extensive scraping or manual monitoring →
  Recommend: monitor manually at small volume until signal-to-pipeline is validated,
  then automate or buy.

---

Output format

## Signal prioritization — [date]

ICP trigger event: [from Layer B]
Sales motion: [SLG / MLG / PLG]
Signal budget: [$range]

### Trigger-to-signal mapping
  Trigger: [event]
  Proxy signals:
    [Signal 1] — Precision: [high/med/low] — Class: [own/monitor/buy]
    [Signal 2] — ...

### Signal inventory
  Own (activate now): [list]
  Monitor (set up workflow): [list]
  Buy (passed buying test): [list with vendor category, not specific vendor]
  Skip: [list with reasons]

### Signal weighting model (feeds outbound-motion scoring)
  [Weight table as above, customized to this ICP]

### Build vs. buy decisions
  [Per signal gap: recommendation with rationale]

### Success metric
  In 90 days: [specific pipeline metric that validates this signal stack]

---

Anti-patterns

Anti-pattern	Why it fails	Fix
Buying intent data before defining the trigger event	Intent data for undefined triggers is a firehose; generates leads but not pipeline	Define ICP trigger event first; use it to select which topic clusters and keywords to monitor
Treating all intent signals equally	Tier 1 (first-party) is 5–10× more precise than Tier 3 (proxy); equal weighting dilutes scoring	Apply the weighting model; first-party signals dominate score
No falsification criterion for signal spend	After 6 months, you can't tell if the data is working	Define the pipeline trace before buying; set a 90-day review checkpoint
Building a complex 15-signal stack at early stage	Too many signals produce analyst paralysis; scoring becomes meaningless	Start with 2–3 signals max; validate signal-to-pipeline before adding more
Vendor-locking the signal stack	Specific vendor → single point of failure	Recommend signal CATEGORY (intent data / technographic / job data); let team choose vendor

---

Validation criteria

• All recommended signals are mapped to the ICP trigger event — not generic best practice
• Paid signal recommendations have passed the 4-criteria buying test
• Signal weighting model sums to 100 and gives priority to first-party signals
• Output includes a falsification criterion: how to evaluate the stack in 90 days
• Output specifies signal category, not specific vendor names

---

Benchmarks (2025–2026)

Benchmark	Value	Source
Improvement in pipeline quality with signal-triggered outbound	2–4×	Demandbase 2025, 6sense 2025
B2B intent data subscription cost (mid-tier)	$1,500–$5,000/month	G2 pricing research 2025
Average signal-to-meeting rate with Tier 1 (product engagement)	8–15%	Winning by Design 2025
Average signal-to-meeting rate with Tier 3 (proxy triggers only)	1–3%	Outreach benchmark data 2024
Contact data decay rate	2.1% per month	B2B data research 2025

---

References & Sources

Tier-1 frameworks:

• 6sense / Demandbase category research — B2B intent signal effectiveness
• Winning by Design — signal-to-pipeline tracing methodology

Tier-2 operator synthesis:

• Signal stack prioritization synthesis (research synthesis, 2026-06-13): tier framework, weighting model, buying test